1

Contents

Foreword
Acknowledgements
Introduction – Hooked on the Center Game - In search of a
practical answer to 1...e5
Part I - A history of the Center Game
Chapter 1 - From the Middle Ages to Y2K
Chapter 2 - The Center Game in the New Millennium
Part II - The theory of the Center Game
Chapter 3 - Paulsen Variation 4. e3: old main lines
Chapter 4 - Paulsen Variation 4. e3: lines with ... e7 and 5. d2
subtleties
Chapter 5 - Paulsen Variation 4. e3: other 4th moves for Black
Chapter 6 - New alternatives for White and minor black moves
Epilogue
Bibliography

2
Foreword
This is a book about the Center Game, an old chess opening
characterized by the moves 1.e4 e5 2.d4 exd4 3. xd4.
At the highest level, the Center Game has never been a popular
answer to 1.e4 e5. But things are moving. The world’s number four
at the moment of writing, Arjun Erigaisi, is one of several top-rated
players who employ it regularly these days, and some of the most
advanced neural network chess engines evaluate 2.d4 (followed by
3. xd4) as almost equal to the alternative 2. f3. Interesting, isn’t
it?
My aim with this book is to show the reader why the Center Game
is a great opening, how it evolved, which players were ahead of
their time, and how it could, perhaps, become a more popular
opening in the near future. It certainly deserves to be taken seriously
as a dangerous surprise weapon. But this is not merely a theoretical
exercise or a repertoire book. It is also a personal account and a
declaration of love. I hope the reader will catch some of the
enthusiasm and joy I still feel for this opening after playing it for
almost 30 years.
I have played the Center Game in thousands of games, over the
board and online, sometimes against very strong opposition. I have
always scored well with it, but I have also suffered painful losses.
My reason for playing this opening is not that I like memorizing
theoretical lines. The Center Game (we largely use UK spelling in
this book, but we make an exception for ‘Center Game’ since that is
the generally accepted term) is a practical opening with a lot of
room for creativity and inventiveness, and players from the past
have used it as a competitive weapon to create relatively unknown
positions from the first moves onwards. While writing this book, I

3
realized once again how much there is still left to discover, and how
many ideas are still unexplored.

Despite its rich history, only a handful of books on the Center Game
have been written. In 2020, Marek Soszynski published the digital
book The Centre Game Re-examined. In 2023, Chessable published
an online repertoire course called The Daring Center Game by
Michael Gorny. A third title well worth mentioning is Andrew
Greet’s extensive and excellent section on the Center Game in the
Everyman chess book Dangerous Weapons: 1.e4 e5 from 2008.
There have also been a few titles in other languages, notably Mario
Ziegler’s 2010 monograph Paulsen Eröffnung (in German), in
which he also offers a historical perspective.
My own approach is somewhat different. I delve more deeply than
these authors into the opening’s ancient origins and its evolution
over time: how did its ideas take shape, and how did players from
the past handle the Center Game? How does this reflect on the way
the opening is treated now? I will end by considering the plethora of
new ideas for White that have been explored in the past few years,
such as moving the queen not to e3, but to c4 or d3.

In the Introduction, I talk about my personal connection to the

Center Game: how I came to study it, play it, and love it. Over the
decades, I’ve had many interesting exchanges with Center Game
fans and critics, and these encounters have strongly influenced my
perspective on this opening. They have also shaped me as a chess
player. I hope, and trust, that my search for a simple yet aggressive
weapon against 1.e4 e5 will be familiar to many players. With the
Center Game, we avoid having to reproduce long lines of theory,
while we still obtain attacking chances. That’s why I have always
loved this opening.

4
Part I of the book is devoted to the history and evolution of the
Center Game from past to present, celebrating its most important
pioneers along the way. We will travel back in time to the end of the
Middle Ages, when the queen (as a chess piece, that is) was given
new powers. I will look at the most essential ideas and themes, and
explain how these were assessed in their time. We will see some
legendary players of the late 19th and early 20th century in action
and then look how a new generation of Center Game proponents
treated the opening at the dawn of the new millennium.
We will end this first part of the book with a game by Magnus
Carlsen in which he played the Center Game. Throughout the book,
I will be regularly referring back to this historical part. In my
opinion, you can’t learn to love the Center Game without
understanding its history. It would remain just another interesting
but ultimately random and anonymous opening, of which there are
too many already.
Part II is the theoretical section. Illustrated by relevant games from
all eras and all levels of play, we will look at the current state of
theory of the Center Game. There are old lines to be revised (and
refuted) and fascinating new ideas to be examined. We will first
look at the Berger Variation (3... c6 4. e3 f6), which takes up
the vast majority of Center Game theory. There are two main ways
to play this for Black: by developing the bishop to b4 (Chapter 3), or
by developing it to e7 (Chapter 4). Next, in Chapter 5, we will check
fourth move alternatives for Black. In Chapter 6, we look at modern
alternatives for White on move four. Many of these lines are still
new, and I think this is one of the most exciting parts of the book. In
the last game, we look at an encounter between Topalov and Anand
from the Leon Masters, 2024, which was played just two days
before my deadline for this book. In the Epilogue, I draw some final
conclusions.

5
I didn’t want to write a typical ‘repertoire book’, in which a limited
and one-sided view of the opening is given, presenting it to the
reader as a cure-for-all. The Center Game is a great surprise
weapon, but it is objectively equal. If you’re thinking of
incorporating it into your repertoire with White, my most important
practical advice is to vary your approach. Don’t just play the
traditional lines and hope for the best, but also give the new ideas a
try. The good news is that many variations can be played on the
basis of the knowledge of just a few basic concepts. (Experienced
Center Game players might need to ‘un-learn’ many of their old
habits and routines when playing these new lines.) On the Black
side of this opening, my general recommendation is never to go for
passive development and hope for the best. Playas actively as
possible in the centre or on the queenside.

So, who am I to tell you all this? I’m a FIDE Candidate Master, and
my national rating has been close to 2300. My own games are
hardly interesting to the general reader, but I have taken a few
liberties in this book, quoting from my practice. This is my first
chess book, but I am not new to chess writing. I was a regular
columnist and book reviewer for the Chessvibes website, and later
for Chess.com. As you can read in the Introduction, I am fascinated
by many different episodes in chess history. But I am not a chess
historian, if such an occupation exists at all.

I hope this book will inspire readers to take a closer look at the
Center Game.

Arne Moll
Weesp, August 2024

6
Acknowledgements
Several people have helped me during the writing of this book. In
particular, I would like to thank Peter Doggers, Paul Janse and
Merijn van Delft for their invaluable advice and suggestions. Also
thanks to the Max Euwe Centre in Amsterdam, which has been a
source of information and inspiration for me since the late 1980s.

This book wouldn’t have been possible without my countless

conversations with friends and fellow members of my two chess
clubs, Caissa Eenhoorn in Hoorn and Tres Peons in Barcelona.
Thanks, all of you, for putting up with my ramblings about forgotten
gambits and obscure sidelines over the years!

I’ve tried my best to be as complete as possible in my research, but

inevitably there will be several crucial sources that I have ignored or
misinterpreted. If you, dear reader, notice any mistakes or
inaccuracies, please do let me know. I am always open to different
or dissenting views.

7
Introduction –
Hooked on the
Center Game
In search of a
practical answer to
1...e5
Well there have been better plans
But none that I could ever understand
The Sisters of Mercy, Lights
All chess players, except one, grow up.
When I was about ten years old, I learned the rules of chess. Around
that same time, my dad took me to the cinema to see the movie
Amadeus, which awakened a life-long passion for music in me.
Soon, I was so busy practising the piano that I pretty much forgot
about chess. At the age of thirteen, I started to get interested again,
but I didn’t really know how one was supposed to study chess, nor
did I ask anyone. It didn’t occur to me to join a club. Instead, I went
to a local bookstore and bought a cheap Dover version of Morphy’s
Games of Chess (1957) by Philip W. Sergeant. The life story and

8
games of Paul Morphy made a lasting impression on me. Morphy’s
playing style greatly appealed to me, but I was mainly intrigued by
his openings, particularly the romantic variations in the Open
Games after 1.e4 e5. Studying Morphy’s games and obsessing over
forgotten gambits – what a way to start a chess career!
One of the first chess events I visited was the 1988 Hoogovens
tournament in Wijk aan Zee. By this time, I had actually joined a
chess club, and was playing more seriously. Excited to finally see
some famous grandmasters in action, my father and I entered the
playing hall. I looked to the left, I walked to the right. And then I
nearly bumped into Mikhail Tal. ‘Watch out for the World
Champion!’ someone joked to me. I was starstruck. Later that day, a
friend of my dad’s, also a member of my chess club, told me some
anecdotes about the ‘Magician from Riga’, who wasn’t just
uncompromising behind the board, but also in life. Tal was my
second chess hero. It was now clear that I was going to be an
attacking player myself.
Not so long after that chance encounter with Tal, I heard about the
‘Latvian school of chess’. It consisted of chess players from Latvia
with a specific playing style: attacking, imaginative chess, not just
following the trodden theoretical paths but exploring original ideas
early on in the game, a bit like Morphy had done in his day. There
was Tal himself, of course, and his trainer Alexander Koblencs.
There was the old master Alvis Vitolinsh, and then there were the
young stars Alexander Shabalov and Alexei Shirov. I studied their
games and, unoriginally, copied many of their openings.
Early on in my chess career, I discovered that my capacity to
remember opening variations was far from impressive compared to
some of my peers who seemed able to memorize entire opening
books. I developed a slight phobia for ‘mainstream’ opening theory
after 1.e4 e5, which required concrete knowledge. There were
endless lines in the Ruy Lopez, the Italian and the Petroff to be

9
learned by heart, and I found out that I simply wasn’t able to. I
needed something more practical and tried to follow a piece of
advice from Tal, who once said that ‘young players are very fond of
trying to catch their opponents in prepared variations’. I
experimented with some offbeat gambits, but there were just too
many ways for my opponents to deviate from my prepared lines.
In the 1990s, my attention swerved away from chess as I became
interested in Victorian children’s literature. I was especially
fascinated by Lewis Carroll (1832-1898), Oxford don, logician and
author of Alice’s Adventures in Wonderland. Carroll also liked
chess and included a chess problem in the Alice sequel Through the
Looking-Glass, in which Alice meets several chess pieces. My
favourite character was the White Queen, who told Alice she could
believe ‘six impossible things before breakfast’. I also got involved
in music again. This time, I dived into a different musical rabbit-
hole. I wore black T-shirts of Queen, Metallica and The Sisters of
Mercy, the last one being my favourite band. The group had been
founded in the late 1970s by lead singer Andrew Eldritch, an elusive
figure with a deep voice who was always dressed in dark clothes
and wore large sunglasses, even in dimly-lit and smoke-filled
concert halls. His band was supported by a legendary drum machine
called Doktor Avalanche and it was particularly popular at ‘gothic’
music events and parties. In those years, I attended a lot of them.
One day, I was browsing through a New in Chess magazine when an
interview with Alexei Shirov caught my attention. At some point,
the interviewer asked him what he thought was the difference
between World Champion Garry Kasparov and himself. Shirov
explained how his thought process was different: ‘(...) Sometimes
during the game I don’t think about the position I have. I think about
other things, even when it’s my move. I may think about anything. I
may recall some songs of Sisters of Mercy or things like that. That’s
one of my favourite groups.’

10
The Sisters of Mercy? What a curious coincidence, I thought. I was
even more surprised when, sometime later, I learned that it had been
his compatriot Shabalov who had drawn Shirov’s attention to the
band. Two of my favourite chess players liking the same music as I
did: that felt a little special.
In the summer of 1996, I visited the Donner Memorial Chess
Tournament in Amsterdam, where I happened to witness the
conclusion of a game of Shabalov, who was one of the participants.
When he stood up, I decided to ask him whether he was still into the
music of The Sisters of Mercy. He said he was, and we got talking.
During the rest day of the tournament, I took him to downtown
Amsterdam to visit some record stores in search of new music. We
also talked about chess, of course, and I told him I still hadn’t found
a good way of playing against 1.e4 e5.
‘You should try the Center Game,’ he said casually, and he told me
the first moves:
1.e4 e5 2.d4 exd4 3. xd4

11
‘It’s a very easy opening to play, always underestimated by Black.
You castle queenside, your queen goes to e3 and then to g3, your
knight goes via e2 or h3 to f4, you push your kingside pawns and
ultimately you mate Black on g7.’ He showed me some lines,
mentioned a few games I should study and shared some novelties
with me. I felt like Salieri, assisting Mozart to compose the
Requiem.
After this conversation, I feverishly started studying this wondrous
opening in which White’s queen enters the centre as early as move
3. Perhaps my fondness for the White Queen in Through the
Looking-Glass made me take a liking to this opening right from the
start. Using early computer programs such as Fritz and ChessBase, I
created my first digital opening preparation. I found it surprising
how little was known about this opening – and how little I knew
about it myself. Most commentators seemed to be prejudiced against
it, giving harsh verdicts on White’s early queen adventures.
I tried the opening in blitz and rapid games at my local chess club,
in Amsterdam chess cafes and on the Free Internet Chess Server
(FICS), which was the go-to place for online chess back then. As I
gained my first successes with it, I got more and more confident that
this was the line that solved all my problems against 1...e5. It was
astounding, beyond belief: many of my opponents simply had no
idea. Here is an early rapid game I played with the Center Game,
against one of the strongest players of my club. (The game doesn’t
do this outstanding player justice, but he was the first to laugh about
it afterwards.)
Arne Moll
Roy Dieks
Amsterdam rapid 1996
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 b6?! 5. c3 c5?! 6. g3
f6? 7. d5 1-0

12
I remember the thrill I felt when I played the Center Game for the
first time in an important league match. I pushed my d-pawn in
response to 1...e5, then brought my beloved White Queen out on the
very next move. Some of my team members, strong veterans with a
classical chess education who had decades of experience at the
highest national level, shook their heads at such naivety. But for
once, I persisted, and gradually even my most sceptical teammates
had to admit that there was something to this unorthodox opening.
First studying it with members from my club, Max Euwe
Amsterdam, I later also had the honour to analyse the Center Game
with grandmasters. The strongest of them was Alexei Shirov, fellow
admirer of The Sisters of Mercy and former number two in the
world. I talked to him a couple of times during tournaments in the
mid- and late 1990s and sometimes played blitz games against him
on the Internet Chess Club. Despite having played the Center Game
once himself (see Chapter 1) and trying to keep an open mind,
Shirov was very critical of it. I learned a lot from him about the
opening – especially that Black should never be afraid to sacrifice
material for counterplay.
13
Because of Shirov’s sceptical views, I came to believe at the time
that the opening actually had more flaws than perks, and that it was
too risky to play in serious games. In March 1999, I wrote a critical
article about the most problematic lines for White in Kaissiber, the
German chess opening magazine edited by Stefan Bücker. For lack
of suitable alternatives to 1...e5 , I continued to play it from time to
time until the mid-2000s, and in blitz and rapid games, it was still an
excellent weapon. But I didn’t really trust it anymore. There were
new developments, for sure, but I’d somehow lost interest. And so it
was with my musical tastes. The Sisters of Mercy had long stopped
making studio albums, and my interests turned to other genres, and
other aspects of life. I got married and became a father.
In the summer of 2009, my attention was drawn to the opening
again. Someone emailed me the score of a game with the Center
Game played by the young Russian player Ian Nepomniachtchi. I
felt a pang of nostalgia. Inspired, I wrote a piece called Finding
Nepo (on an Old Laptop) for the Chessvibes website, in which I
recounted my earlier experiences with the opening and took a look
at Nepo’s contributions. ‘After all those years,’ I wrote, ‘I finally
feel like I don’t have to be ashamed anymore of my preference for
this crazy opening, even if in the end it turns out to be incorrect.’
For many years, I avoided the Center Game in my serious games.
Only during the COVID-19 pandemic, which started in early 2020,
did I change my mind for good. While the whole chess world was
forced to stay at home and play from behind a screen or on their
smartphone, I realized that there were still so many impossible
things to discover ‘before breakfast’ in this beautiful opening that I
should finally stop caring about whether it was correct or not. I was
hooked on the Center Game, and it was time to embrace it once
more. And so I did. I rediscovered the joy of playing it again and
found that a new generation of grandmasters had started to
experiment with new interpretations. I was beginning to think this

14
development might be a great reason to write a book to present a
new view on this old opening.
Or maybe my real reasons were less obvious, safely hidden in
Mikhail Tal’s deep dark forest ‘where 2+2=5’. Elusive and abstruse,
like the lyrics of The Sisters of Mercy. For nearly thirty years, I
have treated the Center Game as my own private Neverland, the
imaginary island of Peter Pan, where children don’t grow up. But I
was wrong.
The Center Game shouldn’t remain a secret place, hidden from plain
sight. It deserves to be seen in broad daylight.

15
Chapter 1
From the Middle
Ages to Y2K
The White Queen came running wildly through the wood, with both
arms stretched out wide, as if she were flying.
Lewis Carroll, Through the Looking-Glass

Early queen adventures

Medieval chess was a slow game, in which the queen and bishop
were insignificant, static pieces. But around the last quarter of the
15th century, the rules changed: these two pieces, especially the
queen, gained increased powers, transforming the game
dramatically. In the context of the Center Game, the evolution of the
queen’s mobility – and also thinking about the queen’s mobility – is
of particular interest.
The French manuscript Le Jeu des Eschés de la Dame, Moralisé,
written by an unknown author and dated around the year 1500,
describes a game played between the Devil as White and a Lady
(whose soul is at stake) playing with the black pieces.

Satan, perhaps hoping for a quick mate on f7, tried his luck with:
Satan
Lady
1.e4 e5 2. h5

16
The queen comes out menacingly enough, but the Lady didn’t mind:
she played...
2...g6
... and immediately, Lucifer lost his calm, producing the howler...
3. c4?? gxh5
... and the Lady won easily.
I find it hard not to sympathize with the devil: haven’t we all
succeeded and failed to deliver the Scholar’s Mate at some point in
our early chess career?

Everyone who learns the rules of chess is told that the queen is the
strongest piece on the board. Her powers are amazing. She is worth
nine pawns; she is majestic and extremely mobile, and has
seemingly unlimited powers. Then comes the big disappointment:
we hear that it’s actually unwise to move your queen too early in the
game. It is, to quote Kasparov and Keene (in Batsford Chess
Openings 2, 1989), a ‘breach of the basic rules of development’.
You should develop your knights and bishops first, occupy the
17
centre, bring your king into safety, and only then start thinking
about your queen. This is what everyone learns, and it usually
sticks.

It’s also a big misunderstanding.

Consider castling queenside. This is a move that is completely

routine in many openings, perhaps most prominently in the ever-
popular Open Sicilians, where White often castles queenside before
finishing the development of his kingside pieces. But you can’t
castle queenside without moving your queen. So how bad can it be,
really, to bring out the queen so early?
John Watson discusses early queen excursions in a separate chapter
in his book Chess Strategy in Action (2003). ’Often an early queen
development comes with gain of space and staking out territory in
the centre,’ he writes. ‘(...) It seems to me that most early queen
moves are in positions with one’s centre pawns on the second or
third ranks, and possibly but not always with another pawn on the
fourth rank.’ Watson mentions the Scandinavian Defence as a prime
example. Indeed, it’s the most well-known opening in which one
side brings their queen out very early in the game, sometimes with
the purpose of castling queenside.
The Scandinavian is, in fact, one of the oldest openings known to
chess historians. It features prominently in the ancient Scachs
d’Amor manuscript. Written in Valencia, Spain, around 1475, this is
often regarded as the first book about modern chess. The game
described in the manuscript, allegedly played between two gods
from antiquity, Venus and Mars, demonstrates the newly gained
powers of the queen and bishop. It starts with the moves:
1.e4 d5 2.exd5 xd5

18
3. c3 d8
The new movements of the bishop and the queen, characterized by
what chess historian Peter Monté called diagonal dynamics, changed
the nature of the game completely. This was especially true of the
powers of the queen – so much so that, in order to distinguish the
new game from its medieval predecessor, it was sometimes called
dela dama, de la donna or de la dame (‘of the queen’ in Spanish,
Italian and French respectively). It is often assumed that the Spanish
Queen Isabella of Castille (1451-1504) inspired the movement of
the new queen. In other early sources, the new game was called ala
rabiosa or enragée (raging or enraged), emphasizing the ‘mad’
movement of the queen.
One might expect a lot of ‘mad queen’-related openings (such as the
Scandinavian Defence) in the oldest chess literature, but further
research is disappointing in that respect. In the Scachs d’Amor
manuscript, Black’s daring second move is actually criticized, and
his third move, cautiously retreating the powerful queen, is praised.

19
It is an early example of a way of thinking about chess openings that
persists until this day.
The move sequence introducing the Center Game, 1.e4 e5 2.d4 ,

was first mentioned in the Rotilio Gracco manuscript from around

1580, but not analysed further. The Traitté de Lausanne (1690)
mentioned various openings and gave general advice to players. One
of them was a warning ‘against acquiring too great a fondness for
[the] Queen’. About half a century later, Captain Joseph Bertin
advised readers in his The Noble Game of Chess (1735): ‘Never
play your Queen till your game is tolerably well opened, that you
may not lose any moves.’ Bertin may thus have been the first chess
author to explicitly state that it’s bad to move your queen too early
in the game.
Around the same time, a contemporary of Bertin’s broke the trend
of not taking early queen moves seriously. The first chess author to
publish an analysis of the Center Game was Philipp Stamma.

The first pioneers of the Center Game

20
Philipp Stamma (ca. 1705-ca. 1755) was an enigmatic figure, but
an important one in the history of chess. He was born as Fathallah
Shtamma in the city of Aleppo, in the Ottoman Empire (currently
Syria). As a translator of Oriental languages, probably low on cash,
he came to Paris as a young man to work for the French
government. He was also a strong chess player and published a book
on the game in 1737: Essai sur Le Jeu des Echecs. A few years
later, he went to London. His book was translated into English as
The Noble Game of Chess in 1745, the same book title as Bertin’s.
Stamma was a regular visitor to Old Slaughter’s Coffee House, a
place frequented by many strong chess players. In 1747, Stamma
lost a famous match against Philidor there. (I went to see the
location where Old Slaughter’s Coffee House used to be – before
the building got demolished in the 19th century – in St. Martin’s
Lane, London. There is still a coffee shop, but it’s a Pret a Manger
now.)
Stamma’s most important contribution to chess was the introduction
of algebraic notation, but he was also an excellent opening
theoretician, who was the first to see the value of the Queen’s
Gambit. Stamma called the opening starting with 1.e4 e5 2.d4 the
‘Pawn-Close Game’. In a just world, the Center Game would have
been called the Stamma Opening. Although he was dismissive of
taking back the pawn on d4 with the queen, no doubt influenced by
earlier negative sentiments, his analysis of it was quite sophisticated.
It’s worth looking at his variations in some detail, as they feel
relatively modern.
Stamma (analysis)
‘The Noble Game of Chess’, 1745
1.e4 e5 2.d4
Stamma writes: ‘This move, though frequently played, is not a good
one, as it brings out the Queen too early in the game.’
2...exd4 3. xd4 c6 4. e3
21
In the next variation, Stamma analyses 4. d1 c5 5. c4 d6 6. c3
e5 7. b3 g4 8. ge2 h4 9.0-0 g5 10. e3 0-0-0 11. d2 f3+
12.gxf3 xf3: ‘White cannot retrieve the game, as he can not
prevent the Black from playing the Queen to g4 or h3.’

4...g6
A profound reply to White’s set-up, which is still popular today.
5. c4
5. c3 leads to the old main line of the fianchetto variation.
5.h4!? is the modern move, which we will be looking at in Chapter
5.
5... g7 6. e2 ge7 7.0-0 0-0 8.f4
After 8. bc3 White has a decent position.
8...d5!
Excellent! Stamma already understood that this is an important
central break for Black.
9.exd5 xd5 10. xd5?

22
But this is poor. It can’t be right to give up the bishop just like that.
10... xd5 11. bc3 c4 12. d3
12. f2 was better than exchanging the queens and getting an
isolated d-pawn.
12... xd3 13.cxd3 f5
‘Black will ultimately gain the d3-pawn and consequently will have
the best of the game.’
After this, chess literature turned silent again on the topic. In his
famous Analyse du Jeu des Echecs from 1749, Philidor only
analysed early queen moves in the context of the lines 2. c4 c6 3.d4
exd4 4. xd4 and 2.c3 d5 3.exd5 xd5, where he noted that Black
now has the initiative as White doesn’t have the move b1-c3
(while Black does have ... b8-c6 in the Center Game).
The first recorded appearance of the Center Game in an actual game
is from 1836. The game was played in Paris, most likely in a
simultaneous exhibition. The player with White isn’t exactly a
household name among chess players: Victor-Joseph Étienne,
called de Jouy (1764-1846), was a well-known French dramatist at
the time. He served as an adjutant-general during the outbreak of the
French Revolution, but turned to literature when he was accused of
treason. The player with the black pieces was the world’s strongest
player in the first decades of the 19th century: Louis-Charles Mahé
de La Bourdonnais.
The game is interesting from a historical perspective because De
Jouy was the first to castle queenside and thus set up a kind of
‘proto’-Center Game attack.
Victor-Joseph Étienne de Jouy
Louis de La Bourdonnais
Paris 1836
1.e4 e5 2.d4 exd4 3. xd4 c6 4. d3

23
This idea was entirely ignored by later theoreticians, but has become
topical in recent years.
4... f6
4...d5! is the critical move, as we will see in Chapter 6.
5.f3
f2-f3 is often part of White’s plan, but here it’s not to be
recommended. Nor are White’s next moves.
5. g5 d6 6. c3 was played in a correspondence game between the
City of Buffalo and the City of Albany in 1895.
5. c3 leads to the Vienna/De Jouy Hybrid, which we discuss in
Chapter 6. The first game with this move went 5... e7 6.f4? ( 6. f4
) 6... b4 L.Paulsen-Berger, Breslau 1889.
5... c5 6. e3? e7
Mising the simple 6... b4, winning at least a pawn.
On the next two moves, this knight sortie would still have been
winning for Black.
7. e2 0-0? 8. c3? e5? 9. d2 xe3 10. xe3 d6 11.0-0-0
24
This looks like a regular Center Game. White is better.
11... e6 12.b3?
The thematic 12.g4! gives White a very nice game and is now a
standard way of conducting an ideal Center Game attack.
Also possible was 12.f4.
12...a5 13. a4
Heading very much the wrong way.
13.a4 was better.
13...b5?!
13... fb8 was almost winning.
14. xb5 fb8

25
15.c4?
Losing a piece.
After 15. c3! it’s actually not at all clear how Black continues the
attack.
15...c6
It’s all over.
16. b6 cxb5 17. xa8 xa8 18.cxb5 a4 19. b2 axb3 20.axb3 d5
21. a1 xa1 22. xa1 a3+ 23. b1 dxe4 24. c2 a2+ 25. d1
xb3+ 26. e1 d3+ 27. f1 b1+ 28. e2 e1#
There were only brief references to the Center Game in chess
publications in the following few decades. Carl von Jaenisch, an
influential Finnish-Russian chess theorist, mentioned the opening in
his Chess Preceptor (1842), saying that 3. xd4 ‘loses you at least "a
time" ’ (meaning a tempo) and that Black, after 3... c6 4. d1 c5,
‘has two pieces out, while yours are all at home.’ The English chess
author William Lewis classified the Center Game (rather
amusingly!) under the chapter ‘Irregular Openings’ in A Treatise on

26
the Game of Chess (1844). He wrote that 2.d4 ‘is not a very good
move, but may be played without much danger.’ He then mainly
looked at 3.c3, not deeming 3. xd4 worthy of attention. Howard
Staunton faced the opening in 1851 in a game against Bernhard
Horwitz, who played 4. d1? and lost.
In some lines of the Philidor Defence ( 1.e4 e5 2. f3 d6 ), which
were played regularly in the 1850s, White played his queen to d4
and then to e3 at an early stage.
Here is the beginning of a game Moheschunder Bannerjee
(sometimes spelled as Bonnerjee) played in a match with the
Scottish master John Cochrane.
Moheschunder Bannerjee>
John Cochrane
Calcutta 1855
1.e4 e5 2. f3 d6 3. c4 e7 4.0-0 f6 5.d4 exd4 6. xd4 0-0
7. c3 c6 8. e3

27
By move transposition, we have reached a position that will be
analysed in Chapter 4, when we look at ways to combat the set-up
with ... f8-e7 against the regular Center Game.
Very little is known about the Bengal player Moheschunder
Bannerjee, also known as ‘the Brahmin’. He came from a local
village near Calcutta and originally only played traditional Indian
chess, in which pawns could only move one square from their initial
position and castling was not allowed. His style of play in European
chess proved revolutionary. Bannerjee stood at the foundation of
modern ‘Indian’ defences after 1.d4 and was also the first to play the
Pirc Defence. His contribution to modern chess opening theory
remains undervalued and under-investigated to this day.
Paul Morphy, in a game against Johann Löwenthal, also took out his
queen out to d4, albeit one move later than in the Center Game.
Paul Morphy
Johann Jacob Löwenthal
Paris 1858
1.e4 e5 2. f3 d6 3.d4 exd4 4. xd4

28
4... d7 5. e3 f6 6. c3 e7
Now, 7.0-0-0, analogous to White’s main plan in the Center Game,
would have led to a good position for White. Instead, Morphy went
for a plan involving kingside castling, and lost. (This game may
have been the reason why I never seriously studied the Center Game
during my preoccupation with other 19th century openings in my
early chess years.)
Only from the late 1870s onwards did the Center Game become
accepted as a ‘normal’ opening. It was still called ‘Center Gambit’
in those days, for example by William Cook in 1874 – maybe
because, like many others before him, he considered 3.c3 (a ‘real’
gambit) the only proper way to continue. In Dutch, the opening is
still called Middengambiet. In any case, the opening was now
increasingly played and analysed by leading players.
Two Center Game pioneers from this period deserve special
mention: Wilfried Paulsen and Szymon Winawer.
Wilfried Paulsen (1828-1901) was the elder brother of Louis
Paulsen (who played matches against both Paul Morphy and Adolf
Anderssen). He was a potato farmer from Nordrhein-Westfalen,
Germany, and published various articles on potato farming in
scientific magazines. Wilfried Paulsen first played the Center Game
at the 1881 Chess Congress in Berlin. This tournament proved a
milestone for the Center Game as several others played it there too,
including the German player Fritz Riemann (1859-1932).
Here is Paulsen’s game against Johann Berger, in which he castled
queenside and conducted a standard attack against the black king.
Wilfried Paulsen
Johann Berger
Berlin 1881
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 d6
Passive, but not bad. The move is frequently seen at lower levels.

29
In the February 1884 issue of Deutsche Schachzeitung, Berger
analysed the stronger 4... f6! which is still called the Berger
Variation, though, in fact, the move was first played by the
Hungarian chess master Josep Noa (1856-1903) in this very same
tournament. Berger was the first to articulate that Black doesn’t
need to fear 5.e5 ...

analysis diagram
... because of 5... g4! 6. e4 ( 6. e2 d6! and Black has a good
game –Taubenhaus-Gunsberg, Hamburg 1885) and now Black can
simply take on e5: 6... gxe5! ( 6...d5 7.exd6+ e6 is what they
played in the old days, and is also OK for Black; it first occurred in
Schallopp-Minckwitz, Berlin 1881) 7.f4 d5 and Black’s lead in
development is too great.

30
5. d2 e7 6. c3!?
An original but slow set-up. In Chapter 2, we will look at set-ups
with b2-b3 and c1-b2.
There was nothing wrong with 6. c3 f6 7.0-0-0, which first
occurred in a correspondence game between the German cities of
Halle and Magdeburg in 1887.
6... f6?!
6... f6 7. d2 0-0 8.0-0-0 e8 and Black has a pleasant position.
7.f4

31
7... e7?!
7...d5! 8.exd5+ e7 9. xe7+ cxe7 10. xf6 xf6 and Black will
likely equalize in due course.
8. b5?!
The bishop usually goes here only when Black has castled
queenside.
8. f3 gives White a slight edge.
8... d7
Better was 8... f5!, profiting from the unprotected position of the
queen on e3.
9. f3 h6?! 10. xf6 xf6 11. c3 0-0?
Berger should have castled queenside: 11...0-0-0 and White is still
better, but it isn’t the end of the world yet.
12.0-0-0

32
We have a typical Center Game position with opposite-side castling.
White has a simple attack on the kingside.
12... fe8 13.h3 e7 14.g4 f5?!
Passive defence was called for: 14... f8! 15.g5 f5 16. f2 fe7,
though White’s attack is still strong after 17.f5.
15. c4+ e6 16. d5 xd5 17. xd5+ h8 18.gxf5
18. g5! fxe4 19. he1 followed by d2, and White will soon crash
through.
18... xf5 19. c3?!
The white queen is often nicely tucked away on f2 in this opening:
19. f2! with a huge advantage.
19... g3 20. he1 d8
20... h5 was the only move to hold on.
21. g5 h5 22. f3 f6 23.e5 dxe5 24. xe5 f8 25. de1 h6

33
26.h4!
Another common theme in many lines in the Center Game: leaving
the knight on g5 en prise, and here the white rook isn’t even on h1.
26... c6?!
26... xe5 27.fxe5 b4 loses to 28.c3 .
27. xc6
Even stronger was 27. f7+! h7 28. d3+ g6 29. g1 g7 30. xc6
bxc6 31. xe8 xe8 32. e5 and Black can no longer defend g6.
27...bxc6 28. e6 b4 29.c3 b6 30. g2 e7
White has played a nice game so far, but now the quality of his play
goes downhill.
31. g6?
The beautiful but difficult-to-spot 31. d8! xe5 32. f7+ h7
33.fxe5 would have been much cleaner.
31... b8?

34
After 31... f2!, attacking f4 and threatening 32... b8, the position
would actually have been rather unclear.
32. 5e2??
A very strange blunder. Perhaps White played the other rook to e2,
but this doesn’t match with the rest of the game.
32... a6??
Black doesn’t see it either. 32... xe6! actually wins for Black.
33. b1 d5 34. a1 a4 35. g4 d7? 36. c5 xc3 37. xa4
Black resigned. With the other rook on e2, ... d1 would now have
been mate.
Paulsen tried ‘his’ line (i.e. 4. e3, which we will call the Paulsen
Variation from here on) in other tournaments as well, but didn’t
score well with it. Looking closer at his games, my conclusion is
that Paulsen didn’t really have any consistent opening set-up in
mind. He was simply trying to develop his pieces creatively and get
a playable position, without any apparent ‘masterplan’. This is a risk
that every Center Game player should be mindful of. You can only
play this opening if you know what you’re doing.
Just look how Paulsen completely messed up in a game against his
compatriot Emil Schallopp (who also played the Center Game from
time to time):
Wilfried Paulsen
Emil Schallopp
Hamburg 1885
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 f6

35
5.h3?
Simply bad.
5... g4 was not a threat, but Paulsen apparently thought it was, as
earlier, in the 7th and 10th rounds, he had tried 5. e2?! against
Tarrasch and Riemann respectively. Perhaps he had an early g2-g4
in mind.
5... b4+ 6.c3
The move c2-c3 isn’t often seen in the Center Game, but here it is
definitely best.
6... a5 7. g3 e7

36
8. xg7?
Hitting the self-destruct button (avant la lettre). Best was something
like 8. e3, but clearly White’s opening ‘experiment’ has failed
miserably. Black has a big lead in development and will push ...d7-
d5 soon.
8... g8 9. h6 xe4 10. e3 d5

37
After just 10 moves, Black is already completely winning. This is,
sadly, a fate that awaits many too overly creative White players in
the Center Game. Beware!
11. d1 f5 12. e2 0-0-0 13.g4 d4!
Paulsen could have resigned here.
14.cxd4 b4 15.gxf5 e1+ 16. c2 b4+ 17. b3 d1+ 18. a3
b5 19.b3

38
19... g6!
A nice interference move. Schallopp wants to play ... d3 without
being bothered by a6+.
20.fxg6 d3 21. d2 xd2 22. xd2 xd2
Followed by ... b4 mate, so White resigned.
This was the last recorded game Paulsen played with the Center
Game, and it’s not too hard to understand why he was fed up with it.
But chess history has been kind to Wilfried: the Center Game with
4. e3 is still called the Paulsen Variation. There’s even a book (in
German) on the Center Game in which the author, Mario Ziegler,
names the entire opening after him: Paulsen Eröffnung (2010).
The second pioneer from this period is Szymon Winawer (1838-
1919). Winawer was born in Warsaw, Poland, and his name is, of
course, still well-known in modern chess culture because of the
dangerous variation in the French Defence ( 3. c3 b4 ) named
after him. Winawer played the Center Game at various times in his
career and managed to beat the two strongest players of his time
with it. In the second international Vienna 1882 tournament, he won
39
a nice game against Mikhail Chigorin (who played the rare move
3... f6 ). Winawer went on to finish shared first in the tournament
(together with William Steinitz).
Even more impressive was Winawer’s victory over Steinitz in the
Nuremberg 1896 tournament, a game praised as ‘audacious and
elegant’ by Tarrasch in the tournament book, and still theoretically
relevant today.
Szymon Winawer
William Steinitz
Nuremberg 1896
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 f6 5. c3 b4 6. d2 0-0
7.0-0-0 e8!

Steinitz chooses the critical set-up for Black against the Center
Game, immediately posing the threat of ...d7-d5 thanks to the nasty
pin along the e-file. It was first played by Henry Bird.
8. c4

40
This natural developing move was a novelty at the time. In later
years, it was somewhat neglected as other moves were tried for
White. But the latest engines rather like it.
8.f3?! d5! was good for Black in Fleissig-Bird, Vienna 1882.
The sharp 8. g3 was first played by Tarrasch and will be seen in
the next game.
The tricky 8. f4 is an idea introduced by Judit Polgar in the early
2000s, which we will see in Chapter 2. After 8... xc3 9. xc3 xe4
10. c4, we are back in the current game.
8... xc3
8...d6 was played by Janowski later that year, and will be discussed
when we look at the game Tartakower-Reshevsky, Stockholm 1937.
9. xc3 xe4

It’s not immediately clear what the downsides of this pawn grab are,
but...
10. f4!
Stepping out of the discovery and targeting f7.

41
10... f6
10... e7 is the main alternative, after which White can win back his
pawn with 11. xf7+ xf7 12. xe4.
11. f3
White has fantastic bishops and a lead in development.
11...d6 12. g5 e6

13. d3!?
Very ambitious. Winawer wants to create weaknesses on the
kingside.
13. xe6 fxe6 14. he1 would have led to a typical position where
White has more than enough compensation for the pawn and is in no
rush at all. The black pawn centre will become very vulnerable in
due course.
13...h6 14.h4?!
Spectacular, but questionable.
Objectively better would have been 14. xe6 xe6 15. f5 e8
16.h4 d5 17. h3! with compensation.
42
14... d5?!
It was better to play the other knight: 14... e5! 15. xe5 dxe5
16. xe5 and now the very accurate 16... d5! 17. f4 e7 would
have led to a better position for Black.
15. h7+ h8 16. xd5! xd5 17. e4 f6?
A huge mistake by Steinitz.
17... xe4! was the only move.

analysis diagram
Black understandably feared 18. f5 ( 18. xe4! e5 19. d1 xe4
20. xe5 h7 21. xf7 f8 22. xf8+ xf8 23. d4 with equality),
attacking the d5-bishop and then the e4-rook, but White’s attack is
not decisive after 18...hxg5 19.hxg5+ g8 20. h7+ f8 21. xg7+
e8 and Black wins, although this is by no means obvious.
18. xd5 fxg5 19.hxg5

43
White is winning. The two bishops are simply too strong to resist.
19... e5 20.g6
Black resigned as 21. xh6+ will decide.
In the last years of the 19th century, many new ideas in the Center
Game were discovered and tried out in practical games by the
strongest players in the world. Chigorin played the opening on a
number of occasions with White. One of Chigorin’s greatest rivals,
Siegbert Tarrasch, too, contributed to the theoretical
developments.
Tarrasch launched one of the most important (and long-lasting)
ideas in the following game, which he analysed in his famous book
Dreihundert Schachpartien ( 300 Chess Games, 1895):
Siegbert Tarrasch
Max Kürschner
Nuremberg 1889
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 f6 5. c3 b4 6. d2 0-0
7.0-0-0 e8 8. g3!?

44
Tarrasch introduces a concept that would grow into one of the most
important variations of the Center Game in the 20th century: ‘White
sacrifices a pawn to start the attack.’
8... xc3?
A weak move, and Tarrasch rightly gives it a question mark.
8... xe4 was first played by Semyon Alapin in a game against
Eftifeev (St Petersburg 1906). White can’t take on e4 here.
Although still complicated, it would later be considered the
refutation of Tarrasch’s idea and greatly diminished the popularity
of the Center Game for decades.
8... xe4! 9. xe4 xe4 will be seen in Marshall-Pillsbury, Buffalo
1901.
9. xc3 xe4 10. d3 e8 11. f3

45
These kinds of positions made the Tarrasch Gambit popular in the
following years.
11...d6 12. g5
Objectively better was 12. he1 and White has a strong initiative.
12...h6 13. h7+

46
13... f8
If 13... xh7, 14. e6 ends the game immediately.
14. e4 g4??
A blunder which Tarrasch doesn’t comment upon.
The only move to maintain the balance was 14... h5! and now
15. f3 e5, intending 16. xh5? g4.
15.h3 ge5 16.f4 g6 17. xg6 fxg6 18. xg6 e7 19. xd6
A pretty move, but not a necessary one.
19...cxd6 20. de1

47
20... e5?!
20... f7 21. xg7+ xg7 22. xe8#;
20... e6 loses to 21. xe6.
It was better to return the piece by playing 20... f5, but White is
still winning after 21. xf5+ f7 22. h7.
21.fxe5 d5 22.h4
Preventing 22... g5+. The rest is a piece of cake for Tarrasch.
22... e6 23. h7 ec8 24. h8+ f7 25. ef1+ g6 26.h5+ g5
27. d2+ g4 28. h7 1-0
This was a crucial moment for the theory of this opening, and we
will return many times to the ‘Tarrasch Gambit’ throughout this
book.
The last decades of the 19th century also saw the emergence of
another way of playing for White. The Hungarian Isidor Gunsberg,
who lost a World Championship match against Steinitz in 1891,
experimented with the move 4. c4 in combination with 0-0-0,
which has subtle advantages over placing the queen on e3.

48
Isidor Gunsberg
James Mortimer
London 1887
1.e4 e5 2.d4 exd4 3. xd4 c6 4. c4

This was played once before in a game between (another) Mortimer

and Von der Lasa in 1837, but Gunsberg was the first to combine it
with the idea of castling queenside. Having the queen on c4 has the
advantage that an early ...d7-d5 is prevented.
4... f6 5. c3 b4 6. d2 d6
6...0-0 7.0-0-0 e8 is the critical test, like in the Paulsen Variation,
but here we see the advantage of having the queen on c4: 8.f3 is
now possible, as ...d7-d5 is less of an issue. White ideally follows up
with g2-g4 and h2-h4.
7.0-0-0

49
According to my database, this perfectly natural position didn’t
occur again in practice until more than 130(!) years later, in 2019.
7...a6
The plan with ...b7-b5 is less effective here, as the knight on c6 will
hang.
7...0-0 8.f3 e6 9. e2 was played in Wei Yi-Bok, PRO League
rapid 2023.
8.f4 d7
Another slow move.
9. f3 b5 10. e2
10. d3 was even better.
10...0-0 11.h3 e8 12.e5

50
12... h5?
Hoping for a fork on g3, but Gunsberg ignores the threat.
12...dxe5 13.fxe5 d6 14. f4 with an unclear position.
13. g5! g6?
If 13... g3, then 14. d3 with a double attack on h7 and the knight
on g3, or 14... f5 15.g4, also winning.
13... f6 was the only move, but White is better after 14. ce4.
14. ce4?!
Perhaps Gunsberg didn’t entirely trust 14.g4! g3 15. e1 xh1
16. h4 h5, but White is simply winning after 17.gxh5 f5 18.hxg6
xg6 19. d3 and there is no defence.

51
14...f5?
14...dxe5 had to be tried, though White is still winning, for example:
15. xb4 xb4 16.f5! f6 17.fxg6 hxg6

analysis diagram

52
and now the elegant and unusual 18. xd7! xd7 19. xh5! gxh5
20. xf6+ g7 21. xd7.
15.g4!
White’s attack is winning.
15...fxe4 16.gxh5 dxe5 17.hxg6 hxg6 18. xb4 xb4 19. xe4
xa2+ 20. b1 f5 21. xd8 xe4 22. xe8+ xe8 23. xe4
It wouldn’t have been shameful to resign at this point.
23... b4 24. f6+ f8 25. xe8 xe8 26.fxe5 c6 27. d3 e7
28. e4 1-0
The move 4. c4 came to be known as the ‘Hall Variation’ after the
little-known British player John Edmund Hall (1853-1941), who
played it twice, after Gunsberg, at the 1888 meeting of the West
Yorkshire Chess Association. The line has been almost completely
neglected by theory until very recently.

The Center Game comes of age

At the turn of the 20th century, there was an explosion in chess
publications: opening theory entered the ‘modern’ era. There were
now many standard theoretical works available, such as Steinitz’s
Modern Chess Instructor (1889) and Von Bilguer’s Handbuch des
Schachspiels (the 1891 edition was prepared by Schallopp and Louis
Paulsen, Wilfried’s brother).
A game played just after the turn of the century between two
American giants, Frank Marshall and Harry Nelson Pillsbury, turned
out to be of major theoretical importance.
Frank Marshall
Harry Nelson Pillsbury
Buffalo 1901
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 f6 5. c3 b4 6. d2 0-0
7.0-0-0 e8 8. g3 xe4! 9. xe4 xe4

53
Black’s play makes perfect sense. He simply takes a pawn and asks
White: where is your compensation?
10.c3
The most natural move. During the great ‘revival’ of the opening in
the 1990s, this was generally considered to be White’s best try.
10. f4 is the main alternative. It was first played by Mieses in his
match against the great José Raúl Capablanca in Berlin 1913. The
Cuban responded strongly with 10... f6! 11. h3 d6 12. d3 d4!
13. e3 and now Capablanca, who had played a model game so far
and was close to refuting the Tarrasch Gambit, went wrong:
13... g4? (correct was 13... f5! 14. f3 e8 15. g5 e5 and Black
had a great position in Wagner-Näter, ICCF email 2018), when the
tactically alert Mieses responded attentively:

54
 analysis diagram
14. g5!.
An unexpected shot which turns the tables.
14... xe3 15. xg4 e2+? 16. xe2 xe2 17. e4 xe4 18. xe4
g5+ 19.f4 and White was winning, though he later lost.
10... f8
A reasonable retreat, but there is a better one.
Black’s could have dealt a real blow to White’s set-up with
10... e7!. The problem is the threat of 11... h4. We will look at
this in more detail in Chapter 3.
10... d6 11.f4! creates the kind of compensation White is aiming
for.
11. d3 e6
Black can get away with 11... a4!, but that would not be everyone’s
cup of tea. White still has practical compensation after 12. h3.
12. f3 h6 13.h4 d5

55
14. h2?
A typical manoeuvre, but here White had to try another ‘typical’
move: 14. g5!. In the spirit of the position! The situation is highly
unclear after 14...hxg5 15.hxg5 g6 16. h4 g7 17.f4, when White
is a full piece down but has a dangerous attack.
14... d6 15. g1 f6 16. g5!
White’s best practical chance, and Marshall grabs it.
16...hxg5 17.hxg5 d8 18. h7+ f8 19. f5

56
19... e8?
After the cool 19... e7!, White doesn’t have any real compensation.
20. xc8?
20. e1! e6 21.f4! would have been highly unclear.
20... xc8 21. h8+ e7 22. e1+ e5 23. xe8+ xe8 24.f4
White wins back the piece, but the endgame is still good for Black.
24... g4?
Pillsbury should have played 24... e6! 25.fxe5 d7.
25.fxe5?
25. c5!.
25... d7 26. h2?! e8 27. f4 c8 28.g3 d4 29. c2 dxc3
30. xc3 e6 31.a3 f6 32.gxf6 gxf6 33.b4 xe5 34. b1 d5?!
Allowing counterplay due to the pin on the e-file.
35.g4 h8 36. c1 c6 37. xe5 fxe5 38.g5 h2 39. c2 h1+ 40. c1
e4+ 41. b2 g2+ 42. b1

57
42...b6?
A blunder. Pillsbury sees that White can’t take on c6, but forgets
about the pawn on e5.
42... h2 would have won.
43. xe5 xc1+ 44. xc1 f1+ 45. d2 f2+ 46. d3 f3+
47. e3 f5+ 48. e2 d7 49. d3+ e6
Or 49... xd3+ 50. xd3 e6 51. e4 c5=.
50. c4+ d6 51. d4+ e6 52. e3+ f7 53. f3 g6 54. xc6+
xg5 55. c7 e4+ 56. d2 ½-½
One of the first loyal 20th-century supporters of the Center Game
was Jacques Mieses (1865-1954). Mieses is sometimes regarded as
the ‘eternal third’ of late German 19th and early 20th century chess,
after Tarrasch and Emanuel Lasker, the second World Champion.
Mieses was a fierce attacking player who could be dangerous to
anyone. The Center Game, with much room for creativity, suited his
style, and he played it against later World Champions Alexander
Alekhine and José Raúl Capablanca, though he lost both games.

58
There are similarities between Mieses and Wilfried Paulsen in the
sense that they both persisted in playing the opening despite
suffering setbacks with it. Their track record should be a warning to
readers who consider picking up this opening and may have become
over-optimistic about White’s chances.
Let’s look at Mieses’s loss against Alekhine. In Alexander
Alekhine’s Best Games (Algebraic Edition, 1996), the fourth World
Champion commented on White’s third move: ‘It is quite evident
that such displacements of the queen at an early stage in the opening
are not likely to reap any advantage. However, Black is compelled
to play with precision, in order to give his opponent no time to start
an attack against the kingside, or even in the center. For, no doubt,
the white queen installed at g3 (via e3) would exercise pressure on
Black’s kingside if he eventually castles on that side.’
Alekhine counters the Center Game in very natural and harmonious
fashion, developing his bishop to e7 and going for a quick ...d7-d5,
which had already been played thirteen years earlier by Chigorin in
a game against James Mason. The game is a good example of what
can happen if Black is able to free himself and takes over the
initiative.
Jacques Mieses
Alexander Alekhine
The Hague 1913
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 e7
A modest but healthy move, preparing ...d7-d5. Mieses had played
this himself against Walbrodt in Berlin (1894).
5. d2
5. c3 f6 6. c4 is a way for White to prevent ...d7-d5.
5... f6
This is still one of Black’s most solid ways of playing against the
Paulsen set-up.
59
6. c3 0-0
It looks logical to castle, but Black could also have postponed it and
kept the option of castling queenside.
The immediate 6...d5 7.exd5 xd5 was the alternative. White is best
advised to play 8. xd5 xd5 9. e2 followed by either c3 or f4,
as first played by Sultan Khan against Frank Marshall in 1930.
7.0-0-0 d5!

The ultimate liberating move for Black in the Center Game, after
which he usually equalizes. Alekhine comments: ‘This advance,
which at first sight appears somewhat risky in view of the position
of White’s d1-rook, will on the contrary allow Black to extract the
maximum return from his advanced development.’
8.exd5 xd5 9. g3?!
A standard move, but the queen is not ideally placed on g3 in this
line.
9. f3 is analysed in Chapter 4 as perhaps White’s best alterative,
but the move is also of historical importance. After 9... e6 10. f4?!
( 10. ge2 xc3 11. xc3 d6 leads to unclear play),
60
 analysis diagram
Black played 10... xc3! in Lebedev-Rosenkrantz, Moscow 1901. A
move very silimar to the sacrifice Alekhine plays in the game!
11. xd8?! ( 11. xc3 b4 12. c4 e8 ) 11... xa2+ 12. b1
fxd8 and Black was close to winning (though the game was
drawn). It is possible that Alekhine was aware of this game, though
he didn’t mention it in his analysis.
9... h4!
Winning an important tempo against the white queen. ‘Black’s
advantage is now evident.’
9... e6 10. h6 f6 11. xd5 xd5 12. f3 is promising for White.

61
10. f3 e6
Alekhine points out that 10... xc3 11. xc3 g5+ 12. d2 c5!
would have been good for Black.
11. e3 xc3!
See the note to move 9. Alekhine, like Rosenkrantz, prefers piece
activity over his queen.
12. xd8 xa2+ 13. b1 axd8 14. e2 ab4 15. h3 fe8
‘Essential as a basis for all subsequent combinations.’
16. f4
16. d1 was better.
16... f5 17. c1 g6?!
17...h6!.
18.g4! e4 19. h3

62
19... f6?
Alekhine misses a nice tactical opportunity and allows Mieses to get
back into the game.
Strong was 19... xc2! 20. xc2 ( 20. xh4 xe3+ 21. a2 d5 )
20... xf2! 21. xf2 b4, although White can still hold on with
precise play: 22. f1 xc2+ 23. c1 b3 24. e1 with an unclear
position.
20. f3 xf3 21. xf3 e5 22. e2 c5 23. g1 c4 24.h4 d5
25. xd5 xd5 26.f4?
A mistake, weakening White’s control of the e-file.
26. d1! is not mentioned by Alekhine, but was still good for White.

63
26... d3!
‘Immediately exploiting the fact that the advance with f2-f4 has
weakened the e3-bishop’s defences.’ (Kasparov in My Great
Predecessors 1)
27. f3
27.cxd3 was called for. Alekhine now analyses 27... xd3 28. g3
d4 29. c2 xe3 30. xc4 ed8 and evaluates this position as good
for Black, but the computer is merciless and claims a small
advantage for White after 31. e4.
27... b5
‘Decisive!’ writes Alekhine, but that’s not the case.
28.cxd3
28.b3 a5 .
28... xb2+ 29. c1 cxd3 30. d1 c8

64
31.g5??
Collapsing under the mounting pressure.
31. e4!, as pointed out by John Nunn in the English translation of
Alekhine’s work, holds. Black has nothing better than taking the
perpetual with 31... b1+ 32. d2 b2+ 33. d1 b1+.
31... cc2
The rooks on the second rank decide.
32. e1 b1+ 33. d1 c3+ 0-1
Mieses’ victory over Rudolf Spielmann from the year before was an
entirely different story. Spielmann was famed for his romantic
approach to chess and was nicknamed ‘The Last Knight of the
King’s Gambit’. But in this game, it was Mieses who channelled his
inner Romantic against Spielmann’s fianchetto set-up. Mieses’ play
in this game looks fresh and modern, going for the early push h2-h4,
which would make neural network engines proud, as we will see in
Chapter 2.
Jacques Mieses

65
Rudolf Spielmann
Breslau 1912
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 g6
A logical way to develop for Black, already suggested by Stamma.
If White castles queenside, the bishop on g7 will be shining on the
long diagonal.
5. c3 g7 6. d2 ge7?!
The knight belongs on f6 in the Center Game.
7.0-0-0 0-0

8. c4
The direct 8.h4! was even stronger.
8...d6 9.h4 a5?
The knight is badly placed here.
Better was 9...h5 10.f3 e5 11. e2 e6 12. h3 with a nice position
for White.
10. e2 e6 11.h5

66
Now White’s attack is rolling.
11... d7
11... c4 12. xc4 xc4 13. h3.
12. g5!
Attacking the a5 and bringing the queen closer to the juicy h4-
square.
12... c4 13.hxg6
13. xc4 xc4 14. h4 was more accurate.
13...fxg6 14. xc4 xc4 15. h4

After only 15 moves, White has achieved an ideal position with two
heavy pieces on the open h-file.
15... f7?
15...h5! was more tenacious, though White still has an attack after
16.g4! hxg4 17. h7+ f7 18. h6 g8 19.f3, opening more lines
against the king.
16. f3 h5 17.e5 g4 18.b3

67
18. xg4 hxg4 19. g5+ g8 20.b3 f7 21.exd6 cxd6 22. h7!,
followed by e4, was a slightly cleaner way.
18... xh4 19. xh4 d5 20.exd6 cxd6 21. xd5 xd5 22. g5
c3 23. xd6 fe8 24. c4 ac8 25. d7+ g8 26. xc8 xc8
27. xb7
White has a winning endgame. Mieses finishes things off
professionally.
27... xa2+ 28. d2 a5 29. d7 b4 30. d8+ xd8+ 31. xd8 c6
32. b6 f6 33.c3 g5 34. d4 xd4 35.cxd4 f7 36. d3 e6
37. e4 b4 38. xa5 d5 39. d8 g4 40.g3 c3+ 41. e3 d5
42. a5 b5 43. b6 d6 44. c5 e4 45.b4 c3
And Black resigned without awaiting 46. f4.
Mieses stopped playing the opening after 1920, but other strong
players still played it, even if only as a surprise weapon. One of
them was the Punjabi legend Sultan Khan (1903-1966), who played
the Center Game at least three times while visiting Europe in the
1930s. In his biography Sultan Khan (2020), author Daniel King
dismisses the Center Game as a ‘poor system’, and speculates that
Khan played it only because he didn’t have a proper opening
repertoire with White yet. Still, he managed to score 3/3 with it: it
couldn’t have been that poor.
The Polish-French grandmaster Savielly Tartakower (1887-1956),
author of the influential work Die Hypermoderne Schachpartie,
experimented with the Center Game, too. At the 1937 Chess
Olympiad in Stockholm, he played a game that is still relevant for
the overall viability of the Paulsen Attack today.
Savielly Tartakower
Samuel Reshevsky
Stockholm Olympiad 1937
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 f6 5. c3

68
As an aside, it is interesting that Reshevsky played the Center Game
in one of his first ever recorded games. The Polish prodigy, who
emigrated to America in the 1920s, was just five years old at the
time. The game went 5. f3 e7 6. d3 0-0 7.0-0 d6 8.h3 e6 9.c3
d7 10. d4 e5 11.f4 xd3 12. xd3 and Black won in
Reshevsky-Factor, Lodz 1917.
5... b4 6. d2 0-0 7.0-0-0 e8 8. c4 d6
First played by David Janowski against Szymon Winawer in their
second match game, Vienna 1896. Black aims to neutralize the
white bishop with ... e6. This simple move seems to have buried
the c4 line for several decades.
9. f3
The great Paul Keres had faced 9.f3?! in a correspondence game
five years earlier. The problem is that it’s too slow here, as the
bishop on c4 is prone to attacks: 9... a5 ( 9... e5! (Spielmann-
Eliskases, Semmering match 1937) followed by ...c6 and ...d5, leads
to a good position for Black) 10. d3?! d5 and Black was better in
Von Feilitzsch-Keres, correspondence game 1932.
9.h3 e5 10. b3 e6 11. xe6 xe6 is not great for White either
(Ljubojevic-Rosino, Venice 1966).
9... e6 10. xe6 xe6

69
Black is more active and White needs to be fast if he wants to prove
something.
11. g5 e8 12.f4?!
This active move looks logical, but it is not White’s best try.
White’s best chance at this point is 12.a3! a5 13. g3, intending
13...h6 14.h4! with a very sharp position, which will be analysed in
Chapter 3.
12...h6 13.h4
The only chance.
13... c8!
A new move at the time, trying to improve on Janowski.
13... d7 14. f3 g4 ( 14... d4! ) 15. h3 d4 was also fine for
Black in Winawer-Janowski, Vienna 1896.
13...hxg5? 14.hxg5 is clearly losing for Black.
14. f3
The position is very complicated indeed, and it’s not strange that
Reshevsky now makes a mistake:
70
14... f8?
Too subtle, but White doesn’t profit. Reshevsky wants to have the
possibility of ... g8 at hand, but it’s much too slow.
14... d4! 15. d3 c5! was better, leading to an advantage for Black,
though the position remains very complex.
15. d5?
Trying to force matters, but not to White’s advantage. As often,
exchanging the dark-squared bishops is a bad plan for White.
Correct was 15. b1 when the engine evaluates the position as
slightly better for White, though this is not easy to establish with
human eyes. If Black tries the same as he should have done on the
previous move, 15... d4, White has 16. d3 c5 17.e5! dxe5 18.fxe5
xe5 19. h7+ and the annoying check gives White all the attacking
chances he can dream of.
15... xd5 16.exd5 d4

A miserable position for White. Once the d2-bishop is gone, it’s

often downhill from there, and so it is here. The opening has turned
into a nightmare for White.
71
17. d3 e2+
An annoying (and winning) little check. If only the white king
would have been on b1...
18. b1 xd2 19. xd2
It was better to try and complicate matters by means of 19. e6+!?
fxe6 20. xd2.
19... xf4
Black is a pawn up and White doesn’t have enough compensation.
Tartakower tries to complicate matters, but it doesn’t work:
20. h7+ g8 21. f6+ gxf6 22. g3+ g6 23.h5 f5 24.hxg6
fxg6 25. f2 g5 26. c3 e5
Liquidating into a winning ending for Black.
27. xf6 xc3 28.bxc3 g7

Black is clearly winning here. However, the game isn’t over yet.
29. f2 f8 30. e2 ae8 31. he1 xe2 32. xe2 f1+ 33. b2 f7
34.c4 h5 35. c3 g5 36.c5 dxc5 37. c4 g4 38. xc5 g3 39. d4 h4

72
40.c4 b6 41.a4 f2 42. e4 xg2 43. xh4 a2 44. g4 g2 45.a5
bxa5 46. c3 a4 47.c5 f6?!
Good enough, but making things more difficult than necessary.
Cleanest was 47... e2! 48. d3 (or 48. b4 e4+ ) 48...a3!.
48. g8 e5 49. g5+
More dangerous was 49.d6 cxd6 50.c6, after which Black needs to
find the precise 50... a1! to win.
49... f6
49... f4!.
50. g8 f7?
50...a5! was winning: 51.d6 cxd6 52.c6 a1 53. xg2 c1+ 54. d4
xc6 55. a2 a6 56. xa4 e6.

51. g4?
Missing a golden opportunity.
51.d6!! would have saved the game: 51... xg8 ( 51...cxd6 52.cxd6
xg8 53.d7 g1 54.d8 + and, amazingly, White has a perpetual or

73
will win back the rook on a2) 52.dxc7 g1 53.c8 + and here, too,
White has a very unusual perpetual check.
51...a5 52.c6 e7 53. g6 f7 54. g4 f2 55.d6 a3 56.dxc7 a2
57.c8 a1 +
Black promotes with check, and this proves decisive.
58. c4 c2+ 59. b5 f1+ 60. a4 a2+ 61. b3 b1+ 62. c3
c2+ 63. d4 g1 + 0-1
This game was another blow to the reputation of the Center Game,
and theoreticians spoke lowly of the opening in those years. Max
Euwe, the fifth World Champion, wrote in Part 12 of his Dutch
opening series Theorie der Schaakopeningen (1939): ‘the Center
Game was never a common opening and is only seen very rarely
these days. The white queen comes out too early, and because of
this, Black can easily equalize.’
The opening was rarely played in the next three decades and almost
never by strong players. One anomaly was a game between
Alexander Tolush and Mikhail Botvinnik, the sixth World
Champion, played in Moscow in 1944. Tolush managed to outplay
the future World Champion in an attractive game. Another World
Champion-to-be, Boris Spassky, faced the Center Game in 1953,
when he was still a teenager, against Octavio Troianescu in a
tournament in Bucharest (see the notes to Game 16). He won.
In 1970, David Bronstein published 200 Open Games (the English
translation appeared four years later) in which he discussed 1.e4 e5
openings. If Bronstein wrote about something, it was usually worth
paying attention. His view on the Center Game was quite uplifting,
and, as always, highly insightful. Bronstein was ahead of his time in
understanding many hidden concepts in the Center Game, such as
the unusual set-up with b2-b3, which we’ll see later in the book. He
wrote: ‘The QP Attack, 2.d4, is a most dangerous opening. (...) The
2.d4 attack is in complete accordance with all the requirements of
the positional school of chess: lines are opened up for a great
74
number of White’s pieces and his whole position becomes straight
away more refractory. It is also worth remembering that the queen’s
pawn advance forms an organic part of almost any opening variation
beginning with 1.e4 e5, since without d4 it is very difficult to create
any initiative. Thus, it matters little on what move White plays his
d-pawn to d4; the move makes its way imperceptibly into every
opening of an open type, and we are fully justified in considering
the QP Attack a most dangerous fighting weapon, dangerous that is,
of course, for Black, not White.’
An unexpected boost for the Center Game came in 1972. The young
Argentinian talent Juan Carlos Hase (born in 1948) had already
experimented with the Center Game in the late 1960s and scored a
notable result with it at the 1972 Chess Olympiad in Skopje.
Juan Carlos Hase
Anatoly Karpov
Skopje Olympiad 1972
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 d6 5. c3 f6 6. d2 e7
7.0-0-0 0-0

75
8. g3!
The usual move, but a novelty at the time.
A) 8. e2 e8 9. g3 was played in Chigorin-Przepiorka,
Nuremberg 1906, and now 9... d4! would have been fine for Black;
B) Not better was 8.f3 d5 9. e1 d4 10. f2 c5! as played in the
simul game Steinitz-Tresling, Haarlem 1896;
C) 8.f4 was once played by Gunsberg and by Miguel Najdorf (in
a simul game), but also recently (via a different move-order) by
Magnus Carlsen. The problem is that it again allows 8...d5! with the
typical idea 9.exd5 b4!=.
8...a6?!
A common but, here, unfortunately timed plan.
Better was 8... e8 9.f4 and now Black can still go for 9...d5! even
with loss of a tempo, though White is slightly better after 10.e5.
9.f4 b5?!
Here, too, it was better to accept the loss of a tempo in favour of
opening up the position: 9...d5 10.exd5 b4 .
10.e5! d7
Or 10... g4 11. f3 h6 12. d3 when White is also doing great,
for instance: 12... f5 13. xf5 xf5 14. f2 b4 15. e4.
10... e8 11. f3 also favours White.
11. f3 b8
Best was 11...f5 when White should play 12.h4 .
Matanovic mentions 11... b6 in Chess Informant 14, but White has
an overwhelming advantage after 12. d3.
12. d5
12. e3 and White is already at +1.5 according to the engine.
12... c5 13. e3 e4 14. e1

76
14...f5!
Karpov grabs his chance. Though White is still better after this, the
knight on e4 now has a stronghold. Hase now starts to play less
confidently.
15.h3?!
15.g4 was possible right away.
15... e6 16. g1?!
Again, a bit slow.
16... h8
16... xd5 17. xd5 e8, intending ... f7, would have been rather
unclear.
17.g4 dxe5 18. xe7
18. xe5 was better.
18... xe7 19. xe5 xe5 20.fxe5 bd8 21. d3

77
21... d5?
21...f4! would have equalized: 22. xe4 xd1+ 23. xd1 fxe3
24. xe3 and now the ‘Fischer-like’ grab 24... xa2 would have been
possible since, unlike in the first World Championship match game
Spassky-Fischer from the same year, the bishop can’t be trapped.
In this position, a draw was agreed, but in fact White was winning.
The game could have continued 22.gxf5 xe5 and now even
stronger than Matanovic’s 23. h4 is 23. d2! with the double threat
of g4 and c3, hitting g7.
In spite of this, the Center Game was still rare in the 1970s and 80s,
and this was reflected in opening books from those years. A minor
exception was a booklet from Leonard Pickett called Centre Game
and Danish Gambit (1976). But in an influential German chess
opening series from the 1980s, Alexander Suetin wrote, familiarly:
‘The direct push in the center 2.d4 leads to rapid engagement of
tactical battles. White, however, brings the queen into play much too
early (in case of 3. xd4 )...’

78
But after an ice age of almost 80 years, a change was about to come
at last.

Revolution in the 90s: the Shabalov Variation

In the early 1990s, commercial chess programs and databases
established their position in the market. This meant that it was now
much easier to find relevant games and analyse complex openings at
home. The realization dawned that opening theory was not at all
exhausted yet, and that there were many fascinating new ideas
waiting in the depths of the silicon processing units.
From 1991 onwards, the Center Game was not only increasingly
analysed and played by amateurs who wanted to surprise their
fellow members at the local chess club or their opponents in a
tournament, but also by strong players. The Peruvian grandmaster
Julio Granda Zuniga tried the opening a few times in the 1990s (see
Chapter 3 for an important game he played), and his young
compatriot Manuel Muñoz Pantoja (born 1975) followed suit.
Muñoz, now a grandmaster living in Spain, and well-known on the
Catalan chess circuit, became one of the biggest and most loyal
supporters of the Center Game, essaying a huge range of different
ideas for White over three decades. There were even stronger
players looking into it more seriously now. For example, the British
number one, Michael Adams, played the Center Game against
Viswanathan Anand in their 1994 match in Linares.
However, the player who stood at the forefront of the Center Game
revolution in the 1990s was, without a doubt, Latvian-American
grandmaster Alexander Shabalov. We already met him in the
Introduction. Born in Riga (then part of the Soviet Union) in 1967,
Shabalov emigrated to the United States in 1993 and became US
Champion in the same year. Because of his attractive and
enterprising style, Shabalov was a very popular player at the time
(and still is).

79
The following game, played during the 1994 US Championship in
Key West, Florida, didn’t go unnoticed. In it, Shabalov introduced a
novel idea which, it’s fair to say, revitalized the entire opening.
Alexander Shabalov (2600)
Alexander Ivanov (2575)
Key West US Championship 1994
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 f6 5. c3 b4 6. d2 0-0
7.0-0-0 e8 8. g3
Siegbert Tarrasch’s old move, but Shabalov has something new in
mind.
8... xe4
This was by now known to be the ‘refutation’ of 8. g3.
9.a3!?
Shabalov’s idea, forcing Black’s hand. Where does the bishop go?
9. d3 g4! 10. h3 e5, as in Zinn-Sax, Baja 1971, is not what
White wants.
9. g5!? is actually more interesting than was thought at the time.

80
9... d6
One of many options at this point.
9... g4!? is a tricky move which Shabalov faced in the same year
against chess computer Socrates.
The calm retreat 9... a5 was discovered only the next year, and
soon opened a can of worms as it proved to be a very annoying
move for White. I have a personal connection to this line that I
would like to mention here. When these variations were being
discovered, I played a lot of online games on the Internet Chess
Club. I also got to play it against Alexei Shirov, with whom I was
chatting sometimes. One of these games went as follows: 10.h3!?
(preventing ... g4 and preparing the slow g3-h2 and g2-g4)
10...d6 11. d3 e8 12.b4?! b6 13. g5 e5 14. d5 xd3+
15.cxd3? d4 16. f3 e6 17. xd4 xd5 18. f5?! e6? (
18... h5! ).

81
 analysis diagram
Now I could have lashed out with 19. xg7! (I played 19. h6+ and
was lost after 19... f8!, Moll-Shirov, ICC blitz 1998) 19... xg7
20. h4! ( 20. h4+ h8 21. f3 g7 22. g3+ is a draw, which
wouldn’t have been bad either against one of the best players in the
world) 20...h5 21.g4! with a very strong attack for White.
10.f4 e8 11. f3 c5
The alternative is the more solid 11... f8!. Black is actually better
here, but this wasn’t the consensus when Shabalov introduced his
idea.
12. d3 d5 13. de1 xe1+ 14. xe1

82
White has sacrificed a pawn but has very nice development and
active play.
14... e7 15. h4?!
The direct 15. e5 was better, leading to the game position after
move 17.
15... g6 16. f3 e7?!
There was nothing wrong with solidifying the position by means of
16...c6 and Black is better.
17. e5! f5?!
A natural exchange, but now White grabs the initiative in style.
17... d4! is a strong computer move, leading to a slight plus for
Black.
18. xf5 xf5 19. d3 e7 20.g4!

83
Shabalov pushes the g-pawn and Black already needs to tread very
carefully.
20...c6 21.g5 d7?!
Instead, 21... e8, keeping an eye on g7, was more tenacious,
though White can still play for an attack in that case with 22.h4
followed by h4-h5.
22. g4! b6?
It’s remarkable how quickly Black’s position collapses in situations
like these. Only very precise play could have saved him.
22... h8 23.f5 g8 looks extremely passive but keeps things
together for the moment.
23.f5! d7
23... c4 24.f6 g6 25.fxg7 xd2 26. xd2 is also close to winning
for White.
24. f1 d6 25. h3!

84
Shabalov patiently puts his pieces on the right squares. The f5-pawn
is taboo on account of h6+.
25... h8 26.f6
Finally pushing through on the dark squares, a recurring theme.
Shabalov finishes in style.
26...gxf6 27.gxf6 g6 28. h6 f8

85
Desperately defending against the pressure on g7, but...
29. g7+ xg7 30.fxg7+ xg7 31. h6+ h8 32. f6 1-0
A model attacking game by White, highlighting once more the
danger that awaits Black on the dark squares.
With the inauguration of the ‘Shabalov Variation’, a short period of
enthusiasm for the Center Game began. Suddenly, the old lines with
8. g3, first played by Tarrasch, seemed like a promising way again
to get a dangerous attacking position with White against 1.e4 e5.
The young mega-talent from Russia, Alexander Morozevich, played
it at the Lloyds Bank Masters in 1994, crushing the experienced
1...e5 player Mark Hebden with it.
The following year, Shabalov produced another Center Game
classic.
Alexander Shabalov (2570)
Igor Shliperman (2320)
Newark 1995
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 f6 5. c3 e7 6. c4
An important move to fight against ...d7-d5.
86
6...0-0 7. d2
7. ge2 occurred in a game from 1953 involving the young Boris
Spassky: 7... g4!? 8. d2? ( 8. f4 is unclear) 8... c5 9. d1 e7
10.f3?! h4+ 11.g3? ge5 12.gxh4 xf3+ 13. f1 d5! with a clear
advantage in Troianescu-Spassky, Bucharest 1953.
7...d6
A small victory for White: Black has played ...d6 with his bishop
still on e7.
As often in this opening for Black, a more active approach was
called for: 7... g4!, which will be examined in Chapter 4.
8.0-0-0 e5
8... e6 9. xe6 fxe6 was already seen in the old game Soloviev-
Levenfish, Gorky 1950, and now 10.f4! would have been somewhat
better for White.
9. b3 e6

10.f4
A typical push, not fearing 10... c4.
87
10... c4 11. xc4
White prefers to keep his dark-squared bishop.
11... xc4 12. f3 c6
Shabalov had faced 12... e8 the year before against a computer
engine. He should have continued 13. he1! with an edge ( 13.h3?!
b5!?, Shabalov-Comp WChess, 1994).
13. d4 e8 14. g3?!
e3-g3 is a key manoeuvre in this opening, but here the queen
would have been better placed on f3, not hindering the g-pawn’s
advance.
For example, 14. f3 f8 15. he1 followed by g2-g4.
14.b3 would also have been strong.
14... f8 15. he1 c7?
Too meek.
15...b5! would have led to a sharp but balanced position. In general,
Black should strive for counterplay as fast as possible in the Center
Game.
16.b3 a6 17. f5 d7 18. g5!

88
The white queen never tires of running along the dark squares.
18... e6 19. e3 h8
Stepping out of any pins on the g-file, but there is still the h-file to
think about.
20. h3! ae8 21. e3 g8
If 21... xe4 22. xe4 xe4, 23. xd6! wins the house.
22. d4 g6 23. h4
Completing the journey. White is winning.
23...h6 24.g4
Finally, the g-pawn advances with deadly impact.
24... e7
24...d5 25.g5 h7 26. xh6 gxh6 27.f5 d6 28.e5! is a beautiful
winning line.
25.g5 e6

89
26. xg7!
The g7-point marks the spot.
26... xg7 27.f5 d7 28.gxh6
A nice queen sac to finish it off.
28... xh4 29.hxg7+ h7 30. xh4+ h6 31. g1
Black is powerless against White’s pieces on the dark squares.
31... g8 32. f6 d5 33. xh6+
Black resigned as mate with g1-g4-h4 can’t be prevented.
Not everyone was convinced yet. In a Survey for New in Chess
Yearbook 34, Dutch grandmaster Paul van der Sterren wrote:
‘Honestly speaking, I do not believe that an opening like the Center
Game can be good. (...) But as is so often the case, the weakness of
a variation may also be its strength. The resulting positions may be
good for Black but they are not easy and a well-prepared (or a very
strong) opponent may be able to set Black many practical problems.
In the game Morozevich-Hebden, Black was probably caught by

90
surprise, avoided the sharpest line and found himself outplayed in a
complicated middlegame.’
Avoiding the sharpest lines was precisely what some strong players
did with success. In the 1995 Dos Hermanas tournament, Shabalov’s
former compatriot Alexei Shirov tried the Center Game against the
best 1...e5 player in the world: Anatoly Karpov.
As we saw above, Karpov had barely saved the draw against Hase in
1972, but he showed himself older and wiser this time.
Alexei Shirov (2710)
Anatoly Karpov (2780)
Dos Hermanas 1995
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 f6 5. c3 b4 6. d2 0-0
7.0-0-0 e8 8. g3 d6
Karpov doesn’t aim for complications and instead chooses a quiet
but healthy set-up.
9.f3 e5
It was still possible to play 9...d5!?, even with the loss of tempo.

91
10.h4 h8
A novelty at the time, and a strong one.
Stepping away from the g-file is a common idea for Black.
10...c6?! 11.h5 was played in Morozevich-Hebden, London 1994.
Karpov also analysed 10...h6, when 11. xh6? ( 11.h5!; 11. h2!?,
intending g2-g4) isn’t advisable due to 11... h5 and Black is
slightly better.

11. h3
11.a3 a5 12. e1! is an instructive way of playing the position.
White intends, of course, g2-g4.
11... h5?!
11...c6 with the idea ...b7-b5 was better.
12. h2 c6 13.a3 a5

92
14. e2
This move, though it may look perfectly natural to the untrained
eye, is often too slow in the Center Game. In his analysis, Karpov
gives it a question mark, but White is still slightly better even after
this.
The best move for White was 14. f2! (or 14. g1, as indicated by
Karpov). White intends to play g2-g4 and has a good game.
14... xh3!

93
A very pleasing move, disrupting the coordination of the white
pieces.
15. xh3?
Only this was the real mistake. Now Black does take over.
After 15.gxh3! xh4 16. hg1 h6, Karpov gives Black a clear
advantage. In fact, the engines say White is clearly better after
17.f4! with a strong initiative. If 17... xc3 18.bxc3 d7 19. df1
xe4 White has 20. g2!.
15... xc3!
Also giving up the second bishop.
16.bxc3
16. xc3? loses a piece to 16... f4: the bishop is unfortunately
placed on e2.
16... f6?!
16...f5! was stronger, although Black is objectively only slightly
better.
17.c4?!
94
Karpov only analyses the immediate 17.g4, but after 17.h5!,
followed by g2-g4, White is still in the game.
17... b6
17...b5! gives Black a dangerous attack.
18.f4?
Shirov should have gone for the other pawn push: 18.g4!. Now,
Black could play 18... d4 and Karpov thought this was good for
Black, but the engines still see sufficient compensation for White
after 19. b1 xc4 20. xc4 xc4 21.g5, when Black should
probably give perpetual check.
18... ed7

Black’s knights are superior to White’s bishops. More importantly,

White’s king isn’t safe.
19. d3
Better was 19. c3 xe4 20. d4.
19... c5 20.e5 a4 21. b4 dxe5 22.c5 c7 23. c4?! a5 0-1

95
From a public relations perspective, this game felt like a disaster for
the Center Game. Shirov never played it again and was still
suspicious of it when I spoke to him about it a few years later.
Indeed, from Karpov’s analysis in Chess Informant 63, one got the
impression that life was good for Black even in the non-critical
lines. And more bad news was underway. Theoreticians gradually
started to understand how Black should respond to Shabalov’s 9.a3.
By the end of the decade, the Tarrasch Gambit had been abandoned
once again by almost everyone except a few true diehards.
So where does this leave us, as we approach the year 2000 in our
historical odyssey? Was White’s early queen excursion to be
considered a beginner’s mistake after all? Some players refused to
believe it. John Emms still judged the opening mildly in his book
Play the Open Games as Black (1999), writing that ‘(...) it would be
dangerous to dismiss the Centre Game as merely trash.’
In the next chapter, we will see how theory developed in the decades
that followed the short period of hope for the Center Game. We will
see how Judit Polgar paved the way for a new generation of
grandmasters who weren’t shy of experimenting in the opening, and
how neural network chess engines changed everything.

96
Chapter 2
The Center Game in
the New Millennium
I hear the roar of a big machine
Two worlds and in between
The Sisters of Mercy, Lucretia My Reflection

Polgar and Nepomniachtchi lead the way

In May 1997, the world was shocked by the defeat of World Chess
Champion Garry Kasparov in a six-game match by Deep Blue, the
supercomputer designed by IBM. From this moment onwards, it was
undeniably clear to everyone that, going forward, chess programs
(or ‘engines’) would be stronger than humans. Soon, they would
also be the main driving forces behind all future developments of
chess opening theory.
We ended the previous chapter with the rapid rise – and fall – of the
Center Game in the last decade of the 20th century. At the dawn of
the new millennium, fresh ideas were needed for White, as all the
main lines of the Paulsen Variation (4. e3) appeared to give Black
equality if not more. This continued to be the consensus among
most theorists until the end of the first decade. Mihail Marin called
the move 3. xd4 a ‘rather colourless continuation’ in his 2007 book
Beating the Open Games, playing the broken record tune once more:
‘According to the classics, the queen is supposed to be one of the

97
last pieces to be developed, and with good reason. After this
premature development it will be rather vulnerable in the center.’
But hope springs eternal. Some players kept on looking for (and
sometimes finding) improvements and new plans for White in the
old Tarrasch Gambit. Modern chess programs, however, simply
didn’t see enough compensation for White in many of the old
gambit variations, and analysts needed to go very deep down the
lines to find possible ideas behind the horizon of the engines.
Players began to look into alternatives for White, far beyond the
standard theoretical recommendations and analysis. One of these
key innovators was Miguel Angel Muñoz Pantoja, whom we already
briefly met in Chapter 1. A strong and practical player, Muñoz
Pantoja has been a big supporter of our opening throughout his
career, but he never just blindly copied the old lines.
Let’s take a look at one of his games. It allows us to broaden our
scope and to experience some of the richness of the Center Game.
Miguel Angel Muñoz Pantoja (2375)
Candido Martin Torras (2080)
Barcelona 2002
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 d6 5. c3 f6
A quiet but solid set-up for Black. It would have been perfectly
reasonable to go for the usual set-up with d2 and 0-0-0 now, but
Muñoz Pantoja chooses a different path:
6.b3!?

98
This had been tried before at this point, notably by the Icelandic
grandmaster Thorstur Thorhallsson, but Muñoz Pantoja has played it
so often that it would be appropriate to name the variation after him.
The idea of playing b3 and b2 in the Center Game was originally
introduced by David Bronstein. Nowadays, it is also a common
motif in many Sicilian lines where White captures with the queen on
d4 instead of with the knight.
6... e7 7. b2 b6?
Copying White, but this is a poor move. On the other hand, as
White, you better get used to seeing poor moves being played very
early on in this opening!
The more natural 7...0-0 8.0-0-0 e8 was played in Thorhallsson-
Gunnarsson, Gardabaer 1996, when White started a typical kingside
expansion with 9.h3 e6 10.f4 f8 11.g4 with a good game.
Best is the counter-intuitive push 7...d5!=, which to date has never
been played according to the database.
8.0-0-0 b7 9.f4
White already has an overwhelming advantage.
99
9...0-0
As often against a set-up with f2-f4, it is better to castle queenside
with Black: 9... d7 10. f3 0-0-0.
10. f3 c8 11.h3
The immediate 11.g4! was already possible, e.g. 11... xg4 12. e1
f6 13. g1 soon followed by g3.
11... e8 12.a3 f8 13. d3 e7

14.g4?!
A little sloppy, allowing Black’s next move, though White remains
on top.
14. he1 ed5 15. xd5 xd5 16. d2 was winning.
14... ed5! 15. xd5 xd5 16. d2 f6 17. xf6 gxf6 18. he1
d5?! 19.exd5?!
19.e5! was winning.
19... xa3+?

100
After the correct 19... d7!, White’s advantage would not have been
all that big.
20. b1 xe1

21. xe1!
Of course. The queen is headed for the kingside along the dark
squares once again.
21... f8 22.g5 g7 23. a2 d6 24. h4?!
There were many easily winning moves, but this is not one of them.
24...fxg5?
24...f5! was better, although White retains the attack after 25. xf5
a5 26. e5 a4 27.b4!.
25. xg5 h6 26. f3 f8 27. g1 h8 28. g5 e7 29. h5 xg5
30.fxg5 xd5 31.g6 d4 32.g7+
Yet again, the final kill happens on the g7-square. Black resigned.
A major new concept for White was introduced by Judit Polgar,
already mentioned above. Polgar is still the best female chess player

101
ever and one of the most famous and beloved chess personalities of
all time, known for her tactical vision and (nowadays) superb online
live commentary. She had already tried the Center Game a few
years earlier, but during the 2002 FIDE Grand Prix rapid tournament
in Moscow (where she famously beat Garry Kasparov), she
unleashed a surprise upon her strong opponent on move 8.
Judit Polgar (2677)
Rustam Kasimdzhanov (2674)
Moscow rapid 2002
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 f6 5. c3 b4 6. d2 0-0
7.0-0-0 e8 8. f4!?

Polgar wants to move her queen closer to the kingside, but unlike in
the Tarrasch Attack with the queen on g3, the pawn on e4 is still
protected. Moreover, the queen doesn’t block the white g-pawn and
helps prepare the typical f2-f3, g2-g4 storm. This move has been
dubbed ‘Nepo’s Queendance’ by Gorny – and it’s true Nepo played
it quite a few times – but to my knowledge it was tried first at the
top level by Judit Polgar.

102
8...d6
Kasimdzhanov plays quietly, probably fearing some special
preparation.
8... xc3 is played much more often, and will be examined later in
this chapter.
8... d6!? is a dangerous little move if you don’t know how to
respond. The correct way is 9. f3! ( 9. h4? xe4! is a trick to
remember) 9... e5 10. e2, as we will see in the next chapter.
9.f3 e6
Human development, but this move is often not so great for Black.
9...a5 is the engine’s choice, but has never been played in practice.
After 9... e5, 10.h4!? is a new move which promises White a
decent position.
10.g4
10.h4!? is the other agressive option.
10. ge2 is a quieter alternative.

10... d7
103
Trying to anticipate g4-g5 and re-routing the knight to the centre.
The surprising 10...d5! has never been played as far as I can see, but
it seems to solve most of Black’s problems: 11.g5 ( 11.exd5 d6 is
very dangerous for White) 11... h5 12. h4 d4! with a crazy but
balanced position.
11. h3?!
Intending g5, but White doesn’t benefit much from exchanging
this knight for Black’s light-squared bishop. After having seen (and
played) hundreds of games with this type of position, I believe that,
in general, the move h3 is somewhat overrated for White in the
Center Game.
The prudent 11. b1 would have given White a slight edge.
11... de5 12. g5 d4!
Black has comfortably equalized.
13. g3 c4

14. e3?
Very unfortunate.
104
On a good day, Judit would have surely found 14. h3! h6 15.f4!
when White is only slightly worse.
14... xc3 15.bxc3 e2+ 16. xe2 xe2 17. de1 a6
White’s queenside is a ruin and her attack on the kingside is much
too slow.
18.h4 c5 19.h5 c4
19... a5!.
20. f4 f6 21. xh7
White has to gamble, as 21. h3 a5 would have easily won for
Black.
21... xh7 22. f5+ h8 23.h6 g5!
Ending all dreams White may have had about continuing the attack.
The future FIDE World Champion finishes the game effortlessly.
24.f4 xe3 25. xe3 gxf4 26. f3 e5 27. xf4 e2 28. f2 c4
29. d2 e7 30. b2 e8 31.h7 b5 32. h6 b4?!
32... f8.
33. xf6?!
33. xf6+ xf6 34. xf6 would have required Black to show some
endgame technique.
33... xe4 34. g5 bxc3+ 35. xc3 e3+ 36. d3 xd3 37.cxd3
e5+ 38. xe5 3xe5 39. xd6 xh7 0-1
Polgar was ahead of her time and her idea would prove to be a
worthy alternative to the Tarrasch Gambit. But because of the final
result of this game, a familiar story now repeated itself: the move
8. f4 didn’t get much attention and failed to attract followers.
Andrew Greet didn’t mention the move at all in Dangerous
Weapons: 1.e4 e5 (2008). Victor Bologan, writing in 2014
(Bologan’s Black Weapons in the Open Games), devotes just a few

105
paragraphs to the move. Admittedly, I myself also didn’t think
highly of it at first.
That changed when Russian super-grandmaster Ian
Nepomniachtchi all of a sudden tried it not once but twice with
White at the Leuven GCT blitz tournament in 2017. The blitz games
Nepo played in Leuven were not just against random players. One
of them was against the world number five, the Armenian
grandmaster Levon Aronian. The other happened to be against the
reigning World Champion, Magnus Carlsen.
Ian Nepomniachtchi (2732)
Magnus Carlsen (2832)
Leuven blitz 2017
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 b4+ 5. c3 f6 6. d2 0-0
7.0-0-0 e8 8. f4 xc3!
The critical test. The e4-pawn is now hanging.
8... e5 was Aronian’s choice against Nepomniachtchi a few rounds
earlier. It is Black’s second-best move in the position.
9. xc3 xe4!
This move is not seen often at club level or in blitz games, possibly
because, superficially, it looks like White wins a rook after taking
on f6 and on e4.
9... xe4 10. c4 would have led to the game Winawer-Steinitz,
Nuremberg 1896 (see Chapter 1).

106
10. g3
10. xf6? xf4! 11. xd8 xd8 leaves Black a healthy pawn up.
10. d2!? was only tried once, but it is considered White’s best
move by the engines. We will see it in the next chapter.
10...d5 11.f3 e8
Better was 11... e6! 12. e2 d6!. This key idea for Black was
played against Nepo by Vishy Anand in a blitz game in St Louis a
few months later. It is still considered to be Black’s strongest
response to this line and Black is objectively better here.
12. e2 a5!? 13. f4?
A mistake, after which Black is better. 13.a3! b4! 14. b1! is what
the machines recommend. But from a human perspective, it’s less
easy to understand what precisely is going on here.
13... b4!

107
14. xf6?
Such an admission of defeat, giving up the most important piece on
the board, is a sign that things have gone very wrong for White.
14. c4! was the only move, when Black is still better but nothing is
decided yet in the ensuing complications: 14...c6 15. he1 f5
16. xe8+ xe8 17. b3 a4 18. e1! (threatening 19. xe8+) 18... f6
19. xb4 axb3 20.axb3. It may look playable for White, but Black
keeps the initiative after 20... b6.
14... xf6 15.a3
15. xd5 xd5 16. xd5 e6 is rather terrible for White, too.
15... f5 16.axb4?
Severe time-trouble already (remember, this was a blitz game).
16...axb4
Now Carlsen has an easily winning position.
17. xd5 a1+ 18. d2 xb2 19. xc7 xd1+
It’s mate in two, so Nepomniachtchi resigned. A painful loss!

108
It was a little unfair of me to start the section on his contributions
with a loss (even though it was against Carlsen), because the first
successes Nepomniachtchi had with the Center Game could already
be seen during the first decade of the new millennium. This was at a
time when the system with ...g7-g6 was on the rise as a healthy and
ambitious way for Black to fight against the Center Game. Nigel
Davies recommended it in his book Play 1.e4 e5! (2005), and so did
Mihail Marin in Beating the Open Games (2007).
The following game contains all the drama one can hope for as a
spectator.
Ian Nepomniachtchi (2545)
Evgeny Tomashevsky (2595)
Moscow 2006
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 g6
Stamma’s move. We already saw it when we analysed the game
Mieses-Spielmann, Breslau 1912.
5. c3 g7 6. d2 d6
Not allowing e4-e5 – just in case.
7.0-0-0 f6

109
White has more than ten viable moves in this still-popular position,
which first occurred in a game Riemann-Hruby, Nuremberg 1883.
8.h4!
Brave, agressive and... best!
8.f3 is the most popular move in this position, and it isn’t bad, but it
allows Black just a bit more time to organize.
8. c4 0-0 9.f4 was played by Shabalov once in a game against Artur
Yusupov at Chicago 1996.
8... e6
It was better for Black to stop the advance of the h-pawn: 8...h5, and
now play might continue 9. h3 or 9. g3 followed by f2-f3 and
h3-f4, with a small plus.
9. h3 e7
9... d7 10. f4 0-0-0 11. b5! was good for White in the game
Nepomniachtchi-Toma, Warsaw blitz 2011. The bishop pin is a
recurring motif in lines where Black castles queenside.
10. f4 h5
110
11. b5
It was slightly better to wait for Black to castle queenside before
playing the bishop: 11. b1 0-0-0 and only now 12. b5!.
11...0-0?!
11...a6!, questioning the bishop’s intentions, would have been more
logical. Still, White remains better after 12. xe6 xe6 13. a4.
12.f3
Preparing the usual g2-g4 push, even at the cost of a pawn.
12... e5

111
13.g4! hxg4
13...c6 14. e2 hxg4 15.fxg4 xg4 16. dg1 xe2 17. xe2 with a
raging attack for White.
14.fxg4
14.h5?! would have been a little too enthusiastic: 14...gxf3 15.hxg6?
fxg6 and White doesn’t have a clear follow-up after 16. g1 f7!
17. h2 c6 18. a4 c4! followed by ...b7-b5 and nothing great is
available.
14... exg4 15. g1
Nepo finds yet another nice dark square for the queen. Now h4-h5 is
a threat.
15... h5
15...c6 16. e2 b5 17.h5 and White’s attack comes first.
16. xe6?
An inaccuracy. Correct was 16. f1! when White has a dangerous
initative for the pawn.

112
16... xe6 17. e2 gf6 18. f1
The cautious 18. b1 would have been better, not only tucking the
king away safely but also protecting a2.
18... e5!
A powerful move, exploiting the battery on the a1-h8 diagonal.
Black is better now. Over the next moves, both players don’t play
optimally, but the battle remains very tense.
18... xe4 19. xe4 xe4 wins another pawn, but White has some
initiative after 20. f3.

19. h3 ae8
Simple and sound.
19... xe4 leads to madness after 20. xe4 xe4 21. xh5 e5
22. g5.
20. xh5 xh5?! 21. d3?! c6?! 22. xa7 b5 23. b7 c5 24. df3
c4 25.b3 d4 26.a3 c5 27. d1?!
Better was 27. b1 when Black is still better, but not yet winning.
27...f5!?
113
27... a8! was possible, but had to be calculated accurately: 28. xf7
xc3! 29. xf8+ xf8 30. xf8+ xf8 31. c8+ f7 32. c7+ g8
33. xc3 xc3 34. xd6 f3+ 35. c1 xe4 36.a4 bxa4 37.bxa4
f7! and Black ultimately walks out of the checks.
28.exf5 f7 29. a6 xf5 30. xf5 gxf5 31. a5 g3 32. e1 xe1+
33. xe1 d4+ 34. d2 g1+ 35. e1 d4+ 36. d2

36... f1??
A sad blunder. Tomashevsky probably forgot that the white queen
can now go to g5 not just to protect d2 but also to give lots of
checks.
Preventing the check on d8 would have won: 36... h7! and if
37. d8, then 37... g1+ 38. e1 xc3 wins a piece.
37. d8+ h7 38. g5 xd2 39. xf5+ g8 40. e6+ f8
A winning attempt, but now it’s actually White who can start to
push a little.
40... h7=.
41. e2!
114
Tomashevsky must have felt rather miserable at this point.
Fortunately for him, the position is still a draw.
41... xh4 42. xd6+ f7 43. d7+ f8 44. f5+ e7 45. xd2
h6+ 46. f4 h2+ 47. d3 g3+ 48. e2 h2+ 49. f3 h1+
50. g4 g1+ 51. h5 h2+ 52. h3 b2! 53.a4 bxa4 54.bxa4
e2+ 55. g6 e6+ 56. xe6+ xe6 57. f4+ d6 58. f5 c3
59. e2 a5 60. c1 c5 61. a2 e1 62. c1 a5 63. d3+ c4
64. e5+ c3 65. xc6 b6 66. e4 xc2 67.a5 xa5 68. xa5
½-½
A great battle, evidently making the young Ian yearn for more.
In those early years, Nepomniachtchi still experimented with the
Tarrasch Gambit, but he wasn’t very successful with it. A defeat at
the hands of Francisco Vallejo Pons at the Aeroflot Open in
Moscow (2007) did the rounds and was emailed to me by a strong
chess friend. The email subject line read ‘Why do you play this
opening with White?’ Admittedly, Nepo had played a line that was
already known to be good for Black for decades. In Chess Informant
99, Vallejo Pons teasingly rated the move 2.d4 as ‘dubious’. It was a
sign of the times.
Around the same time, the British International Master Andrew
Greet wrote a big section called ‘The Centre Game Revealed’ in
Dangerous Weapons: 1.e4 e5 (2008). It’s worth quoting some
passages from the first part of his study: ‘The Center Game is a
relatively (...) easy-to-learn, non-theoretical, unusual, provocative
yet at the same time highly aggressive and ambitious opening
system, which many Black players are likely to underestimate. (...)
A final positive point is that White gets to implement his idea as
early as the second move. Thus, if you decide to incorporate the
Center Game into your repertoire, you will have the opportunity to
play it in virtually every single game in which your opponent meets
1.e4 with 1...e5.’

115
Nepomniachtchi, meanwhile, must have taken a closer look at
Polgar’s move and concluded it was viable, if not downright
dangerous for Black. He played it several times, though only in blitz
and rapid games. He has returned to the Center Game at several
points in his later career, until the present day.

AlphaZero and the Knowledge Revolution

In late 2017, the chess world was rocked once again: not by IBM
this time, but by DeepMind, the artificial intelligence (AI) company
co-founded by Demis Hassabis. DeepMind published a
groundbreaking paper about the self-learning AI program AlphaGo
in Nature in 2017. Their more generic program, AlphaZero , which
could also play chess alongside Go and shogi, was released shortly
afterwards. Within 24 hours of playing chess against itself, knowing
just the basic rules, AlphaZero had reached a superhuman level of
play. It then crushed the strongest known chess program, Stockfish,
in a 100-game match with 28 wins, 0 losses and 72 draws.
Matthew Sadler and Natasha Regan documented the neural network
‘knowledge revolution’ in their book Game Changer (2019), in
which they made observations about the way AlphaZero plays
chess. They touched on many themes that are very relevant for the
theory of the Center Game. Two of them are:
1) Pushing the h-pawn as far as possible up the board, opening
lines and weakening the enemy king;
2) AlphaZero’s preference for opposite-side castling.
Both points are obvious in the context of the Center Game. We have
already seen quite a few examples of opening up lines against the
enemy king. Sadler and Regan describe AlphaZero’s ‘tendency to
castle on the opposite side to Stockfish’s king’ and note: ‘AlphaZero
is keen to open lines against the opponent’s king (typically on the
kingside), both to keep the king safe, and to ensure that the king
does not affect the mobility of its rooks.’

116
But opposite-side castling offers chances to both sides, of course.
Modern chess engines (modelled after AlphaZero) generally like
White in the Center Game, but they are critical of White’s set-up in
some of the old main lines, where they have found active plans for
Black. In the Shabalov Variation, for example, they came up with
this:
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 f6 5. c3 b4 6. d2 0-0
7.0-0-0 e8 8. g3 xe4 9.a3 a5!?

In the Center Game, the black a-pawn is AlpaZero’s h-pawn!

Taking the bishop doesn’t look like it should lose by force, but it
does. After 10.axb4? axb4 11. b1 d5! (or ...d6) Black calmly
develops and builds up an irresistible attack.
On the flip side, the engines indicated that there were other viable
ways to play the opening with White, and strong players started to
notice this. One of the first examples was seen in December 2019,
when the Chinese top grandmaster Wei Yi played the old Hall
Variation in his Grand Prix mini-match in Jerusalem against former
World Championship challenger Sergey Karjakin.

117
Wei Yi (2725)
Sergey Karjakin (2754)
Jerusalem 2019
1.e4 e5 2.d4 exd4 3. xd4 c6 4. c4!? f6 5. c3 b4
5... e7 6. e3 will be seen in Carlsen-Firouzja, the last game of this
chapter.

6. d2
6. ge2 is the alternative, which we will see in Chapter 6.
6...0-0
6...d6 7.0-0-0 a6 8.f4 was seen in Gunsberg-Mortimer, London 1887
(see Chapter 1). It’s funny how the theory of the Center Game
effortlessly jumps from the 19th to the 21st century and back.
7.0-0-0 d6
Black can’t play ...d5 here, so he has to opt for a set-up with ...d7-
d6. It was perhaps more accurate to play ... e8 first, but it would
likely transpose to the game.
8.f3 e8 9. e2
118
The queen wants to go to f2, but for the moment, it is uncomfortably
placed on e2.
Best seems 9. ge2 or 9. h3.
9... e6
It’s a bit weird that the logical 9...d5! has only been tried once so
far, in a game between two lower-rated players: 10.a3 f8! and
Black seems slightly better here.
10. f2

White’s strategy has succeeded. He can look forward to an

interesting middlegame.
10...a5
10... e5 was an alternative. White continues 11.g4 and the position
is unclear. The Belgian grandmaster Daniel Dardha had this position
twice in 2023.
On 10... c5!?, it’s best to play 11. e3 xe3+ 12. xe3 d5
13. ge2 .
11. ge2 e7?!
119
Better was the aggressive 11...b5 12. f4 c5 13. g3 ( 13. h4!? )
13...b4 14. xe6 fxe6 15. a4 with an edge.
12. f4?!
12.g4! was called for, after which White has the better chances. This
occurred the following year in Karthikeyan-Brinovec, Graz 2020.
12... xc3 13. xc3 b4

This looks very annoying for White, but there was a miraculous
save.
14.a3?
14. xf6! – a rare case of White giving up his cherished bishop:
14... xf6 15. xe6 xa2+ ( 15... xe6 16.c3 ) 16. b1 xe6 17.c3!
and this somehow holds the balance for White, for instance: 17...b5
18. d2! (intending d5) 18...c6 19.c4!.
14... a2+ 15. d2 d5?!
15... xc3 16. xc3 g4! would have led to a practically winning
position for Black.
16. xf6 xf6 17. d3!
120
Now things are not so clear. The black knight is stuck – for now.
17...dxe4 18.fxe4 g6 19. e2 b5?!
19... xe4 20. f3 a4! would still have been better for Black, albeit
not at all easy to play.
20. g3! f6 21. e5 e7

22. xb5
22. c5 would have held the advantage, but after 22... h4 White
would have needed to find 23. e3!!. This is how the engine rolls.
White will follow up with f3 and his position is very solid; in fact,
he is better. Steinitz would have been proud!
22... ab8?
22... ad8 was equal.
23. c5! h4 24. f3 g6 25.g3 h6+ 26. e2
Now White is winning. The rest of the game is still interesting, but I
give it with minimal comments.

121
26... g7 27. f2 xb2 28. xb2 xb2 29. b1 c3 30. xc3 xc3
31. b7 c5 32. e1 d8 33. e3 d4 34. a1 g7 35. a7 a4
36. b7 d4 37. b6 c4?! 38. xe6!
Followed by d3, so Black resigned.

Covid-19 and the online jungle: practical advantages

of the Center Game
Not long after this game, the chess world was in for another, even
bigger shock. As the Covid-19 pandemic started to spread in earnest
and global lockdowns became an everyday reality, millions of chess
players were limited to playing via their online chess accounts, or
had to (re)open them in order to stay active. Suddenly, chess life
happened almost exclusively online, and this usually meant faster
time controls. Titled Tuesday tournaments became very popular and
players started to experiment with openings that were suitable to
these new conditions. The Center Game was one of those surprise
weapons.
I also re-discovered the joys of playing my favourite opening with
White, but I must admit I still held on to the ‘classical’ main lines
and gambits. More progressive minds, however, started
experimenting with yet another queen move: d4-d3. Daniil Dubov
and David Paravyan, never afraid to try out new theoretical
adventures, started playing it in online games. And so did one of the
most dangerous attacking players in the world: Shakhriyar
Mamedyarov.
The Azeri top grandmaster, born in 1985 and often referred to as
‘Shakh’, is known for his creative and daring opening play. He
employed the move d4-d3 in numerous online blitz and rapid
games during the pandemic.
In the following game, which started from a Vienna Game, the
queens are exchanged early on (which happens often in this line),
but the position remains very tactical. This shows that even in
122
seemingly more ‘quiet’, endgame-like positions, there is still plenty
of sharp play for both sides in the Center Game.
Shakhriyar Mamedyarov (2770)
Jorden van Foreest (2701)
Carlsen Invitational Internet rapid 2021
1.e4 e5 2. c3
A small but important subtlety. White often chooses this line even if
he wants to enter the Center Game. If White wants to play the ‘De
Jouy’ Variation, this move-order is particularly smart as it prevents
one of Black’s main responses:
2.d4 exd4 3. xd4 c6 4. d3 invites 4...d5!, equalizing. We will
see how in the theoretical section.
2... f6 3.d4 exd4 4. xd4 c6 5. d3

5... c5!
This must be Black’s most principled response, ‘profiting’ from
White not playing 5. e3.

123
An important line is 5... b4 6. d2 0-0 7.0-0-0 e8 8. ge2! and
White has a much easier game than with the queen on e3, as ...d7-d5
is prevented. We will look at this line in depth in Game 44 in
Chapter 6.
6. f4
The bishop likes to go to f4 in this variation.
6. g3 will be examined in Chapter 6.
6...0-0

7.f3
Allowing ...d7-d5, which Mamedyarov apparently didn’t mind. The
alternative is 7.0-0-0.
7...d5 8. xd5
Going for the endgame.
Bad is 8.0-0-0? d4 9. b1?? (Mamedyarov-Le Quang Liem, St
Louis rapid 2021) 9... b4! 10. b3 a5.
8... xd5 9. xd5

124
Bad was 9.exd5? b4 10. d2 e7+ 11. e2 d8 12.c4 f5,
winning.
9... xd5 10.exd5 e8+?!
What harm can such a check do? Still, it was better to move the
knight immediately.
The tactics continue after 10... b4 11.0-0-0 xa2+ 12. b1 b4
13.c3! when both players still need to be accurate: 13... f5+ 14. c1
c2! 15. d3 e3+ 16. xc2 xd3+ 17. xd3 xf4 18.g3 d6
19. e2 with a level endgame, certainly worth investigating further.
11. e2 b4 12.0-0-0

Can this really be something for White? The engine says yes!
12... f5
If 12... xa2+ 13. b1 b4 14. g3, and Black has minor problems
developing all his pieces. After 14...c6 15.dxc6 xc6 16. d3 e6
17. e4 followed by d6, White can still play for a little something.
13. g3?!

125
Yet again, Mamedyarov is going for forcing tactics in the endgame.
Objectively speaking, it wasn’t White’s best option, but it’s
certainly fun.
After 13. c3, play might develop similarly as the line mentioned
above: 13... xc2 14. b5! ed8 15.g4 g6 16.h4 h6 17.h5 h7 and
now 18. d3 e3+ 19. xc2 xd3+ 20. xd3 xf4 21. e4 d6
22.f4 is again slightly better for White.
13... xc2 14.a3 xd1 15.axb4 xb4?!
Better was 15... e3+! 16. xe3 xf3! 17. d2 xd5 and Black has
equalized, if not more.
16. xd1 e1+
16... ad8 17. c4 c6 18. b3 cxd5 19. c2 .
17. c2 d8 18. c4 xh1 19. xh1 d6 20. xd6 xd6

This position is difficult to evaluate for humans. The knight on h1

looks very far away, but so does the black king. Meanwhile, Black
struggles to create a passed pawn. White is somewhat better, but
with correct play, the game should end in a draw.

126
21. f2 f8 22. c3 e7 23.b4 h6 24. g4 h5 25.h3 c6?!
25...f5 26. f2 h4! was correct, although not easy for a human.
26. d4 f6 27.f4!
Now White has the upper hand. The black rook is terribly misplaced
and from here things go from bad to worse for the Dutch number
two.
27...cxd5?! 28. e2! h4

29. f1!
A cunning move. The black rook is trapped.
29... e6 30.g3 h5 31. e3 b6 32.g4 h6 33. f5 g6 34.b5 h5
35. e2 hxg4 36.hxg4 1-0
Zugzwang! A magnificently played endgame by Mamedyarov.
In other variations, too, things were moving. We already saw the
fianchetto (...g6) set-up in the first chapter, when we analysed the
game Mieses-Spielmann, Breslau 1912. Mieses played the
aggressive h2-h4 on move eight, anticipating later attacking ideas

127
for White in openings such as the Sicilian Dragon and the Pirc
Defence.
Wei Yi, Mamedyarov and Dubov weren’t the only top players who
gave the Center Game a try in the internet chess jungle. It’s time to
look at a game played by the best player in the world, Magnus
Carlsen. Carlsen has always been a player who loves to experiment
with offbeat and eccentric openings.
The following bullet (1-minute) game was played against Alireza
Firouzja, the world number two at the time. I happened to witness
this game live on Lichess, and was greatly impressed by Carlsen’s
intuitive grasp of some of the key themes in this opening. Magnus
shows it all: quick development, opposite-side castling, a pawn
storm against the black king, and a tactical finish on the dark
squares.
Magnus Carlsen (2847)
Alireza Firouzja (2759)
lichess bullet game 2021
1.e4 e5 2.d4 exd4 3. xd4 c6 4. c4
Carlsen has played the Hall Variation several times in bullet games.
4... f6 5. c3 e7!?

128
6. e3
Perhaps the bishop is even more actively placed on f4, though there
are hardly any practical examples. 6. f4 d6 7.0-0-0 e6 8. e2 0-0
9.f3 was somewhat better for White in a recent game Kovalskyi-
Hebden, Cambridge 2024.
6. g5 was played by Hall himself in 1888, and also several times by
Daniel Dardha in 2022. White would like to follow up with f2-f4.
6...0-0 7.0-0-0 d6
7... g4 8. e2 xe3 9. xe3 d6 10.f4 f6 11. f3 xc3 12. xc3
f6 13. xf6 gxf6 was Mieses-Rubinstein, Berlin 1918.

129
White had a substantial endgame advantage and went on to win after
91 moves.
The best move according to the engines was 7...a6!, intending, of
course, ...b5.
8.f3
White didn’t necessarily need to fear the exchange of the e3-bishop.
8.f4 was another attractive idea, as 8... g4 9. e2 xe3 10. xe3
would have transposed again to Mieses-Rubinstein, which is
somewhat better for White. White has easy development in this line.
8... e6
8... e5 was more accurate.
9. e2 a6?!
This is too slow as now White can build up his standard attack, and
the World Champion does so without hesitation.
After 9... e5 10. e1! c4 11. xc4 xc4 12. ge2, White is better,
but Black would still have a normal position.

130
10.g4! b5 11.g5!
Profiting from the fact that 11... h5 12.f4 is already close to
winning for White. As can be seen, having the queen on e2 has
some nice advantages over the more traditional lines.
11... d7 12.h4
There was nothing wrong with 12.f4 .
12... c4 13. f2 xf1
In bullet chess, such exchanges happen automatically.
Better was 13...b4.
14. xf1 ce5 15. b1 c4 16. c1
Carlsen wants to keep his beautiful dark-squared bishop.
16... e8 17. h3 f8 18.h5 c6 19.g6 fxg6 20.hxg6 h6 21. f4
21. g5!.
21...b4 22. ce2 f6
Firouzja should have tried to exchange queens: 22... b6, after
which White would have had to go for 23. d4 .
131
23. h5 e5 24. eg3 ab8 25.f4
White is ready to roll!
25... c5
The queen exchange comes too late now.
26. xc5! xc5 27.f5
A terrible position for Black. He will be crushed on the dark
squares, as so often in this opening.
27... xe4 28. xe4 xe4 29.f6 b7 30.f7+ h8

31. xh6
Removing the last defender. It’s all over.
31...gxh6 32.g7+ xg7 33. xg7 d2+ 34. c1 1-0
And so we have arrived at the present day. In 2023, the Indian
grandmaster Arjun Erigaisi (born in 2003), who is the world’s
number four at the time of writing, added the Center Game to his
opening repertoire, not only as a surprise weapon in blitz and rapid,
but also in serious games. In an interview on the C-Squared podcast

132
from May 2024, Nepomniachtchi said of Erigaisi: ‘Arjun is very
dynamic. He bluffs a lot.’ I think it was meant as a compliment.
Bluffing and being creative have always been two sides of the same
coin. In an email to me, Erigaisi remarked about a particular Center
Game Variation that, while it is objectively good for Black, ‘the
moves are not all that obvious to a human eye and it leads to
interesting positions.’
The young Belgian grandmaster Daniel Dardha, who won the
Belgian Championship at the age of thirteen in 2019, is another
player who regularly experiments with the Center Game. Echoing
Erigaisi’s words, he wrote to me (speaking of the Hall Variation):
‘Black is objectively fine there but has to make some strange and
unnatural decisions (...). If he doesn’t play concretely, things can go
badly for Black very fast.’ It’s a familiar story by now. Forcing
Black to play difficult and unusual moves is worth a little risk with
White. Mikhail Tal himself once said: ‘There are two types of
sacrifices: correct ones, and mine.’ Anyone who wants to play the
Center Game must be part bluffer and part artist.
There are many other grandmasters who have tried the opening in
the past couple of years. Many of them are excellent blitz players,
prepared to jump into the deep end, as long as it creates lots of
practical chances. Here are just a few other grandmasters who have
tried the Center Game (or a hybrid) in recent years: Nodirbek
Abdusattorov, Veselin Topalov, David Anton Guijarro, Venkatesh
Pranav, Vasif Durarbayli, Nijat Abasov, Alexandr Shimanov,
Aleksander Indjic, Xu Xiangyu, Luka Paichadze and Thomas
Beerdsen. We will meet many of them in this book. New games
from both online and ‘over-the-board’ tournaments, containing fresh
ideas devised by amateurs, grandmasters and engines, are now being
published every week.
The Center Game isn’t trendy. It’s here to stay.

133
Chapter 3
Paulsen Variation
4. e3: old main lines
‘You are old, father William,’ the young man said, ‘And your hair
has become very white.’
Lewis Carroll, Alice’s Adventures in Wonderland

In this chapter, we are going to investigate developments in the ‘old

main line’ of the Berger Variation, in which Black plays ... b4, ...0-
0 and ... e8.

134
 position after 7... e8
This is Black’s most natural way of fighting the Paulsen Variation
with 4. e3, developing rapidly and putting immediate pressure on
the white centre. The key idea is to win e4 or play for the break
...d7-d5. It’s clear that the queen on e3 will, at some point, have to
move away from the e-file. White has traditionally met this
challenge by sacrificing the e4-pawn in return for open lines.
Indeed, White often gets dangerous practical chances, but modern
engines show that Black is objectively fine in these lines.

The Tarrasch Gambit

The sharp 8. g3 will be examined in the first two games. There is a
lot of sharp theory here, and both players need to know exactly what
they’re doing. The line with 8... xe4! followed by Shirov’s
10... e7! (see Game 26) is the better of the two possible ways to
win e4. In blitz or bullet games, the Tarrasch Gambit may still work
for White, but in serious games, I doubt it’s a good idea to try it.
Is it a pity that these old and beloved (by me at least) gambit
variations do not seem to be correct under silicon scrutiny? So many
hours by so many great players have gone into the investigation of
these lines! I like to think the beautiful and creative ideas from the
past aren’t lost to us. They are still here for everyone to see.
Alexey Kislinsky (2402)
Denis Shilin (2374)
Kyiv 2004
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 f6 5. c3 b4 6. d2 0-0
7.0-0-0 e8 8. g3

135
8... xe4
Alapin’s move, on which Greet writes: ‘There is no doubt in my
mind that the text must represent the acid test of the Center Game.’
Fortunately, it’s only an acid test of the Tarrasch Gambit.
A) A few words on 8...d6, which is a solid, more quiet approach,
and was Karpov’s choice against Shirov. White should be happy
here, though. With 9.f3 he aims for his typical ‘slow attack’ on the
kingside:
A1) 9... e5 10.h4 is fine for White, see Shirov-Karpov, Dos
Hermanas 1995, analysed in Chapter 1;
A2) 9... e6 is often played automatically, but after 10.h4 White’s
plan is very simple: push the h-pawn as far as possible and then
develop slowly. Black, meanwhile, doesn’t have a clear plan
himself. A sample continuation might be 10... e5 11.h5 h8 12.h6
g6 13.a3 a5 14. ge2 with a dream position for White. g5 and
h4 are already in the air;

136
 A3) The clever 9...d5!, preventing the ‘slow attack’, has only
appeared in a single game so far. Play may continue 10. xd5 xd5
11.exd5 d6 12.f4 b4 13. xb4 xb4 14. f3 and White is better.
B) 8... xc3 9. xc3 xe4 is a much worse version of the 8. f4
main line (see Pirs-Khanas below), as White has an extra tempo
here, which he uses immediately: 10. d3 e8 ( 10... g4?? 11. xf6
now does win) 11. f3 d5 12. he1 with great compensation for the
pawn;
C) 8... h5!? is rare, but actually quite smart. Black might go for a
repetition that White cannot easily avoid: 9. f3 f6 10. ge2 e5 (
10...d6 11.h3 and the queen can hide on h2) 11. g3 h5 and the
only way to avoid a repetition is 12. h3 g6 13.g4!? with an unclear
position;
D) For 8... xe4 see the next game.

9.a3
As we have seen in Chapter 1, this was the move Shabalov used to
bring the Center Game back to life.

137
9. g5!? may be White’s best move objectively, and if you’re happy
with a draw, it’s playable. Black continues 9... xc3 ( 9... e5 10.f4 )
10. xc3 d5! ( 10...h6 11. xf6 xf6 12. xf6 gxf6 13. e2. This
endgame, which is equal, has been played twice by Muñoz Pantoja,
and was first seen in the game Granda Zuniga-Slipak, Buenos Aires
1996). This move has never been tried in a practical game. Black
has a slight endgame advantage after 11. xf6 xf6 12. xf6 gxf6
13. xd5 g4! 14.f3 e1+ 15. d1 xd1+ 16. xd1 f5, as White
can’t play 17. e2? ( 17.c3 ) because of 17... b4!.

9... d6!
This is Black’s best move, but before explaining why, we should
investigate the alternatives:
A) 9... a5 was considered strongest for a while, but after 10. g5!
things are similar to the note on 9. g5, only somewhat more
favourable for White ( 10.h3!? is what I tried a few times in blitz
against Shirov (see Chapter 1), but Black is fine after 10... e8!
11. f3 e7! 12. d3 c6! , intending ... c7; 10.f3 h5 11. g5 (
11. h3? h4! ) 11... xg5 12. xg5 e5 13.f4 e8 Paschall-

138
K.Szabo, Budapest 2009): 10... xc3 11. xc3 d5 12. xf6 xf6
13. xf6 gxf6 14. xd5 g4 15.f3 e1+ 16. d1 xd1+ 17. xd1 f5
18. e2 and now we see the difference with the previous line: Black
no longer has ... c6-b4...! The position is level;
B) 9... g4!? is a strange and concrete move which I’ve never
fully understood. Play is complex and unclear after 10. e3 (
10. h3?! c5 Nepomniachtchi-Ding Liren, Chess.com Speed
Chess 2019) 10... f8! 11.f3 ( 11. e1!? is new and untried, and
leads to an unclear position worth studying; 11.h3 g6 12. d3 (
12. ge2 d5 ) 12...d5! 13. xg6 hxg6 and Black had fantastic play
for the exchange in Varavin-Gusev, Tula 2001. Greet analysed the
position in some detail and claimed White had play even here, but
the modern engines just don’t see it) 11... g6 12. f2 d5 13. d3 d4
14. xg6 hxg6 15. g5 (Nepomniachtchi-Aronian, Mainz rapid
2009), and now 15... e7! 16. ce2 c5 17. xf6 gxf6 18. xd4
f5! is unclear, although I think it’s easier to play for Black;
C) 9... e7!? is a virtually untested exchange sacrifice, but it
might be a great idea. White’s best is 10.f4 ( 10. xe4 xe4 11. e3
d5 gives Black excellent compensation: 12. e1 ( 12.f3?? c5
Koch-Frielingsdorf, freechess.de 2017) and now the engines want
Black to play 12... b8! and claim an advantage) 10...d5 11. d3 e6
12.f5 d6 13. f4, but Black remains better after 13... d7! with the
idea ... c5;
D) 9...a5!? is also completely new but perfectly playable. We saw
it already in Chapter 2. White’s can’t take on b4: 10.axb4? ( 10. g5
is unclear) 10...axb4 11. b1 d5 and Black slowly wins, e.g. 12. d3
g4 13. e3 e6 14. f3 d4 15. e2 a1 16. c4 e4 17. d3 f5.
10.f4 e8

139
It’s perfectly clear that White has compensation for the pawn, and in
a blitz or rapid game, it is perhaps even easier to play as White. But
according to the engines, Black is doing very well here. Analysts in
the 1990s looked differently at the position. Black needs to find the
right piece configuration and, most importantly, should not be
intimidated by White’s attack – which is easier said than done!
Greet wrote that if ‘a final verdict on this variation is ever
uncovered, then I would not be surprised to learn that – with perfect
play – Black possesses enough resources to consolidate his material
advantage.’
11. d3
The most natural move, but perhaps not White’s best chance.
It is possible to play differently by means of 11. f3 f8 12.f5 (not
12. g5? d5 13. d3 h6 14.h4 e7! , which occurred in the father-
son blitz duel A.Dardha-D.Dardha, Titled Tuesday 2022), which
leads to practical chances even though the position is objectively
fine for Black. A few sample lines: 12...d5 13. g5 xf5 ( 13... d6
14. h4; 13... d6 14. b5! ) 14. xd5! ( 14. xd5? e4! Zozulia-

140
Sriram, La Fère 2005) 14... d6 (the queen sacrifice 14... xd5!? is
also not bad) 15. h4 g6 and Black is slightly better.
11... f8
11... c5 12. f3 d5 ( 12...d6?! is passive and White has the strong
13.f5! suggested in Chess Informant 70 and first played in
Paragua-Leyva Proenza, Madrid 2003) happened in Shabalov-
Ivanov, Key West 1994 (see Chapter 1), and here too, White is
fighting for equality after 13. he1 xe1 14. xe1 e7 15. e5 d4!.
12. f3
12. h4 d5 13. f3 e7! and Black will follow up with ...c7-c5 and
is better.

12...d5
12...d6?! is too passive and allows 13.f5!, e.g. 13... e5 ( 13...d5 was
played by Harikrishna against Nepomniachtchi in Moscow 2007,
but 14. g5! would have been strong here) 14. xe5 xe5 15. hf1
and White had promising compensation in Bazaj Bockai-Theissl
Pokorna, Krk 2004.

141
13. he1
13.f5 d4 ; 13. h4 d4 .
13... xe1
13... e7!? 14. f2 c6! ( 14... f5? loses to 15. xe7, Lazic-Pihajlic,
Vrnjacka Banja 1996) 15.h3 b5! was also more than OK for Black,
whose attack certainly isn’t slower than White’s.
14. xe1
14. xe1!? g4 15. f2 might still offer White some practical
compensation, but here, too, the engine doesn’t really see it.

This position occurs 36 times in the database, which is an indication

that White players used to believe in this line quite strongly. But
these days, the engine evaluation is leading, and it simply says -1.2
for Black. The main reason is:
14... e7!
A) 14... g4!? 15.h3?! xf3 16. xf3 (Devallee-Glod, Le Port
Marly 2013) 16... d4! ;

142
 B) 14...d4?! 15. e4 f5 16. fg5 xe4 17. xe4 gave White
compensation in Papatheodorou-Nielsen, Panormo 2002, although
he later lost the thread of the game.
15. e5
15.h3 was White’s best move, but was only seen in two games by
lower rated players. Still, after 15... f5! Black has a serious
advantage.
15...c5
15... f5?! 16. xf5 xf5 17. d3 e7 18.g4 with compensation for
White.
16. f2

16... c6?!
Black could have capitalized on his material advantage with 16...c4!
17. e2 e6 when nothing seems to work for White, e.g. 18. e3 (
18.g4 d4 19. b5 c3 20.bxc3 dxc3 21. xc3 g6 and the white king
is horribly exposed) 18... f5 19. d1 h5! 20.h3 xe3 21. xe3 c7
22.g4 c5 23. g3. It may still look like something for White, but
143
alas, Black is again first with 23...h4! (or 23...d4 24.g5 h4! 25. xh4
dxc3 26.gxf6 b6! ) 24. xh4 e3+.
17. h4! e6?!
17...h6! 18.g4 c4 and again, Black comes first.
18.f5 d7?
18... xe5! was mandatory, after which White only gets vague
compensation with 19. xe5 d6 20. e1 d7.

19. g4?
Missing his chance!
19. xf7! was already pointed out by Greet, who analysed 19... xf7
20. xd5! xf5 ( 20... xd5 21.f6 with a proper mess, which is
objectively equal) 21. xf5 xd5 22. e6+ xe6 23. xe6 xe6,
which is, again, evaluated by the engines as 0.00.
19... xg4 20. xg4 c4 21. f1 d4?!
21... e7!, with the idea of ... c6, was better and would still have
retained an objective plus.
22. xc4
144
22. d5 c3 23.bxc3 a5 may have looked scary indeed.
22... f6?!
22...dxc3 23. xf7+ h8! 24. xc3 e7 was better, but things were
already pretty messy.
23. e4! xf5 24. h4!
Now White has real compensation.
24... e5?!
24... e8 25. g5 xe1+ 26. xe1 e5 still favoured Black slightly.
25. f1
25. g3! was the way to do it; the position is roughly balanced.
25... g4 26. xg4 xg4

Black is still slightly better in this endgame. The rest of the game is
of no significance to our understanding of the opening. White
managed to draw on move 47.
Ori Taichman (2299)
Yuan Qingyu (2373)

145
Titled Tuesday blitz 2023
The Israeli FM Ori Taichman’s name appears many times in the
database in Center Game positions. He has been trying hard to make
the Tarrasch Gambit work for White, but in this game, he was not
successful.
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 f6 5. c3 b4 6. d2 0-0
7.0-0-0 e8 8. g3 xe4 9. xe4 xe4

10.c3
What else could be tried?
A) Playing yet another piece twice just to attack the pawn on c7
can’t be it: 10. f4?!. Curiously, this move was also played once by
Nepo against Vallejo Pons, but he lost badly: 10... f6! (already
played in Mieses-Capablanca, Berlin 1913, as mentioned in Chapter
1) 11. h3 ( 11. xc7? d6 12. xd6 h6+ 13. b1 xd6 ) 11...d6
( 11...d5 is equally good. Best for White seems 12. g5 e5 13. f4
which ended in a repetition in Paragua-Brain, New York 2023, but
in fact Black could play for a win with 13... e7! 14. d3 and now

146
the typical 14... d4! ; 11...h6!? ) 12. d3 d4! 13. b1? e2
14. f3 (Nepomniachtchi-Vallejo Pons, Moscow 2007)

analysis diagram
and now Black had the nice winning move 14... a3!;
B) 10. xb4 was once played by a young Muñoz Pantoja and is
not that bad according to the engines, but it’s hard to root for White
after 10... xb4 11.a3 e4 12. d3 e8 13. f3 d6 and there just isn’t
enough compensation. In his book A Complete Repertoire for Black
after 1.e4 e5 (2020), Yuriy Krykun remarks, rather sarcastically: ‘To
me it is not obvious at all why White gave up a pawn but that is
something he does sometimes in this line hoping to develop
quickly’;
C) After 10. g5 e7 11. xe7 xe7 White essentially plays a
position a pawn down hoping for vague compensation. It just isn’t
there.
The text move is White’s best practical chance, despite the fact that
the engines rate it lower than alternative bishop moves.

147
10... e7!
This move is probably White’s biggest headache in the Tarrasch
Gambit. The first game with this move in my database is from 1999,
but Shirov suggested it to me two years before that. I tried it shortly
afterwards in a game against a chess friend (see below). The text
move is not very intuitive: it feels a bit scary to cut off the rook on
e4, and, therefore, it may not be found over the board by an
unprepared opponent. It is perhaps worth risking in blitz or rapid,
but in an important encounter, this is not what you want to face.
Black threatens 11... h4, and this is surprisingly problematic for
White.
A) 10... f8 11. d3 was analysed in Chapter 1 (Marshall-
Pillsbury, Buffalo 1901). Now, Black should have played the bold
11... a4! when 12. h3! is a new move and White’s only chance to
stay in the game. A sample line would be 12...d6 ( 12... xa2 allows
13. g5 g6 14. xf7! ) 13. he1 g4! 14. e3 g6! (if 14... xg2,
15. e4 with compensation) 15. f4 e5 ;
B) 10... d6 11.f4 led to a great success for Taichman against the
top-rated player Kirill Alekseenko (Titled Tuesday blitz 2021):
11... f6 12. h3 h6?! ( 12... a4 13. d3! is also difficult to play for
Black) 13. d3 e7 14. f2! g6 15. e4 g7?! 16. xd6 cxd6 17.f5
and White was winning, and won.

148
11. f3
This was Greet’s recommendation in 2008, and it is indeed White’s
best try.
A) 11. d3 h4! 12. f3 e8 and White doesn’t have enough
compensation: 13. f4 ( 13. h5 g6 14. h6 e5 15. e2 d6
(Nadj Hedjesi-M.Pap, Valjevo 2011) was analysed in Chess
Informant 112) 13... e5 14. c2 d5 ;
B) An important detail, which we will also see in the game, is that
after 11.f4 d5 12. d3

149
 analysis diagram
Black sacrifices the exchange and gains a huge advantage: 12... f5!
13. xe4 ( 13. f3 c8! ) 13... xe4 and Black’s attack plays itself;
for instance, 14. f3 a5! ( 14... d7 was also good in Rudd-
Avrukh, London 2010) 15. he1 c4 16. e3 d7 followed by
... b5, etc., winning.
11...d5
Not the only path to an advantage.
A) Greet analysed 11...d6 12.h3 and now continued with 12... f6,
but 12... e5! is a strong improvement for Black, played first around
the same time Greet’s book was published (sic transit gloria mundi):
13. xe5 xe5 14.f4 a5 15. c4 f5 and after 16. b3 ( 16.b4 d5
Arndt-Zschoch, Germany email 2009) there’s that annoying move
16... h4! again. Black has a great position;
B) An even bolder continuation is 11... a4!? when White just
doesn’t have enough;

150
 C) 11...h5!? was played twice by Virgilio Gagliardi in
correspondence games, and is also annoying for White.
12. f4
12. d3 g4 13. h3 h4 14. g3 g4 led to a draw by repetition in
what was possibly the debut of this line, namely a rapid game
between my chess friend (and fellow Center Game player) Paul
Janse and myself (playing Black!) in Amsterdam, 1997. I should
have played on: 15. h3 g6! ( 15...h6 was mentioned by Greet and is
also good) 16. h6 xg2 17.h4 f8 and Black is winning.

12... f5?!
Going for a similar exchange sacrifice as in the note to White’s 11th
move.
Very strong was the hitherto untested move 12...h5!, threatening
...h5-h4. Everything works for Black, e.g. 13. xc7 ( 13.h4 g4
14. d3 a4 15. b1 d7 and White has no compensation) 13... e8
14.h3 f5 15. xd5 h4 16. h2 c8 and Black will soon start
attacking.

151
13. d3 c8 14. xe4 xe4

The engine still evaluates this position as equal, but it’s easier to
play for Black. The light squares around White’s king are horribly
weakened.
15. g5?
Far too optimistic. The knight was needed elsewhere.
After 15. he1 b5! it is again Black who has the initiative, and White
needs to play accurately to keep the balance: 16. e5 xe5 17. xe5
f8 18. e2 a6 19. d2! xa2 20. e1 with an unclear position.
15... g6 16. h3
Trying to find salvation in the endgame but, of course, Black keeps
the queens on the board.
16. xd5 h6 17.h4 hxg5 18.hxg5 e6 19. h4 f6 wins for Black.
16... f5!?
16... d8! 17. f3 d4! with the idea ... d8-d5 was even more
forceful.
17.g4 xg5 18. xg5 xg4 19. d3 xd1 20. xd1 h6
152
Black is two pawns up and White doesn’t have the slightest play on
the g-file. It’s basically over, and White threw in the towel after
move 34.

Polgar’s 8. f4
In Games 27, 28 and 29, we will examine the state of play after
8. f4, which, by the way, is Gorny’s main line. First, we will see an
uncommon but interesting way for Black to avoid theory by
challenging the white queen right away ( 8... d6!? ).
Next, we will turn to lines where Black wins the pawn on e4. With
the queen on f4, White has nothing to worry about after 8... xc3
9. xc3 xe4, which is actually dangerous for Black. Taking with
the rook on move 9 is more promising, but even here, things are not
so clear, as Gorny has shown. I would recommend White players to
investigate the unexplored move 10. d2, as this avoids lots of
complex theory and looks equally as strong as the more traditional
10. g3. In general, Polgar’s move is great for faster time controls,
but before playing it in a serious game, I would think twice.
Gaetano Signorelli (2194)
Sabino Brunello (2510)
Italy team championship 2023
We start with a game in which the Italian grandmaster Brunello
plays a rare but strong idea. I find it all pretty annoying for White,
even though the positions are objectively equal.
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 f6 5. c3 b4 6. d2 0-0
7.0-0-0 e8 8. f4 d6!?
A clever and dangerous idea, which White must be prepared for, or
he risks losing very rapidly! I think this may yet turn out to be
Black’s most practical choice against Polgar’s move.
A) Black’s main alternative if he doesn’t want to win the pawn on
e4 is 8...d6.

153
This push is not bad, but it doesn’t pose White too many problems
either.

analysis diagram
The positions are a bit less tactical and White can often even afford
to ‘lose’ a tempo compared to the Tarrasch Gambit and still play
f4-g3. In my view, it is more consistent (and easier to remember)
for White to play h2-h4 before g2-g4 in these positions. A few
examples: 9.f3 and now:
A1) 9... e5 10.h4!?. This has never been played, but it seems OK
for White. He just wants to push the pawn up the board. A possible
continuation is 10...h5 11. ge2 c6 12. g3 d5! 13.exd5 a5
14. d4, which requires further research;
A2) 9...a5!? 10.h4!? ( 10.g4 is Gorny’s move in his Chessable
course, and is also interesting) 10...d5! is unknown territory again.
The position, whilst complicated, is objectively balanced. Both
11. xd5 and 11.h5 seem possible for White;
A3) 9... e6 and now:

154
 analysis diagram
A31) 10. ge2 is Gorny’s recommendation. White wants to play
an endgame after 10... h5 11. g5 ( 11. e3 c5 12. d3?? e5
13. b5 a6 14. xb7 a7 loses the queen; 11. h4?? is strangely
mentioned by Gorny, but it drops the queen) 11... xg5 12. xg5 and
White is perhaps a tad better;
A32) 10.h4!? is an attempt to go for the attack. What makes this
push attractive in a practical game is that 10...d5 ( 10... e5 11.h5
followed by g3, and White is fine) is now Black’s best move, but
it has never been played. After 11.exd5, Black must take with the
bishop: 11... xd5 ( 11... xd5?? loses to 12. xd5 xd5 13. xb4
xa2 14. c3 ) 12.h5 h6 13. ge2 and the position is roughly equal.
I rather like White’s chances here. The standard set-up is with g3
and f4.
B) Finally, there is Aronian’s move 8... e5!, which sometimes
transposes to 8... d6, but also has independent value. In fact, this
move is the engine’s top choice.

155
 analysis diagram
B1) 9.f3 c6! followed by ...d7-d5 or sometimes even ...b7-b5, and
Black can look forward to an easy game. 9... g6?! 10. g3 d6
11. f2 was somewhat better for White in one of Nepo’s original
8. f4 games: Nepomniachtchi-Aronian, Leuven blitz 2017;
B2) 9. d5 xd2+ 10. xd2 xd5 11.exd5 d6 and although the
position is equal, this seems slightly easier for Black to play;
B3) Gorny recommends 9.h4!? g6 ( 9...c6 10.h5 ) 10. h2 xc3
11. xc3 xe4 12. e1 h5 and says that even though the engines
prefer Black, ‘our practical chances are rather high as we are about
to activate our forces: we will kick the e4 away with f2-f3, push
pawns on the kingside and reactivate the bishop afterward via c3 or
h4. The position is anything but boring.’ Indeed, if Black is not
prepared for this, he will surely suffer. But who will play the rare
8... e5 without being prepared?

156
9. e3?!
White drops the queen back again. Perhaps he was hoping to profit
from the slightly odd position of Black’s bishop on d6, or go for a
repetition against his stronger opponent, but it’s just not working
out.
A) 9. f3! was the move to try, but it invites complications that
are completely new and difficult to assess, especially in a practical
game where Black is prepared: 9... e5 (also interesting is 9... e5
10. ge2 d5 (or 10...a5!? ) with crazy complications) 10. e2 with
an unclear position. A possible continuation is 10... b4 11.a3 a5
12.g4!? exg4 13.f3 e5 14. g2 and if 14...d5, White just
develops: 15. ge2, with compensation;
B) 9. h4? is the automatic move in a quick game, but it fails to
9... xe4!. Oops! I’ve fallen for it myself several times in bullet
games.
9... e5!
How ingenious!

157
Black has the same position as before White’s 8th move, but with
his bishop on e5 instead of b4. The difference? Well, obviously now
both g3 (Tarrasch) and f4 (Polgar) are prevented! That leaves
either 10.f3 or 10. c4, but both moves have drawbacks here.
10.f3
A) White would like to play 10. c4 but Black can continue to
harass the white queen: 10... d4! 11. g3 e5 12. d5 (otherwise e4
falls) 12...d6 13.f3 c6 14. b3 a5 and Black’s attack comes first;
B) 10. ge2 d5 11.exd5 b4 followed by ...a7-a5, and it’s clear
that Black is better;
C) After 10. f3 xc3 11. xc3 xe4 12. c4 xc3 13. xc3 d6
Black is a pawn up but White at least has some activity and open
lines.
10...d5!
Already an extremely depressing position for White, who now
followed up with a bad move as well:
11. g5?

158
11. f2 dxe4 12. b5 was a better attempt, but Black has a large
advantage after 12... e7.
11...dxe4 12.fxe4?
12. f4 d7 13. xe5 xe5 14. h4 h6 and it’s not clear what White
is doing. 15.fxe4 fails to 15... xe4.
12... g4
Liquidating into a winning endgame.
13. xd8 xd8 14. h3 xh2 15. e2 g4

There is no compensation for the pawn. Brunello won on move 54.

Julio Granda Zuniga (2595)
Jose Luis Fernandez Garcia (2470)
Pamplona 1992
We will now see some lines that score extremely well in games with
faster time controls, but which are not so easy for White if Black
plays prudently. I believe that if both sides play correctly here, the
game is very likely to end peacefully.

159
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 f6 5. c3 b4 6. d2 0-0
7.0-0-0 e8 8. f4
I’ve taken a slight liberty with the move-order: in the game Granda
Zuniga played 8. c4 xc3 9. xc3 xe4 and now transposed to the
game after 10. f4.
8... xc3!
The principled move, obviously.
9. xc3 xe4
Almost everyone who sees this line for the first time, thinks Black is
almost winning here already. But then White plays:
10. c4!

Suddenly, Black has a small problem on f7. There are many ways
Black can immediately go wrong here, which makes it a great line
for blitz (for White).
10... e7!
Probably Black’s best move.

160
 A) 10... e7?? is a move even grandmasters have played against
me in bullet and blitz, but it loses on the spot: 11. e1! g5 (
11... xc3 12. xf7+! xf7 13. xe8# ) 12. f3 xc3 13. xe7 xe7
14. xc3;
B) 10... xc3? 11. xf7+ h8 12. xe8 xa2+ 13. b1 xe8 (
13... c3+ 14.bxc3 xe8 15. xc7 ) is less bad for Black than it
may look at first sight (and I’ve lost this position quite a few times
in blitz against stronger opponents), but White is objectively much
better here, e.g. 14. xa2 d6 15. e2! (an accurate move; 15. f3 a5
in Bochkarev-Frolov, St Petersburg 1998, was a bit less convincing)
15... e6+ 16. b1;
C) 10... g5? 11. h3! e6 12. g3 followed by either f2-f4 or
h3-f4, with a winning attack;
D) It is less easy to refute Steinitz’s careful retreat 10... f6, but
here, too, White is doing well by simply developing his pieces and
attacking the black centre slowly: 11. f3 ( 11. h3!? ) and now:
D1) 11... e4?? occurred in no less than 8 games, but loses to
12. xf7+!, when 12... xf7 ( 12... h8 13. xf6 xf4 14. xd8 xf7
15. he1 f8 16. xc7 ) runs into 13. g5+;
D2) 11...d5? 12. xf6 xf6 13. xf6 gxf6 14. xd5 with a bad
endgame for Black due to his weakened kingside (Paichadze-Basso,
Titled Tuesday blitz 2023);
D3) 11...d6 12. g5 e6 13. xe6! ( 13. d3!? was played in
Winawer-Steinitz, Nuremberg 1896 – Chapter 1) 13...fxe6 14. he1
d5 15.a3! (also not bad was 15. g5 h6 16. g6 e7 17. g3 f5
(Warzecha-Dammer, email 2013) 18. f4!? with compensation; or
15. b3!? ).

161
 position after 15.a3!
White has beautiful compensation, completely in the spirit of the
modern engines. All White’s pieces are centralized and his bishops
are pointing towards the black king. A sample line runs 15... d6
16. h4 e7 17. d3 e5 18. b5! d6 19. c4! d8 20.g4! with the
next attacking wave to come.
E) 10... h8!? will probably transpose to the game after 11. xf7
e7. ( 11... f8? 12. xe4 xf7 13. f3 and White, who would
usually be happy to play this kind of position with a pawn less, is
easily winning here.)
11. xf7+
Winning back the pawn, but at the same time simplifying the
position, which is not to everyone’s liking.
The curious 11. xg7!?

162
 analysis diagram
was first played in an old game involving the Argentinian
grandmaster Oscar Panno, and deserves attention, if only as a
surprise weapon. The idea is that after 11... xg7, White has the
slow-looking move 12.f3! and the black king is suddenly
dangerously exposed. If Black doesn’t fancy wild complications, he
should probably take a draw after 12... d6 ( 12...d5!? is what Panno
played, but that isn’t easy either after 13.fxe4! ( 13. xd5? d6, and
Black was winning in Rollansky-Panno, Buenos Aires 1965)
13... xe4 14. g3+ g4 15. c3+ d4 16. d2 and now it’s probably
safest to exchange queens with 16... g5 when the endgame is
equal) 13. g5+ h8.

163
11... xf7
11... h8!? is not silly. White should continue 12. b3 xc3 13.bxc3
d6 14. f3 e6 15. he1 and White had slightly the better chances in
Morovic Fernandez-Gomez Esteban, Las Palmas 1991.
12. xe4 g5+
12...d6 was possible, but White has an easy game after 13. f3.
Instead, 12... xf2? drops at least the exchange after 13. f3 as
13... xg2? 14. d5+ (or 14. hg1 ) 14... h8 15. hg1 g6 16. g5 is
game over.
13. d2
Also possible was 13. d2, but here, too, White can’t really avoid
the endgame: 13... f5 14. xf5 xf5= Degraeve-Marcelin, Issy-les-
Moulineaux 2000.
13...d5
13...d6 is a new move which is perfectly alright as well; for
example, 14. f3 g4 (not 14... xg2?? 15. g1 xf3 16. e8+ f8
17. xg7+ ) 15. xg4 xg4 16. d4 e5 with equality.
164
14. xd5 xd5 15. xd5 xf2 16. f3
Sacrificing a pawn for activity.
It was also possible to play 16. d2, but White is a little more
passive after 16... f1+ 17. d1 xd1+ 18. xd1 g4+ 19. c1 f8.
16... xg2 17. e1
It might have been easier to go 17. e5 as 17... e6 (not 17... xe5?
18. d8+ ) 18. xc6 bxc6 19. d2 xd2 20. xd2 is a dead draw.
17...h6 18.b4
With an equal position.

White has enough activity for the pawn. In the database, the game
score now becomes incomprehensible.
18... g4
I think it’s likely that this logical move was actually played (
18...g6?? is what the database gives, but it’s hard to believe this:
19.a4?? ( 19.b5 wins the house) 19... g4 etc.).
The game may have continued:

165
19. d2 a5 20.bxa5 f8 21.h4 f5 22. c5 d7 23. e4 g4
24. b5 c8 25.h5 f5
And probably somewhere around this point, a draw was agreed.
Matjaz Pirs (2485)
Valeriy Khanas (2391)
ICCF email 2022
This game shows the most critical line for Black against the Polgar
Variation. Things aren’t easy for White here, but Gorny expresses
his faith in the line, having analysed it with strong engines. His
conclusions aren’t wrong, just a bit optimistic – which is OK for a
repertoire course.
My own advice would be to look more closely at the move 10. d2,
which actually looks kind of nice and easy.
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 f6 5. c3 b4 6. d2 0-0
7.0-0-0 e8 8. f4 xc3 9. xc3

9... xe4!

166
The strongest move, and some believe it to be more or less the
refutation of Polgar’s move 8. f4. At the same time, it does require
Black to be relatively accurate and, more importantly, to be aware of
some details that White may have studied more deeply.
10. g3
The automatic move, but Bologan observes that this is the same
position as in the Tarrasch Attack after 8. g3 xc3 9. xc3 xe4 –
with an extra tempo for Black! This means that White can’t
immediately play f1-d3 here. Instead, he usually follows up with
f2-f3, g1-e2-f4, h2-h4 and sometimes slides the queen back to f2,
after which g2-g4 can be played.
10. d2!? has only been played once, according to my database, but
in fact, I think this may be White’s best (and easiest) way to avoid
heaps of complex theory and still get a playable position in the
8. f4 line.

analysis diagram

167
The advantage of this move over the text is that the g3-square is
available for the white knight, for example: 10...d6 ( 10...d5 11. b5!
( 11. xf6 xf6 12. xd5? f5 is a disaster for White) leads to an
entirely new type of position in the Center Game. White has
compensation for the pawn after, e.g., 11...a6 12. xc6 bxc6 13. e2
e8 14. g3 with ideas like g5 and h2-h4 in the air) 11. e2!
(better than 11. d3 as the bishop is vulnerable here: 11... e8
12. e2 e5!? ( 12...a5 may be even stronger) 13. g3 xd3+
14. xd3 d7 and it’s hard to believe White has enough here)
11... e8 12. g3 followed by h2-h4-h5, with decent compensation.
Let’s make a few natural moves just to get a feeling for the position:
12... e6 ( 12... e5 13. e2 e6 14.f4 c4 15. d4 c5 16. f2 )
13.h4 e5 14.b3 a5 15.a4 d5 16. b2 b4 17.f4! followed by f4-
f5, and White’s position looks promising.

10...d5
10...d6 is the less popular alternative, which is nevertheless
evaluated by the engines to be equally as strong as the text move (
10... g4?? is a blunder now due to 11. xf6 ). 11.f3 ( 11. h3!? is a

168
new move, which is worth investigating further) 11... e8 12. e2
(heading for f4) 12... f5 13. f4 (threatening g5; 13. f4!? e6
14. xf6 xf6 15. xf6 gxf6 16. f4 probably gives White sufficient
compensation even in the endgame; 13.h4?! e5! would have been
better for Black) 13... e5 14. f2 ( 14.h4? was played by Carlsen
against Tang in a bullet game (lichess 2021), but it’s a bad choice.
Play continued 14... d7 15. c4? ( 15. b1 ) 15...h6? ( 15... xc2! )
16. he1 h7?? ( 16... xc4 ) 17. xe5 1-0) 14... d7 15.g4.

analysis diagram
Gorny says of this position that ‘the engine thinks it’s roughly
balanced, but I would rather be White.’ I see what he means, as the
position looks optically promising, but one thing modern engines
have taught us in the Center Game is that we should always consider
active options for Black as well. Here, the move 15... a4! is
precisely such an active possibility. Gorny analyses White’s best
chance: 16. d4, sacrificing the a2-pawn, but the line continues even
further: 16... xa2 17.gxf5 c5 18. xd6 xf3! 19. xf3 a1+
20. d2 e4+ 21. xe4 xe4 22. g2 a4 23.b3 e8 24. xe4
169
 xe4. My engine actually slightly prefers Black in this position and
I can understand why: he can follow up with ...b7-b5 and the white
king is slightly exposed. On the other hand, there is still plenty of
play for both sides.
11.f3
Nepomniachtchi has had this position four times with White, and
scored 1½ points with it. To be fair, his two losses were to two
former World Champions: Magnus Carlsen and Viswanathan
Anand.
11. d3? loses to 11... g4 12. f3 d4.
11... e6!
Objectively better than moving the rook back to e8, but less
intuitive.
11... e8 12. e2 occurred in two games with Nepomniachtchi as
White. It seems to me Black has an easy way out here:
A) 12... e7!, intending ... b4. Black wants to simply follow up
with ... e6 and ... ad8. 13. xd5 ( 13. f4 d4 14. xd4 xd4
15. xd4 e3+ 16. d2 e6= ) 13... b4! 14. xb4 xb4 with
equality;
B) Black could also play 12... f5 and now 13.h4 is Gorny’s move
( 13.a3?! was Nepomniachtchi-Bocharov, St Petersburg rapid 2018:
13... e7! ; 13. f4!?; 13. h4!? ). He continues 13... e7 14. f4 (
14. g5 may be better) 14... e6 and ended his analysis here, saying
the position is balanced but that he prefers White, who can look
forward to a ‘pawn storm rolling on the kingside’;
C) 12...a5 was Nepomniachtchi-Carlsen, Leuven blitz 2017. See
Chapter 2.
12. e2 d6!

170
First played by Anand against Nepo in 2017. Gorny outlines why he
is not so worried about the evaluation (Black is better) of the
engines here: ‘We have the pair of bishops and attacking chances
against Black’s king. (...) My honest opinion is that with perfect
play from both sides, the game should end in a draw. But this also
means that Black needs to find all the best moves.’ As we will see,
there is also plenty of room for White to go wrong if he is not well
prepared. Handle this line with care!
13. d4
Not just blocking the d-pawn but also threatening d4-b5.
13. f4? was a mistake by Nepo, punished directly with 13...d4!
14. xd4 d5 in Nepomniachtchi-Anand, St Louis blitz 2017.
13... xd4
Or 13... e7? ( 13... d7? 14. b5 ) 14. d3?! ( 14. b5 and 14. b4
clearly favour White) 14...c5?? 15. b5 c6 16. xf6 xf6
17. c7 Carlsen-Firouzja, lichess bullet 2021.
14. xd4 f5

171
Gorny also analyses 14...b6 15.h4!? with an unclear position (
15. d3 looks OK as well; 15. e5 h5 ).
Black could also consider 14... e6!? but it somehow seems less
natural.
15.h4
15. f2?! d7! 16.g4 e6 17.h4 c8 .

Gorny’s tabiya for the 9... xe4 line. Black has many moves here. I
will give some ideas and lines.
15...h6
With the idea to drop the bishop back to h7 if needed.
A) 15...h5 is possibly less accurate because of 16. g5!, intending
16... g6 17.g4;
B) On the other hand, 15... d7 is entirely plausible and is Gorny’s
main line. The threat is now ... d7-a4. 16.b3 ( 16. f4? xc2!
17. xc2 a4+ and because the d1-rook is hanging, White’s needs
to go to d2, with unpleasant consequences).

172
 analysis diagram
B1) Gorny’s choice for Black is 16... e8, after which his main
line runs 17.h5 c5 18. xf6 xf6 19. c4. This is a very unclear
position which Gorny says is playable for White. I must say I’m not
thrilled by this entire line, which looks rather artificial and even
random to my human eye. Black has plenty of alternatives which
White will also need to be well prepared for. Is this why someone
likes to play the Center Game? Gorny’s line continues with 19...d4
20. de1 a8! and the situation remains highly unclear;
B2) 16... c6!? 17. d2 ( 17. d3 e8 ) 17... e6!. An annoying
move to face with White if you’re not prepared for it. Black’s idea is
... e7 with the double threat of ... e1+ and ... a3+. White’s best
move here is 18. b2 with the possible sequence 18... c8! 19. g5!
c5! 20.g4! cxd4 21.gxf5 e5, which requires further research but is
objectively fine for Black;
C) Also possible is 15... c8!? 16. f4! with interesting play;
D) 15... h5? is bad due to 16. e5;

173
 E) 15...c5!? 16. e5. Gorny gives 16... c6 and 16... d7 here, but
16... e6 may be Black’s best, as the natural 17.h5 e8! is suddenly
annoying for White, e.g. 18.h6 g6 19. e1 c4 20.f4 d7.
16. h2
Better than moving the queen to f4, although the reasons are not
exactly intuitive and are based on concrete lines. This adds to my
suspicion that this line is not so easy to play for White as one might
hope.
16. f4 g6 17.g4 c5! 18.h5 h7 19. e5 e6 favours Black clearly.
16...h5
Trying to block the kingside, but this won’t work.
A) 16... d7!? was also interesting, with a messy position that is
difficult to play with both colours, in my view, for example: 17.g4
h7 18. g1 c6!? ( 18... e7!? );
B) Perhaps the best try was 16... c6!? 17. c3 ( 17.c3 h7
favoured Black slightly in Knöppel-Schubert, LSS email 2022)
17...h5!? and I prefer Black.
17.g4
White’s best option. Clearly, White doesn’t mind giving up another
pawn if he can open up the kingside.
Too slow seems 17. f4 g6 18.g4?! c5! 19. e5 e6 and Black is
clearly better.
17...hxg4 18.h5

174
18... h7!
Very precise play, as one might expect in a correspondence game.
18...gxf3 19. f4 h7 20.h6 allows counterplay.
19. g1 e6!
Bringing some deep back-rank tricks into the position.
20.fxg4
Black’s point could be seen in the following line: 20. d3 xd3
21. xd3 g5+ 22. b1 g3! and now White can’t take on g3 due to
the mate on e1.
20... g5+ 21. b1 e4
Bad was 21... xg4? 22. h3 f5 23. xc7.
22. d3 e7 23.g5 xd3 24.cxd3!
24. xd3 c5 25. f2 xg5 26. g3 f6 27. xd5 e4 28. g2 xf2
29. xf2 e8 is still better for Black.

175
24...c5
Now a liquidation follows, leading to an equal position.
24... xg5 allows White to draw with 25. xg5! xg5 26. g1 f5
27. xg7+ f8 28. g3 f1+ 29. c2 c6+ 30. b3 d1+ 31. b4
and there is no mate for Black.
25. f2 xg5 26. xc5 xc5 27. xg5 ae8 28. g1 xg1
29. dxg1 e1+ 30. xe1 xe1+
White still has to work quite hard for the draw, but Pirs eventually
succeeded.
... ½-½ (45)

Winawer’s 8. c4 and alternatives in the Berger

Variation
In the next game, we will investigate the move 8. c4, sometimes
known as the Winawer Variation. It has never been popular, but the
engines have found many interesting new ideas in this old line to
offer White fair hopes of keeping things alive. Ian Nepomniachtchi

176
has recently played it as well. The fact that this line is objectively
equal makes it more suitable for a serious game, in my opinion. The
downside is that there is an annoying little move repetition if Black
is happy with a draw. Finally, it’s worth paying attention to the
move 6.a3, which leads to a drawn endgame, and the sideline
8. ge2!? which is a fun way to confuse Black.
The last game of this chapter is devoted to some of the quieter ways
for Black to play. These shouldn’t concern White players too much,
but still, they are worth knowing.
Rotom Monotua (2305)
Ilmars Cirulis (2239)
FICGS email 2020
In this game, we will look at Winawer’s Variation as well as some
minor lines.
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 f6 5. c3 b4

6. d2

177
If White wants, he can reach a drawish endgame with the more or
less forced sequence 6.a3 xc3+ 7. xc3 0-0 8.f3 d5 9. g5 dxe4 (
9...d4!? ) 10. xf6 xf6 11. xf6 gxf6 12.fxe4 e8 13.0-0-0 xe4
14. b5 d4 15. d3 f4 16. e2 xe2+ 17. xe2, Aymard-Wagner,
ICCF email 2018.
6. c4 0-0 7. d2? xc3 8. xc3 xe4 was bad for White in the
game Kupreichik-Lein, Voronezh 1969.
6...0-0 7.0-0-0
7. g3? is not to be recommended at this point: 7...d5! 8.0-0-0 (
8.exd5 e8+ 9. ge2 d4 10.0-0-0 xe2+ 11. xe2 e4 )
8... d6 and Black was clearly better in Aronian-Ding Liren,
Chess.com Speed Chess 2021 (this position was actually reached via
transposition from the 5... e7 variation, see Chapter 4).
7... e8
A) 7... e7 is sometimes seen. The idea is also to take the pawn
on e4, but with the queen instead of a rook. I faced it myself in a
league game once and didn’t really know how to respond smartly,
so it’s good to see a few ideas.

178
 analysis diagram
White’s best move is 8. ge2!, which has been played only three
times according to the database. The idea is not so much to develop
a piece, but to be able to bring the other knight to c3 (in my game, I
played in the spirit of the Tarrasch Gambit: 8. g3?! xe4 9. xe4
xe4 10.c3?! d6 11.f4 ( 11. h3 e6! clearly favours Black)
11... a4 12. d3 xa2 13. b1? ( 13. f3 was the lesser evil) and
here my opponent could have obtained a winning position with
13... a1! ( 13... c4?! 14. f3 Moll-M.de Wit, Belgium team
championships 1997/98) 14. f3 a5 ): 8... e8 ( 8...d6 9. f4 ).
White can safely play 9. g3 and now it’s a different ball game:
9... xe4 10. xe4 xe4 11. c3! g6 12. d3! xg3 13.hxg3 h6
14. b5 and White is doing very well indeed;
B) 7...a5!? is not bad at all, but there is almost no practice with it.
It’s an entire world to discover. I would probably go for something
like 8. ge2 e8 9. g3 xe4 ( 9... xe4? 10. g5! followed by d5)
10. xe4 xe4 11. c3 with a complex game ahead.

8. c4
179
This developing move, first played by Winawer, as we saw in
Chapter 1, is objectively White’s most healthy way to play in the
main line. It has been played numerous times by Muñoz Pantoja and
Luka Paichadze. I also think that for players who are new to the
Center Game, it is the easiest to learn against the Berger Variation.
There is, however, one drawback, as we will see.
Let’s check two other moves before we dive into the main line:
A) 8.f3?! d5! occurred in the early game Fleissig-Bird, Vienna
1882. Black has already freed himself and can look forward to an
easy game. White has to play accurately, or he will be much worse
before move 10;
B) 8. ge2 was suggested by Soszynski as ‘maybe objectively the
least worse [sic] of White’s eighth moves’. Well, it’s certainly good
fun! Black again strikes in the centre: 8...d5! 9. xd5 (all other
moves are worse) 9... xd5 10. f3! ( 10. g3?? xe4 Banchio-
Deforel, Argentina 2003), and now Black might scratch his head for
a few minutes before finding the best move:

analysis diagram

180
10... xd2+ ( 10... c5 11. e1! g5+ 12. d2 h4 ( 12... d8
13. e1= ) 13.exd5 e5 14. f4 xf2 15. xf2 xf2 – Soszynski;
10... e5!? is also possible, but slightly more complex: 11. g3 f6!
12. g5 b6! (Black has to spot this accurate move, otherwise there
could be trouble) 13.exd5 c5 14.h3 xf2 15. b3 (Soszynski) and
now Black’s best option seems to be 15... d7!? ) 11. xd2 cb4
12.exd5 xa2+ 13. b1 b4 14. d1. Material is equal, but Black
seems to have the better chances.
8...d6
The most logical move.
A) 8... xc3 9. xc3 xe4 10. f4! transposes to the 8. f4 line
where Black takes on c3, which we saw in the game Granda Zuniga-
Fernandez Garcia, Pamplona 1992;
B) After 8... e5!? 9. b3 a5 the position is very interesting –
almost never tested in practice, and absolutely wild! White should
simply develop his knight: 10. f3! eg4 11. f4! d6! (
11... xf2? fails to 12. g5 ) with crazy complications, which
require further study;
C) The downside for White is that Black can force a repetition
here if he wants to: 8... a5 and now:

181
 analysis diagram
C1) White can avoid the repetition with 9. d3, but this allows the
classic break 9...d5! 10. f3, which happened in Firouzja-Tang,
Chess.com bullet 2021, and now Black could have played 10... c6!
11. g5 h6 12. h4 dxe4 and end up being better, e.g. 13. xe4 xe4
14. xd8 xd8 15. xe4 xe4 16. xb4 g4 ;
C2) 9. f1 c6 at least allows White to choose another line on
move eight (now ten);
C3) 9. e2 and now:
C31) 9... c6. Again threatening ...d7-d5, which leaves White
nothing better than 10. c4 ( 10. f3 e5 ) 10... a5 etc.;
C32) 9...d5!? is possible too, but the bishop on b4 is unprotected,
and this gives White additional tactical resources: 10. xd5! xd5
11. d3! (ingenious play; now, if the knight on d5 moves, the bishop
on b4 hangs, while after taking on d2, the knight on d5 is pinned!
11... f6! 12. xd5 ( 12.exd5 f5 Jacobsen-Urzica, Groningen 1969)
12... e6 and things can still go either way.

182
 D) 8...a6!? is mentioned by Ziegler as a good way to experiment
with Black, and he is right. There are almost no games with it, but it
looks playable. One way to play for White might be 9. h3!? d6 (
9...b5 10. d5! ) 10. g5 e5 11. b3 h6 12.f4 hxg5 13.fxe5 xe5
14.h4 with unclear complications;
E) 8... e7 9. d5! xd5 10. xd5, with a small edge for White,
occurred in a game by another author of a monograph on the Center
Game, Riccardo del Dotto: Del Dotto-Luciani, Imperia 2006.

9. f3
The move g1-f3 (before f2-f4) has never been popular in the
Center Game, as White usually wants to push his f-pawn to either f3
or f4. In this position, however, White wants to play g5 and attack
the weak f7-pawn.
A) The move 9. h3!? introduces the Kupreichik Variation, named
after the Belarussian grandmaster Viktor Kupreichik. It might be a
good surprise weapon, but it leads to positions that are objectively
good for Black. One of the downsides is that Black can simplify the
position in a forcing and favourable way: 9... xh3 ( 9... e5 10. b3

183
occurred in Kupreichik-Estrin, Leningrad 1965, and now Black
should have played 10...a5! ( 10... e6?! 11.f4 c4 12. xc4 xc4
13.e5!? led to complications which Kupreichik managed to shape
into a killing attack later on); the engine’s favourite move is, once
again, 9...a5!? ) 10. xh3 xc3 11. xc3 xe4 12. f5 g5+
13. xg5 xg5 and White doesn’t have full compensation;
B) 9. ge2?! c5 10. g3 e5 and the white queen will feel
rather uncomfortable after Black plays ... f6-h5;
C) We have seen the slow-ish 9.f3?! in the notes to the game
Tartakower-Reshevsky, Stockholm 1937, in Chapter 1. After
9... e5 ( 9... e6?! 10. xe6 fxe6 11. ge2 Paichadze-Le Tuan
Minh, Titled Tuesday blitz 2024) 10. b3 a5! 11.a3 c5 12. e1 b5
Black was doing well in Muñoz Pantoja-Vallejo Pons, Linares rapid
2016.
9... e6
A) 9... c5!? is rare, but playable. White unsually continues
10. f4 ( 10. d3!? ), e.g. 10... e6 ( 10... e7! 11. g3!? ) 11. xe6
xe6 (Paichadze-E.Hansen, Titled Tuesday blitz 2016) 12. g5! ;
B) 9... g4?! looks kind of sensible, but not really: 10. f4! ce5.
This happened in Solovjov-Matlakov, St Petersburg 2007, and now
11. xe5! xe5 12. e2 would have been slightly better for White;
C) 9... xc3 leads to sharp and pointed play: 10. xc3 xe4
11. he1 f5 12. f4

184
 analysis diagram
12... g6 ( 12... d7? 13. d3! d5 and now the only winning move is
the subtle 14.g4! g6 15. xe4! xe4 16. xe4 xe4 17. xe4 dxe4
18. xd7 exf3 and the reason for including g2-g4 becomes clear:
White can now play 19. xc7 without being bothered by ...fxg2!, as
indicated by Solovjov in Chess Informant 100) 13. xf7+! xf7
14. xe4 xe4 15. xe4 ;
D) 9... g4!? is best answered by 10.a3 when ...a5, ... c5 and
... a5 all seem playable. This line requires further research;
E) 9...h6!? was seen in a recent game Nepomniachtchi-
Narayanan, Astana rapid 2024, and after 10.a3 xc3 11. xc3 xe4
White should have played 12. he1! f5 13. f4 with a complex
fight ahead after 13... d7 14. xg7! followed by d3 and h4.
White will win back the piece with xe4 and f2-f3, and the resulting
position is balanced.
10. xe6 xe6

185
10...fxe6 was played only once in more than 40 games, but not by
the worst of players: Vasily Panov, the Russian master who is
mainly known today for his association with the 3. exd5 cxd5 4. c4
line of the Caro Kann Defence. In this game, Gusev-Panov, Moscow
1947, White continued with 11.h3!? ( 11. b1 ) 11... d7 12. e2
f6 13. b1 .

11. g5
The only move that makes any sense.
11... e8
A) 11... e7!? is not that silly. It was played only once, but White
doesn’t have that much of an advantage after 12.a3 ( 12.h4!? is a
way to spice things up) 12... a5 13.f4 e8! when the engines give
0.00;
B) 11... e5!? is Bologan’s suggestion. Black has the move ... e8
at his disposal in some lines: 12.f4! e8 13.a3! (an improvement
over Bologan’s analysis; 13. f3 h6 14.h4 d7= is Bologan’s line,
which continues 15. d5? xd5 16.exd5 d4 when Black indeed
has a great game) 13...a5 ( 13... a5 14.h4 h6 15. d3 ) 14. d3! h6
186
15.h4 d7 16.g4! and White has a mean initiative, certainly in a
practical game. Hardcore analysts will have a field day here, I’m
sure.

This key position has occurred many times since its inception in the
game Winawer-Janowski, Vienna 1896, but White’s next move has
been seen only five times:
12.a3!
A) 12.f4 h6 13.h4 was played in the two oldest games in this line,
and now 13... c8! was Reshevsky’s strong idea in 1937 against
Tartakower (see Chapter 1). After 14. f3 he should have played
14... d4! with a clear advantage;
B) 12.f3?! was tried once by Muñoz Pantoja in 2017. Black is
fine after 12... e5 13. e2 h6! 14. h3 c6;
C) 12. g3?! h6 13.h4 d7! .
12... a5
A) 12... c5!? would lead to a very interesting piece sacrifice for
White: 13. g3 h6 14. he1 (surprising centralization rather than
trying to open the h-file) 14...hxg5 15. xg5. Next up is h2-h4 and
187
White will have a fantastic grip on the position. Unfortunately, there
is a repetition with 15... h5 16. g4 f6 17. g3 h5;
B) 12... xc3 13. xc3 h6 14.h4 is exciting and completely
unclear:
B1) 14... e7 15. g3 a5 16.f4 b5 (Nepomniachtchi-Caruana,
Zagreb blitz 2024, continued 16... ad8 17. b1 (better is 17.b3!,
with the idea 17...b5 18. b2 ) 17...b5 18. xf6 (compared to 16... b5
17. e5!, here 18.e5 fails to 18...dxe5 19.fxe5 d5! ) 18... xf6
19. d5 d4 20. d3 c5 21.e5 g6 and now the position is level
after 22. xg6 ) 17.e5! dxe5 18.fxe5 xe5 19. he1 fg4 20. xf7!
and the computer says 0.00;
B2) 14...hxg5?, as played in Nepomniachtchi-Bok, Champions
Chess Tour CrunchLabs Masters rapid 2024 (three days after
Nepomniachtchi-Caruana), was a big mistake. The game continued
with 15.hxg5 xe4 16. h3, and here Bok had possibly missed that
16... xg5+ could be met by 17. d2!, preventing the defence of
17... h6. After 16...f6 17.g6 f8 18. h8+, White secured a quick
win.
13. g3
Probably the most accurate move.
13.h4!? e5 14. e1!? is worth looking into too. White has ideas
like f2-f4 or c3-d5. The position is extremely sharp and complex.
13...h6
A) Unexplored but also interesting is 13... h5 14. h4 f6
15. he1 h6 16. f3 with a relatively quiet middlegame;
B) 13... e5 was recently tried, but White can be happy after
14.f4! c4 (Remizov-Klimentov, Russia tt 2024) and now 15. e1!
with an edge.

188
14.h4!
As natural as a baby’s smile, but also the only move. 14. h3?
xe4! .
14... b8!?
A typical computer move, if you ask me.
A) 14...hxg5? 15.hxg5 just wins for White;
B) Black’s best (and much more human) move is probably
14... e5 15.f4 eg4 and now the following line seems critical for
the evaluation of the Winawer Variation: 16. d5 xd2+ (or
16... xd5 with some sharp tactics to follow: 17. xa5 de3 18. d3
d7! 19. xe3 xe3 20. xe3 hxg5 21.hxg5 f5! 22. f3! xa5 (
22... xe4 23. h5 ) and now White has just enough counterplay on
the h-file: 23. h3! (not 23. h5? f8! ) 23... f8 24. d7! – the
point: this forces a repetition) 17. xd2 xd5 18. xg4! ( 18. xd5
c8! ) 18... f6 19. f3. There is still plenty of room for
investigation in this line, but with correct play the game is likely to
end in a draw somewhere. Readers can switch on their own engines
at this point;
189
 C) The immediate 14...b5!? is another idea. Again, Black should
often go for activity himself. The position is unclear after 15. xb5
xd2+ 16. xd2 b8 17. c3.
15. b1! xc3
Going for the knight on g5 which, after all, was still hanging.
15...hxg5? 16.hxg5 xc3 17. h4 .
16. xc3

16...hxg5 17. xf6!

Not 17.hxg5? xe4 18. h4 xc3+ 19.bxc3 f8 and White doesn’t
have enough.
17... xf6 18.hxg5 g6 19.f4!
Just supporting the pawn on g5, and never mind that Black can take
on e4.
19... xe4 20. h3 f8 21. d7 e7
21... g8 22. h3 is a draw.
22. xc7

190
White picks up enough pawns to compensate for the piece deficit.
22... e8
After 22... c8 White can win prosaically with 23. h8+, but 23. xd6
xc7 24. d8# is certainly an elegant mate.
23. xd6 c4 24. xc4 xd6 25.f5
Now this high-quality game ended in a repetition:
25... c6 26. h4 e7
Draw agreed.
Ivan Morovic Fernandez (2535)
Jose Garcia Padron (2445)
Las Palmas 1991
In the last game of this chapter, we will look at positions where
Black doesn’t put his rook on e8 but instead tries to develop his
queenside as quickly as possible. These are playable, but not very
ambitious. Still, it’s useful to play through the variations, as you will
likely face these kind of positions with White on many occasions at
your local club or in open tournaments.
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 f6 5. c3 b4 6. d2 0-0
7.0-0-0 d6

191
This move is often played, but White can get his standard set-up
rather easily now.
8. g3
This is almost always played, and that is understandable, but there
are alternatives:
A) 8.f3 was already played in the year 1889 by Minckwitz, and is
possibly even more accurate, going for the direct g2-g4 push:
8... e5 ( 8...a5 9. f2!? ) 9.h4! and I like White’s chances;
B) 8. e1!? is entirely new but looks decent enough. White plans
to follow up with the usual f2-f3/g2-g4 business.
8... e5
8... e8 9.f3 transposes to Shirov-Karpov, Dos Hermanas 1995 from
Chapter 1.
9.f3 c6 10. ge2
The direct 10.h4 h8 11.h5 h6 12. h4 b5 13.g4 is also pleasant for
White.

192
This is a position that has only one entry in the database, but it can
often been seen in amateur and blitz games, sometimes in a slightly
different permutation. White already has a rather promising attack,
while Black’s counterplay is rather slow.
10...b5?!
Black doesn’t have time for this push.
A) Also too naive is 10... c4? 11. g5, clearly favouring White;
B) A better try is 10... h5 11. e1 (the queen retreat to e1 is
something to remember for White) 11... c4 and Black at least gets
the white dark-squared bishop, though White is still slightly better
after 12. d4 xd2 13. xd2;
C) It was still best for Black to activate his rook: 10... e8 when
White only has a small plus.
11. g5!
This is an important move to remember in these slower lines. White
is going to push f3-f4, after which Black’s knight on e5 can’t be
maintained in the centre.

193
11... e7?!
Tougher was 11... d7 12.f4 c4 13. h4, though White is clearly
on top.
12.f4 g6 13.f5 e5 14. f4

White’s position is a dream come true whilst Black doesn’t seem to

have done all that much wrong. This is one of the reasons why the
Center Game is a great practical opening choice at club level.
14...a6
Another slow move.
14...a5 was a better try, though White could then also play
15. cd5!.
15.a3
White’s position is so good that he could already strike directly with
the thematic 15. cd5! cxd5 16. xd5 xd5 17. xe7 xe7 18.f6.
15... a5

194
16. h5?
Careless. The ensuing endgame is good for White, but why go there
at all?
16. cd5! was even stronger here than on the previous move, since
in the end the pawn on d6 will be hanging as well.
16... xh5 17. xe7 xg3 18.hxg3 e8 19. xd6

195
19... xc3?
Black misses his chance. The Dvoretskian ‘minor operation’
19... g4! 20. c5 f6! would still have been better for White, but
Black isn’t lost yet. As it is, White had a big advantage after
20.bxc3, and he won on move 45.

196
Chapter 4
Paulsen Variation
4. e3: lines with
... e7 and 5. d2
subtleties
On the page it looked nothing. The beginning simple, almost
comical.
Salieri in the movie Amadeus

We’ll now turn to lines in which Black tries a deceptively simple

set-up in the Berger Variation: after ... f6 he develops the bishop to
e7.

197
 position after 5... e7
Preparing a quick ...d7-d5. This dry but effective set-up has always
been a source of annoyance after 4. e3. Whereas the old main lines
(even if objectively dubious) give White plenty of room for
interesting and lively play, after ... e7 White basically wishes he
hadn’t moved his queen to e3. If he doesn’t stop ...d5, Black gets
easy play, as the next two games show.
Marek Soszynski (1870)
Nikolay Timkin (1821)
FICGS email 2008
In this game we will see how play may develop if White doesn’t
stop Black from playing the central break ...d7-d5 after both sides
have castled. It’s a nice coincidence (or is it?) that the player with
the white pieces is the author of a recent monograph on the Center
Game.
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 f6

198
4... e7 5. c3 f6 transposes ( 5...d6 will be examined in Chapter 5
under 4...d6).
5. d2
The move 5. d2 is Soszynski’s top choice in his book The Center
Game Re-examined (2020). The author admits, however, that it’s
‘not so lively’.
A) We have seen 5.e5?! in the note to the game W.Paulsen-
Berger, Berlin 1881, in Chapter 1. This weak move is sometimes
called the ‘Paulsen Attack’, but I think this doesn’t do him justice.
Black just needs to know one thing: develop everything with tempo!

analysis diagram
5... g4 6. e4 ( 6. e2 d6! 7.exd6+ e6 and Black has a fantastic
position) 6... gxe5 7.f4 d5 8. e2 ( 8. e3 d4 9. e2 b4+
transposes) 8... b4+! 9.c3 g4 (this is such fast development it
would make Usain Bolt proud) 10. f3 0-0! ;
B) This is a good moment to mention that after 5. c3 Black can
try many moves (there are 20 different ones in the database), most

199
of which transpose to lines we have seen or will be seeing in other
chapters. I once faced 5... e7 (White’s response is the same after
5... g4?! 6. e2 d6 and now White simply develops according to
the principles of this opening, e.g. 7.f3 f6 8. e3 e6 9.0-0-0 )
6. d2 c5 in a blitz game, and didn’t respond correctly to it, so it’s
good to see how White should play: 7. e2! ( 7. g3?! is what I did,
but after 7...d5! Black had already equalized) 7... e7 ( 7... d4
8. d3 ) 8. f3! 0-0 9.0-0-0 with advantage.

5... e7
We will look at 5... b4!? later in this chapter. I recommend either
the transposion with 6. c3 or the interesting 6.a3!?.
5... g4!? was already played by Frank Marshall and is not bad
either: 6. e2 ( 6. g3 d5= Morris-Marshall, New York 1911)
6... d4 7. xg4 xc2+ 8. d1 xa1 9. c3 with a murky position.
6. c3
6. c4? is simply dubious here: 6...d5! 7.exd5 xd5 when 8. g3
fails to 8... d4!.

200
6.c4 is a main move and transposes to Game 34, Muñoz Pantoja-
Escandell Mari.
6...0-0
The big alternative is the immediate 6...d5 7.exd5 xd5, which we
will examine in the next game.
7.0-0-0 d5
The only consistent move.
White doesn’t need to be afraid of 7... e8 8. c4 e5 9. b3 .
The consequences of 7...d6 will be examined in Chapter 5, when we
look at the move 4...d6.

8.exd5
8. g3? was played by Levon Aronian against Ding Liren, but
without including the capture on d5 it is actually a blunder: 8...d4!
and it’s unclear what Aronian had in mind here. Black is close to
winning. Ding went for 8... d6, which was also good: 9. h4 xe4
10. xd8 xd8 11. xe4 ( 11. xd5 xf2 ) 11...dxe4 and Black

201
was a pawn up for nothing, Aronian-Ding Liren, Chess.com Speed
Chess 2021.
8. xd5 transposes.
8... xd5
I’m not impressed by 8... b4 9. c4 e8 and now easiest is 10.a3!?
bxd5 11. xd5 xd5 12. b3 e6 13. e2 .
9. f3!?
I like this move, and I think it is probably White’s best way to try
and make something out of nothing.
A) 9. g3?! h4! was analysed already in Chapter 1 (Mieses-
Alekhine, The Hague 1913);
B) 9. xd5 xd5 10. b4 ( 10.c4? c5 MacKenzie-Lipschütz,
New York 1886) is a sharp way to reach an endgame, but Black has
nothing to fear here: 10... e6 ( 10... g5! has never been played but
is the engine’s top choice: 11. xd5 ( 11.f4? xa2 ) 11... xe3+
12.fxe3 xb4 13. b5 a5 and although it’s objectively 0.00, this
feels simpler for Black) 11. xe6 xe6 12. xe7 xe7 with equality,
Skrenek-Kulovana, Moravia 2002.

202
9... e6
9... xc3 is possible, but it requires some precision: 10. xc3 d6 (
10... g5+ 11. b1 e7 and now the new move 12. h3! leads to a
good position for White) 11. h5 and now the position is actually
quite tricky. Black’s best is 11... d7! ( 11... e8 12. c4
Kharitonov-Bryant, IECG email 2003), intending ... g4, as in the
following line: 12. d3 ( 12. xg7! xg7 13. g5+ forces a draw but
nothing more) 12...h6 13. f3 g4! 14. xg4 xg4 15. e4 fe8
with equality.
9... cb4 10. c4 b6 11. e2 is entirely new and worth exploring.
10. ge2
10. f4?! xc3! was analysed in Chapter 1 (Mieses-Alekhine, The
Hague 1913). After 11. xc3 b4 12. c4 e8 Black is better.
10... xc3
A bit loose is 10... e5 11. g3 g6 12. d4 .
11. xc3
Worth a try is 11. xc3 d4 12. g3 c5 ( 12... h4 13. f4 c5 is
unclear) 13. d3 when Black has many options, and so does White,
e.g. 13... d6 14.f4!? e8 15. hf1 b6 16. de1 and the position is
equal but full of life.
11... d6
Not 11... c8 12. g3 .

203
12. b1
The idea is not only to protect a2 but also to prevent Black from
playing ... g5+ with tempo. Still, it feels a bit timid.
More in the spirit of the opening is 12.h4!? with an interesting
position, which I rather like for White despite the familiar 0.00
engine evaluation. Play might continue 12...a5!? 13. b1 ( 13.h5 h6
) 13... c8! ( 13... b4?! 14.a3 e8!? 15. g3! ) 14.h5 ( 14.g4!? is
also possible, intending 14... xg4? 15. g2 followed by h4-h5
and/or g1) 14...h6 15. h4! with typical Center Game play.
12... g5!
Very alert. Black aims to simplify with ... g4.
After the slow 12... e8 White has 13. f4 with an edge.
13. g3
Soszynski is still trying to make something out of it!
13.h4 g4 equalizes on the spot.
13... g4?!

204
Better, but a bit scary-looking, was 13... b4! 14. xb4 xb4 15. d3
g6 16. e4 ( 16.h4 d5 ) 16... a5 17.a3 e7 18.h4 h5 and Black
has nothing to be afraid of, though it looks complicated in practical
play.
14. d5! g6?!
It was better to play 14... f4! 15. xf4 xf4, though White is
slightly better after the accurate 16. c5! with the idea f5.
15. e3
White actually has something tangible here. Soszynski expands his
advantage nicely in the following phase.
15... fe8 16. d2 e5?
A strange move, especially for a correspondence game.
16...f6 17.f4 .
17. d3 xc3
Or 17... h6 18. xh6 gxh6 19. d2 and White has a huge advantage.
18. xc3

205
18... f6
Sad, but what else?
If 18... e6, then 19. g5 g6 20.h3 traps the bishop.
18... h6 loses to 19.h3 e7 20. c5 d7 21. h5.
19. xf6 gxf6 20.f3 e6 21. b5 ab8 22. h5
Winning a pawn and the game. The rest requires no comment. Black
resigned on move 39.
Vladislav Artemiev (2704)
Vladimir Fedoseev (2706)
Poikovsky 2018
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 f6 5. d2 e7 6. c3 d5
Why not? Black immediately frees himself.
7.exd5 xd5

This position occurred for the first time in the game Chigorin-
Znosko-Borovsky, St Petersburg 1906. It was also played by Akiba
Rubinstein and Frank Marshall in the first decades of the 20th
century. In general, Black achieves equality pretty comfortably,
206
which is one of the reasons this whole complex isn’t played much
with White anymore.
8. xd5
This doesn’t give White anything.
A) The stereotypical and tempting 8. g3? was awarded with an
exclamation mark by Anand in Chess Informant 61, but it is actually
bad on account of 8... cb4!, which frustrates White’s plans: 9. xd5
( 9. xg7? f6; of course, not 9.0-0-0? xc3 10.bxc3 xa2+
11. b1 ( 11. b2 xc3! is even worse: 12. xc3 f6 ) 11... e6 and
Black wins) 9... xd5 10. xb4 e4+ . The point. Black has the
bishop pair for free in an open position;
B) 8. f3 xc3! 9. xc3 0-0 10. d1 e8 is good for Black, e.g.
11. e2 d6 12. h3 e5;
C) White’s most promising option to take the game into somewhat
unsual territory is perhaps 8. e4!?

analysis diagram

207
with the idea 8... f6 ( 8... e6 9.0-0-0 0-0 10. f3 ) 9. a4 with an
unclear position in another correspondence game by Soszynski.
After 9... d7 10.0-0-0 0-0 11. f3 anything is still possible,
Soszynski-Clement, FICGS email 2010.
8... xd5 9. e2
The knight is headed to c3, where it protects a2. Immediate
queenside castling was not possible on account of the capture on a2.
A) 9. f3? is bad because of 9... f5 and Black is clearly better;
B) After 9. d3, easiest for Black is the new move 9... e5! 10. e4
e6 .

9...0-0
A) Also possible is 9... g4 10. f4 ( 10.h3= ) 10... d7 11.f3 0-0-0
12.0-0-0 f5! and again the position is equal ( 12... he8? is,
however, too optimistic: 13.fxg4 b4 14. f2 c5 15. f3 e3
16. d5! and White won in Sultan Khan-Marshall, Liege 1930);
B) Also sufficient is 9... f5 10. f4 ( 10. c3 d4= ) 10... d4 (
10... d7!? ) with equality.

208
10. f4
10. c3 was first played by Tartakower against Rellstab in 1937, but
is slightly less accurate. Black should prevent White castling:
10... h5! ( 10... c5 11.0-0-0 xe3 12. xe3 Tartakower-Rellstab,
Kemeri 1937; 10... a5 was once played by Dutch grandmaster Erik
van den Doel) 11. e2 g6 and White can’t castle queenside: 12.0-
0-0? b4 gives Black a clear advantage.
10... c5
A logical move. Black has equalized.
11.0-0-0 f5!?
Perhaps easier was 11... xe3 12. xe3 d6 with equality, as
happened in the game Stankovic-Atalik, Paracin 2022, which Black
actually won.
12. xc5
12. d3 xd3 13. xc5 xc5 14. xd3 transposes.
12... xc5 13. d3 xd3 14. xd3 b6 15. he1

209
This is a common situation for those lines in which Black manages
to play an early ...d5. The position is perfectly symmetrical and the
two 2700 opponents played it out to a draw. Of course, a mere
mortal would probably lose this with either colour against these
world-class players.
½-½ (60)

The Muñoz Variation: c2-c4

White can prevent ...d5 by playing 5. d2 first and then going for a
c2-c4 set-up in some shape or form.

position after 5. d2 e7 6.c4

This is an old favourite of Muñoz Pantoja and it has recently also
been played by Erigaisi. The engine is not a fan and, as Erigaisi
himself admits, Black actually has a plus after this move. However,
Center Game players have always been pragmatical, and this line
may work well in a practical game.
Miguel Angel Muñoz Pantoja (2366)
210
Juan Carlos Escandell Mari (2161)
Cornella de Llobregat 2022
In this game we see Muñoz play his beloved c2-c4 system against
the ... e7 set-up, but Black comes well prepared and reaches a
winning position. Muñoz’s fighting chess ultimately triumphs, but
only after some incredible adventures.
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 f6 5. d2
Preparing c2-c4 after ... e7.
The immediate 5.c4?! was played twice by the American FM Todd
Andrews, but it fails concretely: 5... b4+ 6. d2 0-0 7. c3
(Andrews-Hamitevici, Titled Tuesday blitz 2023) and now the
simple refutation is 7... xc3! 8. xc3 xe4 and Black wins.
5... e7
A) A more popular reply is 5... b4,

analysis diagram
hoping to get into the classical main lines after 6. c3. If White
doesn’t want to risk that, he can also go for a set-up with c2-c4,
211
trying to transpose to the game. The best way to do so is 6.a3!. This
is more accurate than 6.c3 which, for some reason, is more popular
at the highest level:
A1) 6.c3?! has been played many times by Muñoz, and recently
also by Erigaisi, but is actually inaccurate in view of 6... e7!. This
key move, still aiming for ....d7-d5, more or less forces White to
play 7.c4, leading to the game.
6... a5?! has been played by many strong players, but it doesn’t
make much sense if you know that White wants to play c4 anyway
and exchanging bishops is rather beneficial for him: 7.c4 ( 7. g3?
e7 was simply bad for White in Erigaisi-R.Svane, Dusseldorf
rapid 2023) 7... xd2+ 8. xd2 0-0 9. e2 (Erigaisi’s latest try; 9.0-
0-0 e8! 10.f3?! ( 10. e2 d5 11.cxd5 xd5 12. g3 b6= ) was
tried three times by Muñoz, but none of his opponents played the
obvious 10...d5! and Black is fine; for example, 11.cxd5 xd5
12. d3 f4 ) 9... e8 ( 9...d6!? 10.f3 b4 11. c3 c5 is unclear)
10. c3 d5 11. xd5 xd5 12.cxd5 b4 13. b3 xd5 14. c4 c6
15.0-0 b5 16. xd5 cxd5 17.exd5 b7= Erigaisi-Martirosyan,
Jermuk 2024;
A2) 6.a3 xd2+ ( 6... e7 7. c3 ( 7.c4!? ) 7...d5 8.exd5 xd5
9. g3! leads to a position we have seen in the previous game, but
with a pawn on a3, which prevents the move ... cb4, so this is
slightly better for White) 7. xd2 0-0 and now:

212
 analysis diagram
A21) If White wants to play the c2-c4 system, it is possible to do
so here: 8.c4!? when the pawn on a3 is actually useful since it
prevents the annoying ... c6-b4 in many lines. There are no
practical examples here, but a sample line is 8... e8 9. e2 d5
10.cxd5 xd5 11. g3 f6 (the adventurous 11... db4!? is also
possible: 12.axb4 xb4 13. c1 d3+ 14. d1 xc1 15. xc1 a5
with complicated play) 12. c3 d4 13.0-0-0 , and although the
position is roughly equal after, say, 13... e7 this is a better try for
White than the 6.c3 lines;
A22) 8.0-0-0 is the most popular move: 8...d5 9.exd5 ( 9. gf3!?;
9.f4!?; 9. e2!? ) 9... xd5 ( 9... xd5?! 10. g3 ) 10. e2 e5 (
10... g4!? ) 11. xe5 xe5= Anton Guijarro-So, Carlsen
Invitational rapid prel 2021.
B) 5... g4!? was played already by Frank Marshall and is not bad
either: 6. e2 ( 6. g3 d5= Morris-Marshall, New York 1911)
6... d4 7. xg4 xc2+ 8. d1 xa1 9. c3 with a murky position.
6.c4?!
213
6. c4? is simply dubious here: 6...d5! 7.exd5 xd5 when 8. g3?
fails to 8... d4!.
6. c3 transposes to either of the previous two games.

White’s Maroczy-like set-up was first seen back in 1900 in St

Petersburg in a game between Vasily Tabushnikov and Mihail
Shabelsky. It requires a certain braveness and precision to play this,
so, from a practical point of view, White may gamble (or bluff, as
Nepomniachtchi might say) and give it a try.
6...0-0
6... g4!? 7. e2 d6 8. c3 was played in Erigaisi-Praggnanandhaa,
Baku rapid 2023, and now Black should have gone for 8... d4!
9. d3 c5! with an unclear position, which isn’t bad at all for Black.
7. c3 b4!
Impressive! Juan Carlos Escandell Mari, a much lower rated player,
goes for the optimal line against Muñoz’s own system. It is well
possible that he had prepared this at home.
A) More popular and also good is 7... e8 8.0-0-0 b4 Muñoz
Pantoja-Vasquez Schroeder, Barcelona 2017;
214
 B) After 7... g4!? White’s best move is the new 8. f4 ( 8. e2
d5! 9.cxd5 d4 10. d3 c5 Li Chao-Tan Zhongyi, China team
championships 2021), e.g. 8...d6 9.h3 ge5 10. e3 with complex
play;
C) 7...d6 is possible, but as usual a bit passive: 8.0-0-0 g4
9. e1! followed by h3 and g4 slightly favours White, Muñoz
Pantoja-Soler Lopez, Barcelona 2016.
8.0-0-0 c6! 9.a3
9.e5 g4 10. e1 d6 is good for Black.
9...a5!
Completely in accordance with the rules of modern engine chess.
Black is already better.
9... a6?! 10. d3 is less clear.

10.f3?!
A) Taking the knight would have been dangerous and bad:
10.axb4 axb4 11. b1 d5 and the engine already gives Black a +2
position. White is underdeveloped and his king is simply too weak;

215
 B) The best continuation was 10. g3 d5 11. f3 and now Black
has the nice move 11... a2+! 12. xa2 xe4 with a crushing attack,
e.g. 13. f4 d6 14. e3 c5 15. d4 f6 16. e1 e8.
10...d5 11. ge2
Best was 11. e1, after which Black continues in style: 11... g4!
12. d2 ( 12.fxg4? g5 loses the queen) 12...d4! 13. xd4 xd4
14. xd4 e3! 15. ge2 bc2!, winning the exchange.
11... c5
Black is winning now.
12. g5
Or 12. d4 dxe4 13. xc6 xe3 14. xd8 d3+ 15. xd3 xd2+
16. xd2 exd3 and Black wins, as the knight on d8 is trapped.
12... d3+ 13. c2 f2 14. g3 xd1 15. xd1 e7?!
After a great game, Black now starts to falter.
15...h6 16. f4 dxe4 17. xe4 xe4 18.fxe4 e8 wins for Black.
16. f4 e6 17. e3 dxc4 18. ef5

216
18... e8?
Black was still winning after 18... d7! 19. e2 f6.
19. xg7!
Once again, g7 is where the action is.
19... xg7 20. c3
Now White has dangerous compensation for the rook.

20...h6?
Black finally collapses.
20... g8! was the only move, after which Black is still better,
though it isn’t easy.
21. xc4! g6?
A) 21... xc4 22. f5+ g8 23. xh6 d3+ 24. c1 and White
checkmates on g7;
B) The only move was, again, 21... g8, but White is better now
after 22. d1.
22. d1 c8

217
22... b6 23. f5 and White wins.
23. d6!?
Or 23. xe6 fxe6 24.e5 and wins.
23... d8

24. f5?
The position is so complex for a human that Muñoz also fails to
deliver the final blow.
Correct was 24. h5! xh5 25. xe6 fxe6 26. xf6 and White wins
in all lines, the most beautiful being 26... xd6 27. g4#.
24... xf5 25.exf5+ h7 26. xf6 xf6 27. xf6 d7 28. b1 a4
29. a1?! d1+ 30. a2 d7 31. e4 d6?
31...b5! forces the exchange of queens by 32... d5+ and wins.
32. e5 c5?
32... d2 33.f6+ h8 would still have won.
33.f6+ h8 34. f4 h7 35. e4+ h8 36. e6!?

218
A brave winning attempt from the fighting player Muñoz, even
though it’s objectively losing.
36... d5?
36... d1 37. f4 c1! was possible and winning.
37. f4 f8 38. xd5 cxd5 39. d6 g8 40. xh6+ h7 41. f4
c8?
41... c2 draws.
42. xa4 f5 43. h4+ g8 44.g4 g6 45.f4 c4 46.h3 c2??
46... e4! with the point 47.f5? xf5 would still have equalized.
47. g5+
47. h6.
47... g6 48. xd5
Now Muñoz finishes the game for good.
48... c2 49. e5 b6 50. d8+ h7 51. d7 c5 52.h4 g8 53.h5
c8 54.g5 e8 55.g6+ g8 56. xe8+ xe8 57. b3 1-0
An epic Center Game battle!

White plays f1-c4

The other way for White to prevent Black’s central break is by
means of an early f1-c4 – ‘White’s only realistic try for an
advantage’, according to Greet.

219
 position after 5. c3 e7 6. c4
This seemed a great idea for a while, until it was discovered that
Black obtains active and dangerous chances himself with a well-
timed ... f6-g4!. Even if White is well prepared in this line, the
positions are objectively not better for him, and it seems to me that
he risks more than Black does.
Viktor Varavin (2449)
Alexander Tjurin (2348)
Voronezh 2001
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 f6

220
5. d2
Note the move-order here.
If White wants to stop ...d5 by means of c4, he should go 5. c3
e7 6. c4 here, as in Shabalov-Shliperman, Newark 1995 from
Chapter 1, which transposes to the game after 6...0-0 7. d2.
5... e7 6. c3 0-0
This allows White’s next move.
Black could have played 6...d5 here right away.
7. c4 g4!
A) It was also possible to play 7... e5!?, which leads to great
complications after 8. b3 c6 ( 8...d5!? is new and interesting) when
the natural 9.0-0-0?! (better is 9. f3 d6 10. xe5 dxe5 11.h3; 9.h3!?
) allows Black to grab the initiative in the usual bold way with
9...a5! ;
B) The standard move is 7...d6. Though the position is objectively
equal, White should be relatively happy here: 8.0-0-0 ( 8. f3
transposes to the next game) 8... e5 ( 8... e6 is more common, but

221
is better for White after 9. xe6 fxe6 10.f4 e5 (Külaots-Trent,
Gibraltar 2013) and now White should have played 11.f5! d5
12.exd5 xd5 13. f3! ) 9. b3 c6 ( 9... e6 10.f4 – see Shabalov-
Shliperman, Newark 1995; 9...a5! is a new move, naturally
suggested by the silicon monster. The situation is tense after 10.a4
ed7! 11. ge2!? c5 12. a2 ) 10.f4 ed7 and now I like 11.g4!
with the idea that 11... xg4 12. g3 is very dangerous for Black.

8. e2!
The best square for the queen.
8. g3? fails to 8... ce5 followed by ... h4.
After 8. f4?! Black has the beautiful 8...d5!! ( 8... ce5 9. e2 was
seen in Perez Rodriguez-Bruzon Batista, Merida 2007, and even
here Black could have played 9...d5!, intending the nasty ... d6)
9. xd5 ( 9.exd5 d6 10. g5 e8+! 11. ge2 b4 and Black is
winning due to the many threats, including ...f6 and ...g5) 9... c5!
when Black is better, although White can perhaps pretend to play a
kind of reversed Traxler Gambit with 10. f3!?. Interested readers
are invited to check the ensuing madness with their own engine.

222
8... d4!
Black’s strongest and cleanest move. I like how, in a way, this is
completely consistent with what I’ve been advocating throughout
this book: in order to combat the Center Game, Black has to be bold
and brave, and not fear complications early on.
8...d5!? was also possible here, but it leads to insane complications
after 9. xd5! ( 9. xd5? c5! ) 9... c5! 10.0-0-0! xf2 11.b4! and
diehards will again have lots of fun exploring the crazy variations by
themselves.

9. xg4
White has to bite the bullet now.
9. d1? is simply too passive and all the tactics now work for Black:
9...c6! 10. ge2 d5! 11. xd4 c5! with a clear advantage.
9...d5
Again, cleanest. It also asks White yet again: where do you move
your queen?

223
9... xc2+ has never been tried, but is playable too. The position is a
mess after 10. f1 xa1 ( 10...d5 11. d1 xa1 12. xd5 ) 11. d1
when Black’s best move apparently is 11...b5! with, of course, a
0.00 position.
10. h5?
Enterprising and brave, but objectively wrong. Curiously, Varavin
doesn’t mention any alternatives in his analysis in Chess Informant
83.
White should have gone for the modest but clever 10. d1! dxc4
11. e3 f6 ( 11...c5!? ) 12. ge2 xe2 (or 12...c5 13.0-0=
Bragesjoe-Michalek, ICCF email 2018) 13. xd8 xd8 14. xe2
xc3 15.bxc3 with a dead-drawn position. This might be White’s
most reliable way to reach an equal game after Black’s assault with
7... g4!.

10...dxc4?
Letting White off the hook.

224
Black had to dare White to prove his compensation after
10... xc2+! 11. d1 xa1 12.exd5 ( 12. xd5? loses to 12...c6 )
12...b5! .
11.0-0-0
White has somehow managed to castle queenside, and again,
immediately, he’s somewhat better.
11...c5
Better was the subtle 11... e8!, stepping out of the pin, but 12. e3
c6 13. ge2 was still more pleasant for White, who will follow up
with f4 and the usual pressure on the kingside.
12. f3 g6
12... e6 13. e3.
13. h6 g4

14.h4!
White goes for the standard attack on the h-file.
14... xf3

225
Perhaps it was more tenacious to take on f3 with the other piece:
14... xf3 15.gxf3 xf3 16.h5! xh1 17. f4! ( 17. xh1 g5! ). Black
needs to give up his queen and White is better, but it’s still a game.
15.gxf3 f6 16. e3 g7 17.f4?!
17. d5!.
17... e8 18.h5 b6?
Black had to go for the direct counter-attack: 18...b5!.
19.hxg6 hxg6 20.f5!?
Speculatively opening more lines, Center Game style!
Varavin points out that 20. d5? e2+ is mate in two.
20... xf5 21. f4 d4 22. de1 ad8?
The best move was 22... e5! 23. g3 d8 and Black is still holding
on.
23. h4
The occupation of the h-file is deadly.
23... e6 24. d5?!
It has been a high-quality game so far, but now the level drops
below sea level due to time-trouble.
24. h7+ f8 25. h6 was winning: 25... xh6+ 26. xh6+ e7
27. d5+ d7 28. f4! b5 29.a4! and White wins material.
24... e5 25. e3? f5?
25... f8! was unclear.
26.f4??
Varavin thought 26.c3! fxe4 27. h7+ f7 was unclear, but White is
winning after 28.cxd4! xd5 29. h4!.
26... d6??
Nasty was 26... b3+! and for once, it is Black who triumphs over
the long diagonal.

226
27. h7+ f7 28. xd4 cxd4 29.exf5 xd5??
29...d3! (Varavin) would have saved the game.
30. xg6+ f8 31.f6 xf6 32. xf6+ 1-0
The other way for White is to play the non-standard move g1-f3
and go for kingside castling at an early stage. As we saw in the
historical part, this set-up, recommended by Gorny, was already
tried twice (via a different move-order) by Bannerjee in the middle
of the 19th century. It leads to interesting play, which is perhaps
sufficient for a practical game, but White doesn’t get his desired
queenside castling set-up.
In summary, I think White is caught between a rock and a hard
place in the ... e7 variation.
Mario Napolitano
Valt Borsony
ICCF correspondence game 1956
In this tense and theoretically still relevant correspondence game,
we get to see another interesting way for White to play after
stopping the ...d7-d5 push: with an early g1-f3. The position after
Black’s 7th move can also arise via a different move-order from the
Philidor Defence (starting with 1.e4 e5 2. f3 d6 ) or the Italian
Opening (Hungarian Defence, starting with 1.e4 e5 2. f3 c6
3. c4 e7 ).
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 f6 5. c3 e7 6. c4 0-0
7. f3!?

227
7...d6
The most natural move, but not the only one:
A) 7... b4!? 8. b3 d5 9.e5! g4 10. d2 deserves attention;
B) Gorny calls 7... b4 ‘probably the best move in the position’,
and this may be true, but it’s not likely to happen on the board and
has only been played once so far. It seems paradoxical to move the
bishop twice, but e4 is vulnerable whenever White doesn’t have the
natural defensive move f2-f3 at his disposal. After 8. d2 ( 8.0-0!?
xc3 ( 8...d6 was also possible, but Black has to reckon with
9. d5!? ) 9.bxc3 e8 10. d3! xe4 11. d5 with some
compensation) he analyses the forcing line 8... xc3 9. xc3 xe4!
10. xe4 e8 11. e5 xe5 12. xe5 f6 ( 12...d6? 13.0-0-0 ) 13.f4
d6 14.0-0-0 dxe5 15.fxe5 g5+ 16. b1

228
 analysis diagram
and concludes that the game is equal, but he ‘would rather take
White’. However, Black has 16... g4, which is not hard to find,
forcing a dead draw, e.g. 17. xg4 ( 17. d3 e6 ) 17... xg4
18. hf1 f8 19. d4 h5 20. df4 e6 21. xe6 fxe6;
C) 7... e8!? is another entirely new move which the engines like,
and seems very sensible too. White doesn’t have anything better
than 8.0-0 h6 and now, for example, 9. d2 f8 10. ad1 d6 11. fe1
b4 12. b3 e6 when White should play 13. e2 xb3 14. xb3
a6 15.e5 c5 16. c4, which is equal.
8.0-0
We saw this position already in Chapter 1, in the game Bannerjee-
Cochrane, Calcutta 1855.
8. d2!? is more in the spirit of the Center Game.

229
 analysis diagram
White should be prepared to give up the bishop pair with 8... g4 (
8... e5 9. xe5 dxe5 10. d5 xd5 11. xd5 c6 12. b3 a5 13.a4
and White’s position seems entirely playable) 9. e2 ge5 and now
queenside castling is possible: 10.0-0-0 g4 ( 10... xc4 11. xc4
e6 12. d5 xd5 13. xd5, and again, White’s position looks a bit
more pleasant to play, especially because the automatic 13... f6
14.g4! is good for him) 11. e3!, intending d5.

230
8... g4!?
The most played move in practice, but I don’t think it’s the best one.
A) 8... g4 is Black’s second most popular choice (after 8... g4).
Black wants to simplify by taking on f3 and then play ... e5. Gorny
suggests the clever 9.a3!? with the idea a2 if Black doesn’t do
anything ( 9. d2!? was played by Bannerjee, and is not bad either.
Cochrane continued with the weird 9...b6?! 10.h3 c8!? and now
White played 11.a3!? with unclear consequences. The game
Bannerjee-Cochrane, Calcutta 1855, was ultimately drawn), and if
he goes 9... xf3 ( 9... d7!?; 9... e6!? )

231
 analysis diagram
White doesn’t take with the queen but with the pawn: 10.gxf3!,
intending h1 with the initiative;
B) In the case of 8... e6, White has many options. Gorny analyses
9. d5, and this is certainly possible, but from a practical
perspective I like 9. xe6 ( 9. d5!? ) 9...fxe6 10. g5 d7 11. h3
d4 12. e3 c5 13. xd4 cxd4 14. e2 e5 15. b3+ d5 16. g3 and
although the position is equal, it is still full of life;
C) 8...h6 is the engine’s top choice. It’s a useful waiting move.
Play might continue 9. e1 ( 9.b3!? ) 9... e8 10.b3!? g4 11. d2
followed by b2 and/or d5 at some point.
9. e2 ge5 10. xe5
Not 10. b3? g4! and White was already close to losing in another
Bannerjee-Cochrane game in 1855.
10... xe5

232
Less strong is 10...dxe5 11. h5! , e.g. 11... d4 12. xe5 xc2
13. d1 d6 14. h5 (Gorny) and now 14... xa1? 15.e5! wins for
White.
11. b3

This natural position is worth investigating further. It seems to me

that White, who has a bit more space, can count on a small
advantage. Black, on the other hand, has no weaknesses.
11...c6
In two games by lower-rated players, Black tried the moves 11...a6
and 11... h8, but these don’t exactly inspire confidence.
11... e6 allows 12.f4 xb3 13.axb3 d7 14. f3 and I like White
here.
12. f4
This looks a bit timid, but White will follow up with f4 anyway after
the preparatory f4-g3.

233
12.f4!? appears to me more in the spirit of the opening; for example,
12... d7 ( 12... g6 13. h5 ) 13. e3 b5 14.a3 a5 15. f3 and
White’s position seems easier to play.
12... f6?!
A) After 12...b5 White should go 13. d2! ( 13. ad1 b6= )
13... e6 14. ad1 c7 15. e2 ;
B) 12... e8 13. g3 f8 14.f4 d7 15.f5!? and White has the
initiative.

13. d1?!
A slow move, hard to understand if you look at the game with a
modern engine, though the plan in itself is logical enough: White
wants to bring the knight to f5, but it takes too much time.
Better is 13. ad1 or 13. g3!? with the idea to play f2-f4.
13...a5 14.c3?!
14.a4= was better.
14...a4!
This sort of pawn push is par for the course for modern engines.
234
15. c2 e6
Black has the upper hand here.
16. e3 b6 17. fb1
If 17.b3, then 17...axb3 18.axb3 c5! nicely exploits the
weaknesses on the queenside.
17... fd8 18. g3 c5 19. h1?!
19.a3 .
19...a3!
Thematic and strong.
20.b3 xc3 21. f1

21...g5?
Trying to prevent f2-f4, but weakening his own king.
21... g4! 22. xg4 xg4 23.f3 h5 and Black is a pawn up.
22. ad1?!
22. ac1! with counterplay.
22... g7?!
235
Black also plays too slowly.
Pawn-grabbing was possible here: 22... b2! 23.h4 gxh4 24. h2 and
now Black could have taken over the attack by means of 24... h8!.
23.h4 f6 24.hxg5 fxg5 25.f4?!
A desperate bid for freedom... but an unnecessary one!
25. b1 was still balanced.
25...gxf4 26. xf4 d7! 27. f5 xf5 28.exf5

28...d5?
After the blockading move 28... f6, Black would have had the
better game.
29. de1!
Now the rooks come into play.
29... f7 30. f3 f6 31. h3
White’s attack is in full swing, but it’s not that easy yet.
31...h6 32. e6 ad8 33. xh6
33. xh6! would have finished the game.
236
33... xe6 34. hxe6 c3 35. 1e3 b4 36. g6+ f8 37.f6 d6
38. eg3 e5 39. 6g5 c4!? 40. 3g4?
40. h5! xf4 41. g6! was an elegant win for White. White missed
a few more chances, while Black fought like a lion and rescued the
half-point after 57 moves.

237
Chapter 5
Paulsen Variation
4. e3: other 4th
moves for Black
‘Who cares for you?’ said Alice (she had grown to her full size by
this time). ‘You’re nothing but a pack of cards!’
Lewis Carroll, Alice’s Adventures in Wonderland

In this chapter, we will take a look at other alternatives for Black on

move four against the Paulsen Variation. Most of these moves aren’t
impressive and won’t give White much stress. Some of them are
simply mistakes, others are a confused mix-up of more promising
ways of playing for Black. White shouldn’t be afraid of them.

Miscellaneous fourth moves

The first game is an example of this: 4... b4+ is fine, usually
transposing to lines already examined, but in combination with
... ge7, it is unfortunate for Black.
Xie Jun (2500)
Raset Ziatdinov (2485)
Wijk aan Zee 1997

238
We will now see former Women’s World Champion Xie Jun, who
has played the Center Game a number of times, perform a
masterclass against the set-up with ... b4 and ... ge7 – until she
forgets about one move and allows her opponent to turn the game
into a messy boxing contest.
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 b4+!?
First played by Fritz Riemann against Winawer in 1881, this move
is not bad, of course (it was also played by Magnus Carlsen once).
But it shouldn’t be played in combination with Black’s next.

5. c3
Expecting a transposition.
5.c3?! e7 is the so-called Charousek Variation, which seems a bit
too much of an honour since the Hungarian master only played it
once, in 1897, in a game against Winawer. Black is fine here and it
is once again testimony that the move c2-c3 is usually not
recommended in the Center Game. ( 5... a5 , which was Riemann’s
choice against Winawer, was even better.)

239
5... ge7?
This logical-ish move, once also played by Mikhail Botvinnik, is
surprisingly bad according to the engine, which already rates
White’s advantage at +1.20 at this point. The problem is that Black
can only challenge White in the centre with the move ...f7-f5 here,
but that weakens his entire position.
5... f6 leads to the main line of the Berger Variation.
6. d2
6.a3!? a5 7. d2 0-0 8.0-0-0 .
6...0-0
Postponing kingside castling is also too slow: 6...d6 7.0-0-0 e6
8.f4! and after 8...f5 ( 8... g4 9. e1 ) White has the deadly 9.g4!
with a terrific position.
7.0-0-0 d6

It looks like Black hasn’t done much wrong, but...

8.h4!
Strong, and consistent with the ideas of the Center Game.
240
8. g3!? h8 9.f4 f5 10.a3 ( 10.e5?! was Tolush-Botvinnik,
Moscow 1944) 10... a5 11. d3 or 8.f4! were also good for White.
8...f5
This push is regularly seen with a knight on e7, but it can never
scare White.
A) 8... h8 9. h3! f5 10.h5, followed by f4, is a disaster for
Black;
B) The engine wants to play the strange 8...h6, after which White
is for choice, e.g. 9.a3 (or 9. h3 right away) 9... a5
10. h3/10. g3 .
9. h3
Simple and strong.
A) It was also possible to play 9. c4+!? though White needs to
watch out for ... e5 at some point;
B) The computer’s choice is 9.a3 a5 ( 9... c5 10. g3 ) 10. c4+
h8 11. h3 with a clear advantage.
9... e5
The alternatives were hardly better: 9...h6 10. f4; or 9...fxe4
10. xe4 xd2+ 11. xd2 e5 12. e2.
10. e2 e6?
More stubborn was 10...h6 11.f4 g4 12. xg4 fxg4 13. f2 and
White is still doing very well.

241
A blunder in a very questionable position. Now Xie Jun had the
possibility to play one of the most important standard moves in the
Center Game, but she declined to do so.
11. g5?
Missing Black’s next move, a nasty zwischenzug.
11. f4! would have given White an overwhelming position. If
11... f7 ( 11... d7 12. xe6 xe6 13.exf5 xf5 14. e4!, followed
by g2-g4, almost wins already) 12.exf5 xf5, then White has
13. e4! attacking both f5 and b4, forcing Black’s next: 13... xc3
14. xc3 and now, for example, 14... f6 15. d3 ae8 16. xe5
dxe5 17. d3 g6 ( 17...g6 loses to 18. xb7 ) 18. d5+ f7
19. xb7 and White wins.
11...f4!
The only move, but very strong. White’s advantage has softly and
suddenly vanished away. Not 11... c5? 12. h3 and White wins.
12. d4?
A mistake rarely comes alone.

242
Xie Jun should have temporarily sacrificed her queen: 12. xf4!
xf4 13. xe6 xf2 14. xd8 xd8 15. e1 xg2 16. f1! g4
17. h3 and White is still in the game.
12... 7c6?!
Stronger was 12...c5, forcing White to sac her queen again: 13. xe5
dxe5 14. xe6 b6 15. c4 h8 16. xf8 xf8 .
13. xb4
The third way to ‘sacrifice’ the queen!
13... xb4 14. xe6 d7 15. xf8
The surprising 15. xf4! may have been better.
15... xf8 16.a3?!
Better was 16.h5, though Black would still be doing fine.
16... bc6 17.f3 e7?
17... d4! would have kept the advantage.
18.h5!
Now, White is no longer much worse.
18...c6 19. f1
Intending to re-route the knight.
19... e6 20. b1 b5?!
20...d5!?.
21. e2!
White is a little better again. Black’s next two moves don’t help.
21...g5?! 22.hxg6 7xg6?! 23. d4 f6 24. f5 d8

243
25. c3?
What a pity!
White should have just tried the good old method of doubling on the
h-file right away: 25. h5! with a near-winning advantage.
25... e6?
25... g5! would have prevented h1-h5.
26. e2?!
26. h5! d7 27.g3! was winning.
26... d7 27. h5
27. dg1!, with the idea g2-g3, was still strong.
27...a6 28. dh1 c5 29. h6+?
Now, this difficult but interesting game ends in a repetition.
White had to find the super-human sequence 29.a4! b4 30. xe5!
xe5 31. h6 in order to retain winning chances.
29... f8 30. f5 g8?
30...d5! with counterplay.
244
31. h6+?
31.a4!.
31... f8 32. f5 ½-½
A great fight!
The move 4...d6 (Game 38) is also okay, but very uneventful for
Black. The rare move 4... d6!? (Game 39) is at least interesting: it’s
an underestimated way to avoid mainstream theory and confuse
White. I have successfully played it myself on some of the rare
occasions I’ve faced the Center Game as Black.
Ivan Sredojevic (2253)
David Tai Wei Wu (1960)
Senta 2019
In this game, we will see how White should play against a passive-
solid set-up with an early ...d6. Gorny jokingly calls it ‘The Scared
Dog’, since it may be used by Black players who are scared of the
e4-e5 push. (Perhaps he forgot that one of the best and least ‘scared’
players in the history of the game, Anatoly Karpov, played it once.)
It’s not that bad for Black, just unambitious. As Black doesn’t
challenge White in the centre, the simple advice is to develop our
pieces by the usual means and all will be fine. The game itself is not
of the highest quality, but it offers an opportunity to see how players
of slightly lower levels deal with the opening even after mainstream
theory has ended and the engines give a resounding evaluation.
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 d6

245
5. c3 f6
Note that this position can also arise via the Berger Variation with
5... d6.
A) 5...g6 will lead to fianchetto positions we will examine later in
this chapter;
B) After the passive 5... d7?!, White’s position plays itself.
Gorny looks at the following line: 6. d2 f6 7.0-0-0 e7 and now
his recommendation 8. b1!, with the idea of c1, is worth taking
note of. White has an overwhelming position;
C) 5... e6 makes more sense, intending ...0-0-0, but here too,
White is doing more than fine: 6. d2 d7 7.0-0-0 (a typical move
to try in blitz is 7.f4!? when the number of games I have won after
7...g6?? 8.f5 can’t be counted on two hands; Black should play the
precise 7...0-0-0! and now 8.f5? is actually bad for White due to
8... xf5! 9.exf5? e8 10. e4 f6 11. d3 xe4 12. xe4 d5 ) 7...0-
0-0.

246
 analysis diagram
In such positions, in which both sides castle queenside, White can
often go for the following annoying pin: 8. b5! ( 8.f4 allows 8...d5,
though White is still better after 9.exd5 xd5 10. e1! ) 8... ge7
9. ge2, threatening both d4 and f4, when it’s already hard to
find sensible moves for Black.
6. d2
A) In Chapter 2, we already saw the interesting move 6.b3!?,
which has often been tried by Muñoz Pantoja. Black should go for
6... e7 7. b2 d5! with an unclear position;
B) It’s worth looking into the rare 6. d5!? when 6... e7 is
Black’s best move, e.g. 7. d2 0-0 8.0-0-0 e8 9. xe7+ xe7
10.f3 .
6... e7
6... e6 is possible here, e.g. 7.0-0-0 d5 8.exd5 xd5 and now the
natural move 9. f3! is actually a novelty ( 9. g3 xc3! 10. xc3
d6 11. xg7 f4+ 12. d2 e5 is equal). White has a plus.
247
7.0-0-0 0-0
A typical tabiya, regularly seen in blitz or club practice. From this
position, Black often goes for ...a6 and ...b5, although this is not
necessarily the right plan. Usually, it’s better to play for ...d6-d5,
even if it means losing a tempo.
7... e6?! makes some sense, preparing ...d6-d5.

analysis diagram
White’s best move, but by no means the most played one, is
8. h3! with a classic set-up to follow, e.g. 8... d7 ( 8...0-0 9. f4
followed by f3, g4 etc. and White is much faster; 8...d5 9.exd5
xd5 10. g3 ) 9. f4 0-0-0 10. cd5 with an advantage.
The automatic 8.f4 has been played by experienced Center Game
players like Nepomniachtchi, Muñoz Pantoja and... myself, but it is
not the strongest move in the position. The problem is that Black has
the immediate 8...d5!, which is not that clear. I faced it once myself
in a Titled Tuesday blitz game, which continued 9.exd5 xd5
10. g3 ( 10. f2! ) 10...0-0? Moll-Novak, Chess.com 2020.

248
 analysis diagram
Now, 11. e1! would have won material. A move to remember!

8. g3!

249
First played by Hase against Karpov. Black has many possible
moves at this point, but we will consider just a handful which sort of
make sense.
8.f4 is the most popular move here (also once played by Carlsen in a
bullet game), but it allows 8...d5 9.exd5 b4!= Prelevic-P.Orlov,
Belgrade 2001.
Playable but rare is 8. f3!? e8 9. c4 .
8... e8!
Centralization is usually the best recipe.
A) 8...a6?! 9.f4 d5 ( 9...b5?! 10.e5 d7 11. f3 – see Hase-
Karpov, Skopje 1972 in Chapter 1) 10.exd5 b4 11.a3 bxd5
12. d3 and White is better;
B) Black can always hope for a repetition with 8... h5 9. f3 f6,
but White doesn’t have to go 10. g3 and can instead play 10.h3! ;
C) 8...d5? is simply bad here: 9.exd5 d6 ( 9... b4 10. c4 f5
11. b3 ) 10. h4 e5 11.h3! with the following sample line:
11... g6 12. d4 e5 13. d3 f4 14. xf4 xf4+ 15. b1 and
White has a big advantage;
D) 8... h8?! 9.f4 d5 10.exd5 b4 11. f3 bxd5 12. e5!? .
9.f4
The typical move when Black is not yet challenging the e-pawn.
9.f3 is too slow, and allows Black to develop counterplay with the
push 9...d5! 10.exd5 b4. White would really love to play f3
now, but he can’t.
9... f8
Finally hitting the pawn on e4, but White is again happy to sacrifice
it.
Here, too, 9...d5! came into serious consideration. There are no
games with this move in the database, but it has been analysed by
Gorny. His line continues 10.e5 g4 11. f3 c5 12. d3! f2 (
12... f2 13. h3 g6 14. g5 h6 15.e6!
250
 analysis diagram
with a crazy position, which the engine somehow manages to defend
by means of 15...f5! 16. f7 h4 although White should still be
somewhat better after further complications: 17. xh4 xh4
18. xd5 xe6 19. xh6+ xh6 20. xc7 f7 21. xa8 xa8 22.h3
) 13. g5! and White has a strong attack, e.g. 13...h6 14. h7+ f8
15. f3 xh1 16. xh1 .

251
10. f3!
More careful but also fine was 10. e1 with a pull.
10...g6?
It’s not unreasonable to bring the bishop to g7, but here it’s simply
too slow.
A) 10... xe4? is the wrong way to capture the pawn. White gets a
devastating attack after 11. xe4 xe4 12. d3 e6 ( 12... e8
13. g5 g6 14.h4 and Black won’t survive) and now the engine
suggests the cruel 13. b1! , but 13.f5 is also strong since 13... f6
already runs into 14. g5;
B) The only move was the exchange sacrifice 10... xe4! 11. xe4
xe4, which at least forces the white queen to retreat. White is still
better, though: 12. e1 d5 13. d3 c5 when White can go for a
simple plus with 14. b5 or try the interesting ‘Greek Gift sacrifice’
14. xh7+!? xh7 15. g5+ g8 16. h4 f5 17.g4 d7 18.gxf5
xf5 19. f2 .
11. c4

252
Simple chess. White has all his pieces in the game while Black is
still hesistant about which set-up to undertake. Still, it’s interesting
to see the rest of this game as it shows that even positions where the
engines give +2.0 are not that easy to win in practice.
11... e6
11... g7 was not much better: 12.e5 h5 13. f2 a5 14. e2 dxe5
15.fxe5 g4 16. hf1 with a winning position.
12. xe6 xe6
12...fxe6 13.e5 and White wins.
13. g5?!
The most forceful line was 13.e5! dxe5 14.fxe5 h5 15. h3 xe5
16. xe5 xe5 17.g4! f6 18. f4 and White wins.
13... e8 14. he1
14.h4!? g7 15.h5! xh5 16. d3 was also dangerous for Black.
14... g7 15. b1 b8!?

253
Black could have tried to exchange queens with 15... c8, after
which White should play 16. d3 h6 17. f3 .

16. c1!?
We saw this little move a few times before already.
16.e5! was also strong.
16... c8 17.f5 e5?!
The brave 17...gxf5 was better. White remains better but it’s
actually not that clear.
18. f1!
Switching to the f-file.
18...h6 19. f3 h7
19...g5 20. xg5! hxg5 21. xg5 f8 22. h4, followed by the
simple g2-g4, and White has a great attack for the piece.
20. h4 fd7 21.g4
21. d5 .
21... xf3 22. xf3 d8 23. g5??
254
A hallucination, no doubt.
23... xg5
‘And resigns’, one is tempted to write, but not here! White simply
plays on as if nothing has happened and, lo and behold, miracles do
sometimes occur!
24. f2 e5?! 25. g3 a6? 26.h4!? xh4 27.g5 h5 28. g2

Setting a small trap.

28... g8??
28... c4 was winning.
29. h1 f4?
29... xh1+ 30. xh1 g4 .
30. d5
Suddenly, White is winning again. Black had to give up the queen
and surrendered 24 moves later.
Eric Samuels
John Sheatsley

255
US Amateur team championships 2009
In this game, Black plays an interesting and underestimated idea that
will surely surprise even the most experienced Center Game players.
The engines like it better than some of the more common
alternatives, so it is worth taking seriously.
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3

In this position, almost every move that doesn’t drop material has
been tried according to my database. We will examine a handful of
the most sensible ones:
4... d6!?
Of all the ‘unusual’ fourth move alternatives aimed to take White
out of his comfort zone, I think I like this one best. At first it looks
just stupid, doesn’t it? But actually, it is very cunning
psychologically. Black wants to proceed similar to the Berger
Variation, but avoids lines with either e3-g3 or e3-f4. This
means White can’t play his usual set-up, and to some, this is already
a considerable blow!

256
 A) 4...f5?! is called the L’Hermet Variation, after the German
player Rudolf l’Hermet (1859-1945). It was advocated in the late
1980s by the Hungarian player Szolt Timar. It looks absurd, but it’s
not so bad. White can still be happy if he knows the right set-up:
5.exf5+ (Greet recommends 5.e5, but I think this lets Black off the
hook too easily) 5... e7 6. d3 d5 7. h3!. This is the key move, and
all you have to remember as White. Black can’t win the f5-pawn
back yet and struggles with his weakened kingside. Play might
continue 7... f6 8. f3 b4 9.0-0 xd3 10.cxd3 0-0 and here I like
the simple 11. d4 ;
B) 4...b6?! is a weak move, often played by players who think it’s
beneficial to develop the bishop to c5 with tempo: 5. c3 c5 (
5... b7 may be slightly better objectively, but loses even more time.
White is doing great after 6. d2 f6 7.0-0-0 ) 6. g3 f6 (we
already briefly saw 6... f6?? 7. d5 in the Introduction) and here
White should simply play 7. d2! (certainly not the greedy 7. xg7?
g8 8. h6 xf2+!, which I have fallen for a few times in bullet –
remember, I’m bad at memorizing lines; 7. g5!? h6 8. d2 is worth
a try though) 7...0-0 ( 7...d5 8. b5 ) 8.0-0-0 e8 9.f3! ;
C) 4... f6!? is possible, but White should be happy: 5. c3 b4 (
5... d4 6. d3 Winawer-Chigorin, Vienna 1882) 6. d2 ge7 (
6...d6 7.f4 ) 7.0-0-0 0-0 8.f4! d6 and now the new move 9.f5! is
very strong for White. The idea is to go f4 and simply storm the
kingside, whilst 9...d5 10.exd5 xf5 11. f4 d6 runs into 12. e4!
and wins;
D) 4... b4 was a bit of a big deal around 15 years ago. Dutch IM
Jeroen Bosch wrote an article about the move in Secrets of Opening
Surprises Vol. 12, calling it ‘delightfully direct’. Well, I think if
Black is allowed to move the same piece twice, then White is surely
allowed to move the same piece thrice: 5. e2! (it’s also possible to
dive deeply into the complications of 5. a3 f6 6.e5 fd5 7. e4
d6!? (Thorhallsson-Hebden, Cappelle-la- Grande 2005), which the

257
engines still evaluate as very good for White, but which indeed
looks rather tricky if you’re not made of silicon).

analysis diagram
Puzzlingly, this is only the fifth most popular move for White in this
position, even though computers have indicated it as the best move
for more than twenty years. The plan is very simple too – chase
away the knight with a2-a3 and get on with business: 5... f6 ( 5...b6
6.a3 a6 is nice for a bullet game, but after 7.c4! c6 8. f3 White
is doing well) 6. c3 ( 6.a3!? is a simple alternative. After 6... c6
7. c3 d6 I like 8. e3! followed by 0-0-0, and White is doing well).
This position can also arise after 4... f6 5. c3 b4 6. e2. Here,
Black’s most popular move is 6...b6 (the alternatives aren’t
promising: 6...d5 7.a3! d4 8.axb4 dxc3 9.bxc3 ; 6...d6 7. f4! e6
8.0-0-0!? ): 7.e5! (White should play agressively against the rare
stuff) 7... a6 ( 7... fd5 8. e4 a6 9.c4 ) 8. d1 xf1 9. xf1 and
White has a huge advantage.
5. c3
A) 5. f3 f6 6.e5? loses to 6...0-0!, followed by ... e8;
258
 B) 5.e5 b4+! 6.c3 a5 is fine for Black, e.g. 7. g3 ge7!
8. xg7 g8 9. xh7 d5! with compensation.
5... f6 6. d2 0-0 7.0-0-0 e8

There is only one game in the database at this point (but see
Signorelli-Brunello in Chapter 3!). With a black bishop on b4, this
would simply be the main line of the Berger Variation. Here, both
g3 and f4 are impossible. The good news is that ...d7-d5 is not
possible.
8.f4?!
This may look tempting, but with a rook on e8 it is usually too
weakening. There were two better alternatives:
A) 8. f3 ( 8.f3 e5! is a typical manoeuvre) requires White to
play an entirely different kind of opening set-up, but is objectively
the best move in the position. Play might continue 8... f8 9. c4 (
9. f4!? ) 9...h6 (Black has to prevent g5) 10. he1 d6 11.h3 a6
and athough the engine gives White a very slight plus, I think this is
a perfectly playable position for Black;

259
 B) 8. c4!? is the sharpest move, leading to very interesting play:
8...a6!. Black aims to put pressure on e4 with ...b7-b5-b4. White
must try something fast: 9.f4! ( 9.f3 is too slow: 9...b5 10. b3
a5 ; 9. f3 b5 10. b3 h6 11.g4! is interesting and unclear) 9...b5
10. d3 b4 11. ce2 f8, when the engines give the un-human
12. e1! as White’s best move.
8... b4! 9. g3?!
If only White could get away with this gambit here, all would be
good. But he can’t.
9.e5 was necessary: 9...d6 10.a3 a5! ( 10... c5 11. g3 dxe5 12.fxe5
h5 ( 12... xe5?? 13. h6 ) 13. f3 h4! ) 11. f3 g4 12. e1
dxe5 13. g3! and I leave it to the interested reader to explore this
unclear position further.

9...d6?
This is not the way. As usual in the Center Game, Black should not
be afraid!

260
 A) 9... xe4 10. xe4 xe4 11.c3 leads to positions we have seen
in Chapter 3, but with the extra move f4 for White, which is a clear
improvement;
B) The right move was 9... xe4!, analoguous to the classical
Tarrasch Gambit with 9.a3. However, here there is a pawn on f4,
which makes a real difference. For example, after 10.a3 xc3!?
11. xc3 Black here has 11...d5! since 12. xf6 ( 12. d3 d4 13. f3
f5! ) 12... xf6 13. xd5 now allows Black to take the pawn:
13... xf4!, when White doesn’t have enough compensation.
10. d3
Now, White has the advantage. The rest of the game is of no
theoretical significance and I’m providing it with minimal
comments.
10... h5 11. f3 h4?!
11... f6 12.h3 .
12.g3 g4

13.h3?!
261
13. f2! c5 14. g2! is a queen manoeuvre well worth
remembering. If 13. g2?, Black has 13... xc3 with the idea
14. xc3 xf4, which is why the bishop must be lured to c5 first.
13... xf3 14. xf3 xg3 15. hg1
As often, the open g-file is beneficial to White, even in the
endgame, but the game petered out to a draw on move 43.

The Fianchetto
The Fianchetto System, already analysed by Stamma, was
advocated by several authors in recent decades. It is the only serious
alternative for Black to set-ups with ... b4 or ... e7.

position after 4. e3 g6
In a way, this is the most logical way of countering White’s
intention to castle queenside. On g7, the dark-squared bishop isn’t in
the way of a rook on e8 (as it would be on e7) and looks menacingly
down to b2. In some of the lines involving ...g6, Black castles
queenside. White usually proceeds in attacking style, going for an
262
early h2-h4 (even if Black goes ...0-0-0). I think the modern move
5.h4!? (Game 40) is White’s most interesting option. The positions
are often explosive and require precision from both sides. Still, I
think Black has to be more careful than White.
Nodirbek Abdusattorov (2627)
Gregory Shahade (2476)
Titled Tuesday blitz 2020
1.e4 e5 2.d4 exd4 3. xd4 c6 4. e3 g6
The right move-order to play the Fianchetto Variation. Black should
wait with ... f6 until after d2.
4... f6 5. c3 g6?! is inaccurate here.

analysis diagram
It allows the extremely rare but strong 6.e5! ( 6. d2 transposes to
regular lines after 5. c3 ) and Black is already suffering: 6... g4
7. e2 (less strong is 7. e4?! gxe5 8.f4 d5! 9. xd5 ( 9. e2? d4
) 9... g4 10. xd8+ xd8 ) 7...d5 (or 7...d6; 7... gxe5 8.f4 )
8.exd6+ e6 9. e4 cxd6 10. g5 e7 11. xe6 xe6 12. xe6+

263
fxe6 13.f3 followed by d2 and 0-0-0, and White is better due to his
bishop pair and the potential weakness of the black centre pawns.
4...d6 5. c3 g6 is likely to transpose to lines where Black plays
...d6 before castling.

5.h4!?
Much more popular is the standard 5. c3, but this allows Black to
obtain a favourable version of the Fianchetto System. For the sake
of completeness, we will look at some important lines: 5... g7 (for
5... f6?! 6.e5 see the note to move 4; 5... h6?! is a sly move for a
bullet game, but White is better after the alert 6. g3! ) 6. d2 (
6. c4!? was first played by Chigorin in 1906, and is still an
interesting way to diversify. In 2002, Belgian FM Arben Dardha,
who has played the Center Game extensively in his chess career,
used it successfully to hold the mighty Anton Korobov to a draw):

264
 analysis diagram
A) 6... ge7?! is surprisingly popular, but it is better for White:
7.0-0-0 d6 ( 7...0-0?, which we already saw in the game Mieses-
Spielmann, Breslau 1912, in Chapter 1, is simply bad here because
of 8.h4!, preparing a positional piece sacrifice: 8...d5 ( 8...h5 9.g4!
hxg4 10.h5 plays itself) 9.h5! d4 ( 9... d6 10.exd5 xd5 11. xd5
xd5 12.hxg6 xa2 13.gxh7+ h8 14. c3 ) 10. g3 and White
is almost winning already. Some sample lines to demonstrate this:
10...dxc3 11. xc3 d7 ( 11... e8 12.hxg6 xg6 13. h2 h6
14. xg7 xg7 15. xh6+ f6 16. f3 , which occurred,
impressively, in Garcia Rodriguez-Atares Buisan, Salobrena 2021)
12.hxg6 xg6 13. h2 g5+ 14. d2 xb2+ 15. b1 and wins)
8. ce2!. This new move is a sublime way to exchange the dark-
squared bishops and secure a considerable advantage. It’s also a
useful waiting move to see which way Black wants to castle.

265
 analysis diagram after 8. ce2
Play may continue 8...0-0 (engines suggest the bizarre 8... g8!? as
Black’s best move, in order to be able to play ... f6 after c3.
White can continue 9.f3 and develop normally) 9. c3 e5 10.h3
followed by f2-f4;
B) 6... f6 7.0-0-0 ( 7.e5?! is much less strong here: 7... g4
8. g3 ( 8. e2 d5! ) 8...d6! 9.exd6 cxd6 Solovjov-Matlakov, St
Petersburg 2009) 7...0-0 ( 7...d6 8.h4! e6 ( 8...h5 9. h3!? –
Greet) 9. h3 was analysed in Chapter 2, when we looked at the
game Nepomniachtchi-Tomashevsky, Moscow 2006).

266
 analysis diagram
This used to be the main line of the Fianchetto Variation, already
played a lot in the 19th century and still seen today. Nepomniachtchi
has faced it several times as White. I think Black has relatively easy
and straightforward play and White’s plan of attacking on the
kingside is a bit slow here. I will give some sample lines and
potential directions of investigation:
B1) 8. e1?! was once tried by Nepo, but it’s not convincing after
8... e8 9.f3 and now, of course, 9...d5! when only Black can be
better;
B2) 8. g3 was played no less than four times by Nepo, but after
8... e8 9.f3 d5! White needs to be careful not to be worse;
B3) 8.h4 would be a great move if it wasn’t for 8...d5! 9.exd5 e8
10. f4 e7! ;
B4) The developing move 8. c4 feels like it should be the main
line, but Black has 8...a6!. Since he has not wasted time with ...d7-
d6 yet, he can afford to chase White’s bishop away: 9. f3 ( 9.f3 b5
10. b3 Reinderman-Ghaem Maghami, Beijing rapid 2008, and now
267
strong would have been 10...a5! when Black has all the fun) 9...b5
10. b3 ( 10. d5 b7 ) 10...a5! and I don’t like this one bit for
White;
B5) The rare 8. f3!? is Gorny’s recommendation: 8...d6 9.h4!? (
9. c4 e8 10. he1 and ‘White has the somewhat freer game’
according to Greet, but Black looks perfectly fine after 10...h6,
Rodriguez-Morales, ICCF email 2019).

analysis diagram
In style, but it feels a bit loose with a knight on f3. Gorny’s line now
continues sharply with 9... e8 10. c4 g4 11. g5 e5 12. b3
xd1 13.f4 eg4 14. xf7+ h8 15. xe8 xe3 16. f7+ g8
17. xd8 xd8 18. a4 xc2 19. xc2 xc2 20. xc2 and here I
think Black actually has a slightly more pleasant endgame after
20... e8, although it is, of course, holdable for White. Still, it’s not
clear to me if this is a useful line, given that not only Black but also
White will need to remember a number of only moves.

268
 position after 5.h4
This is not only a dangerous move for Black in a blitz game, but is
the engines’ top choice. Although it is a rare guest in tournament
practice, Abdusattorov wasn’t the first to play it: Carlos Ferron
Garcia already tried it back in 1995 and later it was employed by
WGM Almira Skripchenko. White’s idea is to open the h-file as
quickly as possible, even at the cost of a pawn. Black’s next move
is, therefore, understandable.
5...h5
A practical choice, which I think will often be played in your games.
A) 5... f6 6.h5! (the slow 6. e2?! g7 7.h5 0-0! is surprisingly
good for Black, who is fully mobilized and can easily parry the
threats along the h-file. He will soon play ... e8 and/or ...d5).

269
 analysis diagram
This is the critical test of White’s idea. Played in just two games in
the database, this dynamic pawn sacrifice is entirely in the spirit of
AlphaZero. As preparation for this book, I’ve tried it myself quite a
few times in online blitz and I can say it works wonderfully. The
game becomes spicy right from the start! 6... xh5 ( 6... g7 7.h6!; a
typical computer line is 6... e7!? 7.hxg6 ( 7. c3!? ) 7...fxg6 8. c3
d5! 9. xd5 ( 9.exd5!? b4 ) 9... xd5 10.exd5 b4 11. b5+ d7
12. xd7+ xd7 13. d1 xd5 14. h3+ e6 15. xe6+ xe6 with
an equal, but not uninteresting endgame) 7. c3. White has
compensation, but the position is still unclear and, with just a few
games in the database, almost entirely unexplored. Some sample
lines:
A1) 7... f6 8. d2 d6 9.0-0-0 ( 9. e2?! g7 10.0-0-0 e7
Willow-Hebden, England 4NCL 2021/22) 9... g7 10. h3! followed
by f4, c4 or even b5, with a lot of activity;
A2) 7... g7 8. d2 ( 8. e2!? f6?! ( 8... b4! 9. d2 ) 9.e5 and
the knight needs to go back into its stable, Weishäutel-Ashraf, Titled
270
Tuesday blitz 2024) 8...0-0 ( 8...d6 9.0-0-0 f6 10. h3 ) 9.0-0-0
e8.

analysis diagram
In this position White has several interesting ideas: 10.g4 ( 10. e1!?
is a hyper-subtle move, intending e2 and, sometimes, f2-f3, g2-g4
and e1-h4) 10... f6 11. g3! ( 11. h3!? d6 ) with a very unclear
position, requiring further research;
B) 5... g7 6.h5 is simply good for White. A sample line may be
6...d6 7. c3 e6 8. d2 d7 9.0-0-0 0-0-0 10.h6 f6 11. f3 b8
12. d5 ;
C) 5... h6? fails to 6. c3!, attacking the rook on h8.

271
6. c3
Now White can play freely.
6. c4!? was White’s choice in Skripchenko-Dornbusch, Kishinev
2005.
Or 6. g3!? with the idea g5.
6... g7 7. d2
Again, 7. g3!, intending g5, was a strong alternative.
7...d6 8.0-0-0 e6
Black hesitates to castle kingside, but this was still the lesser evil.
8... f6 9. g3 0-0 10.f3 e6 11. ge2 ( 11.f4!? ) also gives White a
promising attack.
9. h3!?
White doesn’t mind sacrificing the h-pawn, though it wasn’t strictly
necessary, with 9.f3 available.
9... ge7?!

272
 A) 9... xh4 10. d5 is good for White, e.g. 10... d8 ( 10...0-0-0
11. b5 ) 11. hf4 f6 12. c3 xd5 13. xd5 e5 14.f4 eg4
15. g3;
B) The knight is almost always better placed on f6: 9... f6
10. g5 e7 is a better try.
10. g5 d7 11. b1
11.g3!, intending h3 at some point, would have been very mean.
11...0-0-0

Shahade has managed to castle queenside, but White is still more

active and can exploit some weaknesses in Black’s camp. This
position is not unlikely to occur in club practice. White is usually a
bit better because he can obtain the pair of bishops and has more
space.
12. c1!?
Again, we see this move, trying to preserve the beautiful bishop.
Other moves, such as 12. e2 or 12.g3, were also good.
12... he8 13.f3
273
The machine wants to play 13. xe6 xe6 14. d5 because it sees
that White can materialistically win a pawn after 14... xd5 15.exd5
xe3 16. xe3 e5 17. xa7, though this is not a very human
approach. It’s true, though, that 17...b6 18. a6+ d7 19. b5+ c8
20. he1! b7 21. xe8 xe8 22. xb6 xb6 23.f4 wins for White.
13... b8 14. b5
The typical pin when Black has castled queenside.
14...a6 15. a4?!
It was better to take on e6 first: 15. xe6 xe6 16. a4 with an
advantage.
15... h6?!
A) Chasing the bishop away will backfire: 15...b5 16. xe6 fxe6
17. xb5! axb5 18. xb5 and White wins with the manoeuvre d1-
d3;
B) Best was 15... c4!, keeping the bishop and intending ...f7-f5,
with an unclear position.
16. f2 xg5?!
Giving up the bishop pair is bad, but White had a serious advantage
in any case.
17. xg5 c8?
This allows a very unpleasant knight jump.
More accurate was 17...b5 18. b3 xb3 19.axb3 c8.

274
18. d5!
The knight can’t be exchanged due to the pin of the c6-knight,
which is one of the reasons why b5 is often such a nuisance for
Black.
18... g8
18...b5 loses to 19. f6 d8 20. xe8.
19. b4
Destroying Black’s queenside pawn structure.
The surprising 19.c4! with the idea c4-c5 was also very strong.
19...f6 20. xc6 bxc6 21. xa6+ a8 22. d4
22. d3 fxg5 23. d2 also wins.
22...fxg5 23. c5?
Still winning, but White could have delivered the fatal blow already.
23. d3! would have been game over; for example, 23...c5 24. xc5
dxc5 25. xc5 a4 26. a3 and wins.
23...dxc5 24. a4+ b7
275
25. xd7 xd7?
Missing the zwischenzug 25... a8! when Abdusattorov would have
needed to show some endgame technique: 26. xa8+ xa8 27. g7
xa2+ 28. c1 g4 29. e1! ( 29. xg6?? f7 suddenly wins for
Black, who also threatens ... a1+) and because of the passed e-
pawn, White is still better.
26. b3+ a7 27.hxg5 e6 28. c3 b8?!
28...c4.
29. xc5+ b6 30.f4 b7 31.f5
31.a4!.
31...gxf5 32.exf5 b5!
Shahade shows excellent fighting spirit.
33. c3 xf5 34.a4 d5 35. f1 g6 36. a2?
36. e1 with advantage for White.
36... e2?
The pawn grab 36... xg5! was not clear at all now.
276
37. b4+ a7 38. f8
Now, White is winning again.
38... e8 39. xe8 xe8 40. f8
A deadly double attack (or is it a fork?).
40... xg5 41. xe8 xg2 42. xh5
The rest was easy: 1-0 (59)

277
Chapter 6
New alternatives for
White and minor
black moves
All of my words are secondhand and useless in the face of this.
The Sisters of Mercy, Some Kind of Stranger

We will now – finally – consider the alternatives for White on the

fourth move. In the last five years, players have come to realize that
the Paulsen Variation has drawbacks for White that are hard to
overcome. To recap:

• The old Tarrasch Gambit is dubious at best. Both 8... xe4 and
8... xe4 lead to pleasant positions for Black with correct play.
• Polgar’s 8. f4 is playable, but requires deep knowledge from
White as well, resulting in double-edged positions that may well be
better for Black in the end.
• The old 8. c4, although offering interesting options if Black is in
for a fight, allows a move repetition with 8... a5 and 9... c6.
• White can get an equal game after 5... e7 but, with perfect play
from Black, not more.

278
Alternatives to the Paulsen Variation open up an exciting and
largely brand-new area of investigation with lots of concrete lines,
novel ideas and unfinished discussions. I believe most future
research will be focused in this direction. Even after writing this
book, I feel I have only scratched the surface. Many of these lines
have not been popular in the past, and therefore there is very little or
no established theory on them. Let’s dive into them without fear!

The Malmö and Hall Variations

We first look at some minor options for White on move four. Of
these, 4. a4 has been considered relatively best. It was once played
by Bronstein (in a simul) and Bologan even went so far as to say
that of all the fourth move alternatives to the Paulsen Variation, it is
‘White’s safest option’.

White wants to get a reversed Scandinavian Defence with a pawn on

e4, hoping to benefit from the extra tempo – but the queen is too far
from the kingside to scare Black, and the a7-g1 diagonal seems
weakened. I feel Bologan was too optimistic.
Florian Grafl (2387)
279
Marc Capellades Subirana (2233)
Barcelona 2012
1.e4 e5 2.d4 exd4 3. xd4 c6 4. a4
A) 4. d1? gives away any dreams not only of an advantage but
also of equality. It was already mentioned by Stamma, but it is
beyond me why anyone would want to play this. It even feels
inferior to the Scandinavian with 3... d8, because here White can’t
play the solid e2-e3 anymore. Black’s simplest reply is 4... f6! (
4... c5 5. f3 f6 6. d3 occurred in the old game Horwitz-
Staunton, London 1851; after 6...d5! Black would have had the
more pleasant position) when the pawn on e4 is again more of a
problem than an asset: 5. c3 b4 6. d3 d5 is clearly better for
Black;
B) 4. d2?!, already briefly mentioned in Chapter 1, was dubbed
the ‘Bronstein Variation’ by Soszynski. White intends b2-b3 and
c1-b2 or developing by means of d3 and f4 first. But chess is a
concrete game, and it simply doesn’t work here: 4... f6 5. c3 b4!
(after the slow 5...d6 White might indeed try 6.b3!? g6?! 7. b2 ,
Popchev-Drazic, Novi Sad 2021) and Black is doing well already,
e.g. 6. d3 0-0 7. ge2 e8 (Bodescu-Pessi, Bucharest 2003) when
the pin of the knight on c3 is worse than the pin on the e-file after
4. d4-e3;
C) Finally, the curious 4. c3?? was played in no fewer than 56
database games, though never before the year 1993. One wonders
why anyone on earth would first play the sophisticated Center
Game, only to give up the queen a move later. What is perhaps even
more curious is that 4... b4 was played only 41 times out of those
56 (though many of them were online blitz games).

280
 position after 4. a4
This is called the Malmö or Warsaw Variation. It has been played
by strong players from time to time, but it has never been popular,
probably because the queen is offside on a4 and the pawn on e4 can
also become a target.
4... c5!
An obvious move, now that the white queen is so far away from the
battlefield. But it is only the second most popular move here. We
will briefly look at the two main alternatives:
A) 4... b4+ 5. d2 xd2+ ( 5... e7?! 6. c3 Velimirovic-
Todorovic, Yugoslavia 1988) 6. xd2 and in this unclear position,
6... f6!? is Black’s best try, though White can sacrifice a pawn with
7.0-0-0!? xf2 8.e5!? and Black’s position isn’t easy to play;
B) 4... f6 and now:
B1) 5. f4?! c5 (Kalinsky-Rubinstein, Kiev 1903) was good for
Black in the debut of the ‘Center Game in Viking Spirit’ (as Jeroen
Bosch called this line in Secrets of Opening Surprises vol. 13);
281
 B2) Bronstein once played 5. c3 in a simul game and produced a
classic opening trap with it: 5...d5?! (better was 5... b4! ) 6. g5
dxe4 7. xe4 e7 8.0-0-0 xe4?? (necessary was 8... d7 9. f3 ).

White to play and win

9. d8+! xd8 10. xe4 1-0 Bronstein-NN, Sochi 1959;
B3) The most popular move is 5. g5 and now Bologan gives
5...h6 6. h4 b4+!? 7.c3 e7 when Black is slightly better;
B4) 5. f3. Not the only move, as we’ve seen, but the safest one:
5... b4+ 6.c3 ( 6. d2 e7! ) 6... c5 7. d3 0-0 8.0-0 (Kurtovic-
Navratil, Novi Sad 1989), and now Black is better after 8... e8.
5. c3
5. f3 d6 6. c3 transposes. The careless 5. f4? f6! illustrates
how precarious White’s position already is. If the white queen had
been closer to home, this would not have been such an issue.
5...d6 6. f3 d7!
Very accurate! It’s best to chase away the queen immediately.

282
6... f6 7. g5 0-0?! 8. d5! would have been sloppy.

7. c4 ge7?!
I like 7... f6 much more. In the Center Game, this knight belongs
on f6.
8. g5?!
8. e2 0-0 9. e3 would have been better.
8...f6?!
8...h6! .
9. h4 b4?
What’s the bishop doing here?
After 9...a6, intending ...b7-b5, Black is doing great.
10.0-0-0 xc3
There weren’t many alternatives.
11. xc3 0-0 12. c4+?!
12.h3! and g2-g4, with an advantage.
12... h8 13. d4 xd4 14. xd4
283
The queen has ended up on d4 again. The position is about equal,
but White won eventually.
The next game is dedicated to the Hall Variation (4. c4).

On c4, the queen prevents ...d5 and still looks with one eye at the
kingside. It is often re-routed via e2, but there it still blocks the
bishop on f1 and knight on g1. I believe this line, while playable,
should mainly serve as a surprise weapon against unprepared
opponents. It might work well for White in a single game or in
games with faster time controls, but it won’t survive scrutiny at the
highest level. Black has too many relatively simple ways to get
equality – or more. (Interestingly, in the ‘reversed Center Game’ or
‘Elephant Gambit’ 1.e4 e5 2. f3 d5?! 3.exd5 xd5 4. c3 Black’s
best move is actually 4... c5, though White still has a good position
after 5.d4!. )
Daniel Dardha (2612)
Marani Rajendran Venkatesh (2452)
Cannes 2023

284
The young Belgian grandmaster Daniel Dardha regularly plays the
Center Game, which he learned from his father, Arben. Daniel wrote
to me: ‘My father played the move 4. e3 after 3... c6, but after
some analysis I realized that 4. c4 is perhaps an interesting option
from a practial point of view (there was also a game Wei Yi-
Karjakin, Jerusalem 2019 in which White won). [Black] has to make
a few strange and unnatural decisions, e.g. to play ...a7-a5 after a2-
a3 and leave the bishop hanging, and breaking in the center with
...d7-d5 at a specific moment.’
Daniel’s grandfather, Bardhyl Dardha, has also played the Center
Game, making it a family favourite over three generations.
1.e4 e5 2.d4 exd4 3. xd4 c6 4. c4 f6
Black plays analogously to the Berger, but White often doesn’t.
Some points to note are that the white queen often drops back to e2
in this line, and the white dark-squared bishop doesn’t always go to
d2.
A) 4...g6 is a worthy alternative here:
A1) 5. c3 g7 and now White again has plenty of choice:
A11) 6. f3 was played in Shabalov-Yermolinsky, US Senior
online rapid Championship 2020, and now 6...d6= is Black’s best
reply. There are no games in the database;
A12) 6.h4!? f6 (Sadhwani-Kopylov, Titled Tuesday blitz 2023)
7. e2 with unclear play;
A13) 6. d2?! is not very inspiring and Black may already be
better after 6... f6 7.0-0-0 0-0;
A14) 6. e3 d6 ( 6... ge7 7.0-0-0 d6 8.h4 Ponkratov-Chigaev,
PNWCC Bullet 2020) 7.0-0-0 e6 8. d3 f6=.
A2) White’s best may again be 5.h4!? as we also saw in the
Paulsen Fianchetto Variation.

285
 analysis diagram
Again, practical examples are few and far between. 5... f6 (the
strange 5... e7? 6. c3 b4 7. b5 was very bad for Black in Xu
Xiangyu-Romanov, Moscow rapid 2019) 6. c3 g7 7. e2, as seen
above, with an unclear position which I would be happy to play as
White.
B) 4... b4+ allows the extra option 5. d2!? ( 5. c3 f6
transposes to the game) 5... xd2+ 6. xd2 and now Black could try
6... f6!? 7.0-0-0 xf2, though it looks a bit shaky without an
engine to help out after 8. gf3, with obvious compensation.
5. c3
5.a3 would be a nice try to prevent ... b4, but Black has the
immediate 5...d5, equalizing.

286
5... b4
A) 5... e7!? is slightly less logical here than in the Paulsen
Variation, as the idea of playing ...d7-d5 quickly is not as dangerous
with the white queen on c4.
A1) Here, instead of the dull 6. d2, I like the almost untested
6. f4!? with a position that has a lot of room for exploration. A few
lines to test the waters: 6...d6 (the natural 6...0-0 is already a new
move here. Play might continue 7.0-0-0 a6!? (after 7...d6 I like
8.h3!?, intending 8... e6 9. e2 ) 8. f3 with an interesting
position that requires further research) 7.0-0-0 0-0 ( 7... e6 8. e2
0-0 9.f3 ) and now, again, 8.h3!? with an edge;
A2) 6. e3 is also not bad – we saw it already in Chapter 2 in the
bullet game Carlsen-Firouzja, 2021: 6...0-0 7.0-0-0. Now, Black
should have gone after the white queen immediately with 7...a6! (
7... g4 8. e2! and, contrary to the Paulsen Variation, the dark-
squared bishop isn’t missed much here. White was better after
8... xe3 9. xe3, Mieses-Rubinstein, Berlin 1918) 8. h3 b5 9. e2
d6 10. f4 e5 with chances for both sides;

287
 A3) 6. g5 was played three times by Daniel Dardha in 2022.
However, none of his opponents found the strongest move:

analysis diagram
A31) 6...0-0 7.0-0-0 Dardha-Demchenko, Titled Tuesday blitz
2022;
A32) 6...d6 7.0-0-0 (Dardha-Osmak, Cap d’Agde rapid 2022),
and now best was 7... e6! 8. b5 ( 8. d3!? ) 8...h6! ( 8...0-0?!
(Shabalov-Proleiko, US Online Swiss blitz championship 2020)
9.f4! ) 9. xb7 d7 with mutual chances;
A33) 6...d5! 7.exd5 xd5 8. xd5 ( 8. d1 xg5 9. xd5 0-0 is at
least equal for Black) 8... xg5 9. c5 and now simplest is 9... e7
10. xe7 xe7+ 11. xe7+ xe7 12.0-0-0 e6, which is equal.
B) 5...d6 is likely to transpose to lines mentioned above, but has
independent value after 6. f4 ( 6. g5 e7 transposes to Line A32
above) 6... e6 7. e2 and here, Black can either go for the
fianchetto set-up or try to castle queenside, or both. Let’s briefly
look at the concrete options:

288
 B1) 7... e7 8.0-0-0 0-0-0?! ( 8...g6 9. d2 ) 9. d5! xd5
10.exd5 xe2 11. xe2 e7 12.c4 and White is better;
B2) 7...g6 8.0-0-0 g7 9. d2! (an important little operation to
remember) 9... d7 10.f3 a6 ( 10...0-0-0 11. b5 ) 11. ge2 0-0-0
12. d5 and White has a small plus.
C) 5...d5!? is always a possibility to reckon with, and so it is here.

analysis diagram
This move was already played by Leonhardt against Mieses in 1920,
and also by Anna Muzychuk in 2023. White’s best choice here is the
never-played 6.exd5! with the possible continuation 6... b4 7. d3
fxd5 8. e4! e6 9.a3 with further complications ahead.
The problem with 6. xd5 is that it allows 6... e6! (better than
6... xd5 7.exd5 b4 and now 8.a3! ( 8. e4+?! e7 was played in
E.Pähtz-A.Muzychuk, Baku rapid 2023, and in Mieses-Leonhardt,
Berlin 1920) 8... xd5 9. f3 gives White a minimal advantage)
7. g5 xd5! 8. xd8 b4+ 9. d1 ( 9.c3? xc3! ) 9... e3+
10.fxe3 xc4=;

289
 D) 5... b4 6. e2 transposes to a line we’ve already seen in the
previous chapter, after 4. e3 b4 5. e2;
E) 5...g6 6. g5 g7 7.f4! ;
F) Finally, 5...a6!? is also interesting here, e.g. 6. f4 b5 7. e2
b4.

6. d2
Usually played without thinking, but we should check the
alternatives:
A) 6. d3?! is recommended by Soszynski, but it’s not so simple
after 6...d5! 7.exd5 xd5! when White struggles to equalize.
7... xd5 8. xd5 (Soszynski) is also slightly more pleasant for
Black after 8... xd5;
B) 6.a3 is a somewhat better try, but Black can’t complain here:
6... xc3+ 7. xc3 0-0, e.g. 8.f3 ( 8. g5? xe4 ) 8...d5 9. g5 dxe4
10. xf6 xf6 11. xf6 gxf6 12.fxe4 e8 with a slightly better
position for Black. This occurred in three (!) correspondence games
from 2018 in which Michel Aymard tried (and succeeded) to hold
this with White;
290
 C) The objectively best move seems to be 6. ge2!?. White would
like to go for a different set-up altogether, involving c1-g5. But
again, the move 6...d5! is a sound equalizer ( 6...d6 7. g5 h6 8. h4
e6 9. d3 c5 (Santagati-Basso, Brescia 2023) 10.0-0-0!; 6...0-0
7. g5 d5 ( 7... e7 8.h4!?; 7... e8 8.0-0-0 ) 8.exd5 xd5 9. xd5
xd5 10.0-0-0 xc3 11. xc3 xc3 12.bxc3 e6 13.a3 and
although White has the bishop pair for now, Black will be able to
hold easily after the accurate 13... a5! ): 7.exd5 xd5 8. d2 xc3
( 8... e6!? 9. d4= ) 9. xc3 xc3+ 10. xc3 0-0;
D) 6. g5?! has been played several times online by Olga
Prudnikova, but it isn’t great here: 6...h6 7. h4 g5 8. g3 d5! 9.exd5
xd5 .
6...0-0
6...d6 7.0-0-0 a6 8.f4 was seen in Gunsberg-Mortimer, London 1887
(see Chapter 1). Again, it’s funny how the theory of the Center
Game effortlessly jumps from the 19th to the 21st century and back!
7.0-0-0

7... e8
291
7...a5!? is a good and active move here, obstructing White’s dreams
of an easy game. The natural 8.f3 ( 8. f3!? d6 9.a3 – as usual,
Black can simply ignore this: 9... e8! and he is certainly not worse)
may seem risk-free, but it’s not, as Black continues to build on his
lead in development: 8... e5! 9. e2 e8 10. f2 c6! and White’s
attack is slower than Black’s: 11.g4 d5! 12.g5 d4 and although the
position is still very difficult, Black is objectively better.
8.f3
This move is exclusively played here. 8. ge2 b5! 9. xb5 b8
10. a4 c5! is not what White wants from the opening.
8...d6
Again, 8...a5 was a serious alternative, likely transposing to the note
on 7... e8.
8...a6 9.g4 was OK for White in Kosakowski-Grabarczyk, Poland
team championships 2020.

9. e2

292
A human move – the queen aims for the cosy spot f2 – but not the
best!
A) After 9.a3 Black must play very bravely in the spirit of
Dardha’s comment above: 9...a5!. It leads to very unclear play, e.g.
10. e2 e5 ( 10...d5!? ) 11. h3 ( 11.g4?! b5! with a dangerous
attack) with a very difficult position (Narciso Dublan-Valenzuela
Gomez, Barbera del Valles 2022), but it is easier to play with Black
after 11... xh3! 12.gxh3 c5! ;
B) 9. b1?! is a bit slow: 9... e6 10. e2 ( 10. d3?! d5?! was
fine for Black in Howell-Harsha, Titled Tuesday blitz 2021.
10... e5! was even stronger) 10...d5 when after 11.a3

analysis diagram
Black’s position is so good already that he can even afford a little
gallery play: 11... xa3!? 12.bxa3 b5! and in a practical game,
White’s position is not to be envied;
C) Magnus Carlsen tried 9.g4 in an online bullet game against
Andrew Tang, but it’s not a good move: 9... e6 10. e2 and now

293
simplest is 10...d5! ( 10... d4? 11. f2 Carlsen-Tang, lichess bullet
2021), e.g. 11.g5 d7 12.exd5 xd5 13. f2 b6 ;
D) The best move was 9. ge2 and now, again, Black can go for
active play directly: 9...a5! ( 9... e6 10. b5 occurred in Shuvalova-
Harika, Dubai 2023, and now Black should have played 10...a6!
11. xb7. The engine gives 0.00, but the game has only just begun)
10. f4 e5 11. e2 c6! when play is complex, but Black is
objectively fine.
9... e6?!
This move is often seen, but blocking the e-file prematurely is
almost never a great idea in this opening.
A) 9...d5! was the obvious move: 10.a3 f8 and Black is better
here (in the only game with this line, Bezeyko-Sandor, Brasov 2023,
Black erred with 10... a5? and White immediately profited by
means of 11. xd5 );
B) 9... e5!? is new and also good;
C) 9...a5 10.g4 was unclear in the game Dardha-Korneev, Sao
Joao da Madeira 2023;
D) Even 9... d7!? was better than the text move. After 10.g4
Black still has 10...d5 with an edge.
10. f2
As we saw in Chapter 2, White now has what he wants. Also
playable was 10.a3!? but, as usual, it comes with pros and cons:
10... a5 ( 10...a5 11. h3 ) 11. e1! unclear, Wei Yi-Bok, Pro
League Stage rapid 2023.
10... e5
A) 10...a5 11. ge2 was Wei Yi-Karjakin, Jerusalem 2019 (see
Chapter 2);
B) 10... e7 11.g4 ( 11. ge2! ) 11... ad8 12.g5 d7 13. ge2
with a lively game in Dardha-Fressinet, Samarkand 2023;

294
 C) 10... c5 11. e3 (not 11. g3?! (Dardha-Gozzoli, Cap d’Agde
rapid 2021) 11... h5! 12. e1 d5! ) 11... xe3+ 12. xe3 d5
13. ge2 .
11.g4 c4?!
Again, this move is not good here.
A) 11...b5!? 12.g5 c5 13. g2 leads to unclear play;
B) The surprising 11... fd7! was Black’s best move, heading for
the queenside. The position is complex and roughly equal, e.g.
12. h3 b6 13.a3 a5! (of course) 14. f4 c4 with a very difficult,
un-human position, where White’s best move according to the
engine is 15. fe2.
12. xc4 xc4 13.h4

Dardha had this nice (for White) position twice in one year, which
perhaps says something about the likelihood that it may appear
again.
13...c6?!

295
 A) 13... e7?! 14.h5 d7 15. h3 ( 15.h6! ) 15... e5? 16.g5
was Dardha-Lenaerts, Titled Tuesday blitz 2023 ( 16.h6 g6 17.f4 is
close to winning for White already);
B) Best was 13... c5! 14. e3 xe3+ 15. xe3 when White is only
slightly better.
14.h5 d5 15.g5
White’s position looks very attractive, but Black is holding on.
15... d7 16.h6 c5 17. e3 b6?
A bad move.
Mandatory was 17... xg5! 18.f4 e7 19. h3 xe3+ 20. xe3 g6 .
18. xc5 xc5
White now spoils a near-winning position.
19. h3?
19. d2! followed by b2-b3 or simply hxg7 was very strong.
19... ad8 20.exd5 cxd5 21.hxg7?!
21. d4 was slightly better for White.

296
21...d4!
With dangerous counterplay.
22. e4 xe4
22... xe4!? 23.fxe4 d3, followed by ... a5, was also a possibility.
23.fxe4 d3
Heading for an equal endgame.
24. xb6 axb6 25.b3 a6 26. f2 dxc2 27. xd8 xd8 28. h6
Now the game fizzles out to a draw, a result that was agreed upon
after 28 more moves.

The De Jouy Variation and the Vienna Hybrid

In the last three games of the book, we look at what in my view is
the most interesting (and surprising) new variation in the Center
Game: De Jouy’s old idea of playing the queen to d3 instead of e3.
In one sense, this is ‘just a move’: on d3, the queen isn’t placed
badly, but also not fantastically. It can still go to g3, but it blocks the
bishop on f1 and also stands in the way of the rook on d1 in the case
of 0-0-0.

297
 position after 4. d3
What does this remind you of? Another reversed Scandinavian! But
this time, it’s the one with the queen on d6 ( 1.e4 d5 2.exd5 xd5
3. c3 d6 – often referred to as the Tiviakov System, after the
Russian-Dutch grandmaster Sergei Tiviakov). With an extra tempo
compared to this, White would like to go for the same set-up by
playing the traditional 2.d4 followed by 3. xd4 and 4. d3, but
there is a slight problem as Black can immediately equalize with
4...d5!.

298
After White takes on d5, Black quickly plays ... b4 and wins back
the pawn, exchanging queens along the way and achieving an easy
endgame.
The modern way to reach the desired positions is to go 2. c3 first
and only after 2... f6 3.d4! exd4 4. xd4 c6 to drop the queen
back: 5. d3 .
(see diagram next page)

299
With a knight on c3, White can castle quickly in case of ...d5. Still,
things are not clear at all after, for instance, 5... c5, attacking f2 in
case White wants to castle. Perhaps more fundamentally, the move
2. c3 introduces another opening, namely the Vienna Game, which
requires White to prepare for moves like 2... c6 (stopping d2-d4) as
well.
There are so many ideas left to discover in the ‘Vienna Hybrid’ that
I can only encourage readers to follow the top players in the coming
months and years and also undertake their own original research.
Artem Bardyk (2238)
Melih Yurtseven (2333)
Titled Tuesday blitz 2024
1.e4 e5 2.d4 exd4 3. xd4 c6 4. d3

300
De Jouy’s move. On d3, the queen still eyes the d5-square, whilst
not stepping into a potential pin on the e-file, and flirts with ideas
such as g3. Moreover, the e3-square is now available to the
bishop. There is only one downside:
4...d5!
This move, not often played in practice, spoils the fun for White
here, as it gives Black an simple and equal game. Still, there are
some things to know for both players.
A) 4... f6 5. c3 transposes to the next game ( 5. f3?! c5! is
already slightly more pleasant for Black);
B) Let’s look at some positions in which Black plays an early
...d6: 4...d6. As usual, this is an unambitious but common and solid
set-up. Simplest is 5. c3 f6 (here, too, fianchetto set-ups are
possible: 5...g6 6. e3 g7 7.0-0-0 followed by f2-f3 with an easy
game for White. Because the dark-squared bishop is on e3 here, the
white queen can go to d2 and from there re-route to f2 and h4 if
needed, or threaten to exchange bishops on h6). Now, I like 6. f4 (
6. e3!? is virtually new but possibly even better. Some sample

301
lines: 6... e7 ( 6... e6 7.0-0-0 g4 8.f4 and White has a lot of
space and a pleasant position) 7.0-0-0 0-0 8.h3

analysis diagram
followed by d2 and g2-g4. This type of position is what makes the
4. d3 line rather attractive for White) 6... e7 7.0-0-0 0-0 8.h4 (
8.h3!? ) 8... e6 and now White should go for the space-grabbing
9.h5! with a pleasant advantage (unnecessarily slow was 9.a3?
e5 , Durarbayli-Shankland, St Louis blitz 2023). Familiar ideas
like g3 or f2-f3 and g2-g4 are in the air;
C) 4... c5 is also a good move here, of course. Now the simplest
is 5. c3 ( 5. g3!? looks nice but 5... f6! 6. c3 b4 is irritating.
See the note to move 6 in Game 45. There are many transpositions
here) 5... f6 6. f4, which transposes to a line we will examine in
the next game;
D) 4...g6 is actually dubious here on account of the surprisingly
strong 5. f3! ( 5. c3 is also possible, of course) 5... g7 6. g5!
when White already has a steady advantage, e.g. 6... ge7 7. c3 h6

302
8. e3 d6 9.0-0-0 when 9...0-0?! 10. d2! h7 11. d3 is simply bad
for Black.

5. xd5
A) I don’t like 5.exd5 b4 6. b3 xd5 when the endgame can
only be better for Black;
B) The same is true of 5. f3 dxe4 6. xe4+ e7;
C) 5. f4? was played in the game Savitha Shri-Ju Wenjun,
Kolkata blitz 2023, perhaps confusing the move with a possibility
for White after c3 and ... f6 have been included (see the next
game). But it failed to 5... f6! and Black already had a huge
advantage;
D) After 5. c3 Black can profit with 5... b4! 6. e2 d4 7.a3 dxc3
8.axb4 cxb2 ( 8... d4 ) 9. xb2 xb4+, which is not what White is
after.
5... xd5
5... e6? 6. xd8+ xd8 7. b5 doesn’t give enough compensation.
6.exd5 b4 7. b5+

303
The only move to keep the game going; if 7. a3 xd5 8. c4
gf6 .
7... d7 8. xd7+ xd7

This can be considered a tabiya if White insists on playing the

4. d3 variation without including c3 and ... f6. The endgame is
actually more interesting than it looks, and can become quite sharp –
but only Black can hope for an advantage, as he becomes active fast.
In the game, we will see how quickly White can stumble here.
9. d1
Better than 9. a3 e8+ 10. e3 ( 10. e2? xd5 ) 10... f6 (
10... c5!? ) 11. f3 ( 11.0-0-0 xa2+ ) 11... bxd5 12.0-0-0 c8
and Black has the better game.
9... xd5
Perhaps 9... f6!? is even more subtle ( 9... c5 10. f3 is unclear).
In the case of 10.c4 Black has 10... d3 11. h3 e8 ( 11... e4 is a
draw: 12. c2 b4+ 13. d1 d3 ) 12. c3 b4 and now the typical

304
computer line 13. c2 e1+! 14. b3 xc3 15. xc3 e4+ 16. b3
a5 when 17. g5! is the only move to keep the balance.
10. f3 d6
I think 10... d8 ( 10... e8 11. e1= )is Black’s best practical try. The
position is still equal but White needs to be accurate: 11.c4 (
11. d2!? c8 12. e1 c5 13. e5 d4 forces White to find
14. d3! f6= ) 11... b4 12.a3 c8+ 13. d2 c6 14. c2.
11. e2?!
The wrong direction.
Better was 11.c4 df6 12. c3 e8 13. c2 and the position is
balanced.
11... e8+ 12. f1 gf6

Now, Black is better, of course, due to his lead in development and

the poor position of the rook on h1.
13.c3 e7?!
13...a5! was a good move, grabbing space on the queenside just as if
it was a middlegame position in the Paulsen Variation.
305
14. g5 he8 15. bd2 f4 16. xf6 xf6 17.g3 xd2?!
Black should have kept pieces on the board: 17... h6! .
18. xd2?
A careless recapture. Stronger was 18. d1! and White’s troubles
would have been nearly over.
18... e2
Of course. Now Black is winning.
19. d1 c8 20. g2?
The inhuman 20. g1! g4 21. g2 would have been more tenacious.
20... g4
The rest of the game needs no commentary.
21. hf1 e3+ 22. f3 xd1 23. xd1 e1 ... 0-1 (54)
Georgios Souleidis (2419)
Vahe Baghdasaryan (2257)
Titled Tuesday blitz 2023
In this game, we will see how the German chess streamer Georgios
Souleidis, also know as ‘The Big Greek’, conducts one of the
critical and currently most popular lines in the Vienna/Center Game.
He recently published an online course on this line and told me in an
email: ‘For me [this] is just a blitz weapon but I think it can be
played also in classical games as a surprise weapon and for players
up to around 2200 regularly.’
1.e4 e5 2. c3!?
An attempt to get a favourable version of the De Jouy Variation,
which we saw in Chapter 2. Of couse, White actually plays a Vienna
Game now, so he needs to be prepared for sidelines on move 2.
2... f6
This move is by far the most popular against the Vienna.

306
White does need to have something against 2... c6, against which,
for the love of chess history, I will recommend the Steinitz Gambit:
3.f4 exf4 4.d4!? ( 4. f3 is the normal move) 4... h4+ 5. e2, which
is a lot of fun, not as bad as it was once thought, and has been
played by strong players such as Mamedyarov, Howell and
Durarbayli.
3.d4
Of course!
3...exd4
3... b4? is just bad: 4.dxe5 xe4 5. g4! and White wins.
3... c6 4.dxe5 xe5 allows White to transpose back with 5. d4!? (
5.f4! ) 5... c6 6. d3.
4. xd4 c6 5. d3

The starting position of the Vienna/De Jouy Hybrid System.

5... b4
This is the most popular move. Black naively assumes there is no
difference with the Paulsen Variation.
307
 A) Here, too, 5...d5 is possible. Play may become drawish if both
players know what they’re doing: 6. f4!? ( 6.exd5 is nothing much:
6... b4! 7. e2+ e7 was equal in Balog-Urkedal, Durrës 2023;
6. d2?? shows one of the downsides of having the queen on d3:
6... b4 and Black wins):

analysis diagram
A1) 6...dxe4 7. xd8+ xd8 8. xc7 b4 ( 8... e6!? 9. e5 with
an interesting, roughly equal endgame) 9.0-0-0 and White had a tiny
edge in Dubov-So, Lindores Abbey rapid 2020;
A2) Black’s best move is 6...d4, which leads to extremely messy
and rich positions after 7. b5 b4+ 8.c3 dxc3 ( 8... a5!? 9.0-0-0
e7! 10. xd4 xd4 11. xd4 b6! with mutual chances) 9. xd8+
( 9.bxc3!? ) 9... xd8 (Bellahcene-Erdös, Abu Dhabi blitz 2021),
when White has several attractive options, all leading to unclear
play. Least messy is 10.bxc3 a5 11.0-0-0+ d7 12.f3 Gavrilescu-
Erdös, Douglas 2021;
A3) Also possible is 6... b4 7.0-0-0 dxe4 ( 7...d4? 8. b5 occurred
in a recent game Pranav-Puranik, Dubai 2024. White was already
308
much better here. After 8...0-0, the most accurate move is 9. f3! )
8. xe4 and now Black must find 8... e7! ( 8... xd3? 9. xf6+ gxf6
10. xd3 with a clear advantage, Souleidis-J.Heinemann, Titled
Tuesday blitz 2023) 9. xf6+ xf6 10. e4+ e7 11. b5 with an
edge.
B) 5...d6 was examined in the previous game under 4...d6;
C) 5... e7 is less strong here than in the Paulsen Variation, e.g.
6. f4! 0-0 7.0-0-0 d6, which transposes to variations examined in
the previous game. White can either play 8.h3 or 8.h4!?, which was
tried in 2023 by Durarbayli;
D) 5... c5! will be the subject of the next and final game.
6. d2
6. f4!? is entirely new. After 6...0-0 White should first play 7. ge2
( 7.0-0-0? xc3 ), when 7...d5 8.0-0-0 xe4 9. xe4 dxe4 10. xe4
is interesting.
6...0-0 7.0-0-0 e8
7...d6 8. g3! transposes to lines in the Paulsen Variation where
Black plays ...d6, and which are fine for White (see the game
Morovic Fernandez-Garcia Padron, Las Palmas 1991, in Chapter 3).
Here, unlike in the Paulsen Variation where the queen is on e3,
White can play:
8. ge2!

309
This is the difference: Black can’t play ...d7-d5.
8... e5
The best move, chasing the queen away.
8...d6 9. g5! ( 9.h3 a5 – 9... b8! is the engine’s choice – 10.a3 c5
11.f4 b5 12. xb5 led to a wild game in Indjic-Rasulov, Titled
Tuesday blitz 2024) is unpleasant for Black.

310
 analysis diagram
White’s plan is simply to go f2-f4 and h2-h4 and start putting
pressure on the kingside. We will look at some sample games in this
very topical position:
A) 9... e6 10.f4 h6 11. h4 ( 11.h4!? was played by Souleidis in
an online game, and is also interesting) 11... g4 12. d5 ;
B) 9...h6 10. xf6 xf6 11. d5 d8 12. xb4 xb4 13. d2 a5
(less accurate is 13... c6 14.f3! D.Horvath-Tarasova, Titled
Tuesday blitz 2023) 14.a3 ( 14. c3 e6 15.a3 a2+! 16. xa2
xa2 Souleidis-Travadon, Titled Tuesday blitz 2023, when trapping
the bishop with 17.b3? ( 17. d3= ) is not good due to 17...a4
18. b2 axb3 19.cxb3 xb3! 20. xb3 xe4 with a dangerous
attack) 14... a6 15.f3!. The most accurate move. White had a slight
plus here in De Winter-Cnejev, Mamaia 2022; if 15. c3, 15... c5
is equal, Mamedyarov-Koneru, Chessable Masters rapid 2021.
9. g3

311
9... h5!
The only good move here.
A) 9...d6? 10. g5! is already pretty terrible for Black, e.g. 10...c6
11.f4 g6 12.a3! followed by e4-e5, and White is close to winning;
B) 9...c6 10. g5 is not much better;
C) After 9... c4, White again plays 10. g5.
10. g5
Forcing a trade of queens is usually not what White has in mind in
the Center Game, but there was no choice. In any case, the resulting
position is slightly easier for White to play in my view.
A) 10. e3 was the alternative. Black’s best is 10... f6, when
White can repeat or insert the moves 11.a3 f8 before going 12. g3
h5 13. g5, analogous to the game. There’s no great difference, I
think;
B) 10. h3!? looks risky but is playable, e.g. 10...g6 11.f4 (
11.g4!? ) 11...d5 12.f5.
10... xg5

312
10... f6? 11.f4! is horrible for Black.
11. xg5 h6 12. d2

12...c6
If Black plays 12... c4?!, then White has the nice retreat 13. e1 as
in the game.
13.h3!
Souleidis rightly remarks in his course on this line that White’s
position is slightly easier to play.
13... c4?!
Better was 13... f6 when Black is close to equality.
14. e1 c5?!
14... f6 15. f4 .
15. d4
The standard of play in this blitz game has so far been very high, but
now Black makes a big mistake:
15... d6?
313
15... e5 16. b3.
16.f3?!
16. b3! b4 17.f3! with a2-a3 to follow, and the knight on d6 is
close to being trapped in the middle of the board!
16... f4?
16...b5! provides an escape square for the knight on b7.
17.g3
17. g3! was winning.
17... e6 18. b3 e3+
Or 18... b4 19.a3 xc3 20. xc3 b5 21. xb5 cxb5 22. d5 .
19. b1
The knight on d6 is in dire straits.
19... b5 20. xb5 cxb5 21. xb5
White is winning. Souleidis didn’t give anything away and hauled in
the point on move 70.
Veselin Topalov (2727)
Viswanathan Anand (2751)
Leon Masters rapid 2024
1.e4 e5 2. c3 f6 3.d4 exd4 4. xd4 c6 5. d3 c5!

314
We already saw this logical move in Chapter 2. Souleidis even goes
so far as to wonder if this might be ‘the refutation’ of White’s set-
up. It’s not that grim, I’d hope, but this is the most principled move,
‘profiting’ from the fact that the queen is not on e3.
6. g3!?
I like this practical move, but it’s by no means compulsory.
A) The main alternative is 6. f4, when there are several
possibilities for both sides:
A1) 6...d6. As Souleidis remarks, it’s smart to postpone castling
kingside. 7. d2.

315
 analysis diagram
A recurring move in this line. White protects f2 before castling.
7... e6 (castling transposes to the lines below) 8. f3!?. I think this
is one of those situations where the move f3 is appropriate ( 8.0-0-
0?! b4 is annoying, Mamedyarov-Ding Liren, Chess.com Speed
Chess 2021; 8.f3!? is Souleidis’ recommendation. He continues the
line as follows: 8...d5! 9.0-0-0 dxe4 10. e1 e7! 11.fxe4 and here
11... b4! is Black’s best, since 12.a3 allows 12... xa3! and White
can’t capture the bishop. After 13. b5 b4 14. xc7+ xc7
15. xc7 xe1 16. xe1 0-0 it seems to me that the position is easier
to play for Black): 8... b4 ( 8...0-0 9.a3!? with equal and unclear
play in Vlasak-Laukola, ICCF email 2021) 9. d3 ( 9. g5!? ) 9...d5
10.exd5 xd5 11.0-0 xc3 12.bxc3 f6. A critical position which
occurred in Thusberg-Voveris, ICCF email 2021. Now 13. g3
xc3 14. xc3 xc3 15. xc7 is roughly equal. White has the pair
of bishops but Black has an annoying knight on c3. There’s still
plenty of play in the position;

316
 A2) 6...0-0 leads to similar positions with lots of possible
transpositions. A few examples: 7.0-0-0 ( 7.f3!? d5 is Mamedyarov-
Van Foreest, analysed in Chapter 2)

analysis diagram
A21) 7... g4!? 8. h3 xf2 9. xf2 xf2 10.e5! with interesting
compensation in Stefanova-Kashlinskaya, Bydgoszcz 2022;
A22) 7... xf2 8.e5 h5 9. d2 with very dangerous compensation
(Safarli-Milanovic, Titled Tuesday blitz 2022). A funny triple attack
would occur after 9... xe5? 10. e2!;
A23) 7... e8 8.f3 d6 9. d2 e6 10. b1 a6 11. ge2 b5
(Nepomniachtchi-So, St Louis online blitz 2020) 12.h4 ;
A24) 7...d6 8.f3 ( 8. d2!? ) 8... e6 9. d2 (Mamedyarov-Adams,
EU-Cup tt online rapid 2021) leads to a lively position in which
both sides have chances. Best now seems 9...a5 10. ge2 ( 10.g4 a4
Souleidis-Warmerdam, Titled Tuesday blitz 2023) 10...b5! 11. g3
b4 12. a4 with an unclear game.
B) One move to avoid is 6. g5??, which actually occurred in 7
games; it’s game over after 6... xf2+;
317
 C) 6. e3?! is not as bad as it looks but after 6... b4! ( 6... b6?
7.0-0-0 Abdrlauf-R.Svane, Titled Tuesday blitz 2023) 7. d2 xe3
8.fxe3 0-0 Black has nothing to complain about;
D) 6. d2?! looks like a blunder after 6... b4, but it’s not that
clear after 7. b5 when Black must be satisfied with ‘merely’ a good
position after 7... xc2+ 8. d1 xa1 9. xc5 d6 10. g5 e6!.

6...d6
Black has many alternatives here.
A) The untested move 6... b4! is also mentioned by Souleidis and
seems more in the spirit of Black’s active play. Now White must
either go in for a dubious gambit or accept an equal game:
A1) 7. c4!? may look interesting but Black should simply
develop: 7...0-0! ( 7... xc2+ 8. d1 xa1 9. xg7 f8 10. d5 is
less clear) 8. b3 d5 when Black’s position is preferable;
A2) 7.e5 also doesn’t work concretely: 7... h5 8. f3 xc2+
9. d1 xa1 10. xh5 d6! ;
A3) 7. d3 d5, e.g. 8. ge2 dxe4 9. xe4 xd3+ 10.cxd3 xe4
11.dxe4 0-0 and Black is certainly not worse.
318
 B) Interesting is 6... xe4!? 7. xe4 e7, which leads to an equal
but complex game after 8. e2 xe4 9. xg7 d4, Abdrlauf-
Krivonosov, Titled Tuesday blitz 2021;
C) After 6...0-0, the best move for White is 7. g5 when things are
unclear, e.g. 7... b4! ( 7. e3?! b4!; 7. h6 h5 8. g5 xg5
9. xg5 f6 10. d2= ) 8. d3 xc3+ 9.bxc3 d5 10.exd5 e8+
11. d2!? ( 11. e2 e5 Souleidis) 11... e5;
D) Also possible is 6... e7 7. e3!, intending 7... xe4? 8. xg7
f8 9. d5 and White wins;
E) 6...d5!? 7. g5! requires further investigation.
7. e3
7. xg7?? fails to 7... g8 8. h6 xf2+.
7. d2 g4 8. h3 h5 is very unclear.
7... xe3
Interesting and, again, more testing was 7... d4!? 8. d3 (less
accurate is 8.0-0-0 h5 9. g5 xg5 10. xg5 e6 , when f2 is also
hanging) 8... h5 9. g5 xg5 10. xg5=.
8. xe3 0-0 9.0-0-0

319
A natural position has arisen, usually seen with the dark-squared
bishops still on the board, which offers chances to both sides.
9... e8?!
This loses a bit of time.
Better was 9...a6! 10.f3 b5, intending ...b5-b4.
10.f3
White is a little better now, but Black also has opportunities on the
queenside.
10...a6
After 10...d5 White plays the typical 11. f2 followed by the usual
g2-g4.
Or 10... e5 11. ge2 c6 12. d4 a5 13.a3!, both with an edge for
White.
11. ge2
11.g4!?.
11...b5 12. f4?!

320
12.g4 would still have been pleasant for White.
12... e5 13. e2 b4 14. cd5 xd5 15. xd5 a5 16. b1 b7
17. d2
The direct 17.h4 was perhaps more to the point.
17... d7 18. f4 xd5 19. xd5 e7 20. hd1

Optically, White has a nice position, but things aren’t so bad for
Black yet.
20... e5?
Heading in the wrong direction.
20... b6 was still equal.
21. d2! h4?!
The queen is misplaced here. Now Topalov grabs the initiative.
21...h6 22.f4 d7 gave White just a small advantage.
22.f4! g6 23. b5! xe4 24. c6

321
Beautifully hitting both rooks.
24... a7 25. b5 e7 26. b8+ f8 27.g3
27. b5! .
27... h5 28. e1?!
Very strong was 28.a4!, creating some air for the king. The idea is
28...bxa3 and only now 29. e1! xe1+ 30. xe1 g6 31. e7 and
there is no mate on d1; White wins.
28... e6?
28... xe1+ 29. xe1 g6 was the only defence, though White is still
much better after 30.a4.
29. d7?
29.a4! was again the right move, winning.
29... xe1+ 30. xe1 g6
This is the difference. Now White can’t move the queen on account
of mate on d1.
31.b3 g7

322
Anand has managed to equalize, but things soon get messy again.
32. b5 e6 33. c4 d5 34.g4?
34. d3=.
34... xg4 35. xd5 xf4?
35... a6! .
36. e5+ h6 37. xf7 c5?!
37... g1+ 38. b2 xh2 .
38. c4 g5 39. e8 xe5 40. xe5 c7
Black is still somewhat better, but Topalov holds the draw:
41.a3 h5 42.axb4 axb4 43.c3 bxc3 44. c2 f6 45.h3 h5
46. xc3 f4 47. f1 d7 48. xc5 d1 49. c4 xh3 50.b4 f2
51. c7 e4+ 52. c2 d4 53.b5 d6 54. d3 b4 55. c5 e4
56. c6 g5 57.b6 f6 58. d2 d7 59. c7 e5 60. c5 xb6
Draw.

Third and second move alternatives

323
In closing the theoretical part, let me say a few words about second
and third move alternatives for Black. My recommendation is to not
lose any sleep over all these obscure sidelines, which occur rarely
anyway.
The move 3... f6 is sometimes seen and was once played by
Chigorin.

White can play either 4. e3 or 4. c4 (targeting c7) and can be

more than happy in both cases. There is no theory to learn here.
Black may try to exchange queens, but he will always be worse due
to White’s lead in development. I once faced 4. e3 b6 5. g3
g6 when White can simply develop: 6. c3, with a big plus. After
reading this book, I trust you’ll have a good idea of where the pieces
need to go, and that’s all you need to know.
The only other real alternative is 3...c5 .

324
Here, it’s important to know the move 4. e5+! , which practically
forces another miserable endgame for Black: he has many
weaknesses and White has easy development with c3, f4, 0-0-0,
etc. There are no other serious third moves for Black except 3... f6
and 3...d6 (which usually transpose).
A final word of caution: if you plan to play the Paulsen Variation,
you should never pre-move 4. e3 in online bullet games since
Black can try the sly 3... c5??! , which I have stupidly fallen victim
to quite a few times.
Alternatives on move 2 are more commonly seen. I won’t go into
detail on these as I consider them to be different openings. Life is
tough, and we can’t always get a Center Game on the board,
unfortunately.
The first possibility is 2...d6 .

325
White can still try 3. c3, hoping for a delayed capture on d4, but I
recommend the simple 3. f3 , transposing to the Philidor Defence,
which is theoretically good for White. A nice bonus option after
3...exd4 is the move 4. xd4!? . Why not try to get some kind of
Center Game after all?
Then there is 2.... c6 .

326
There are several good options here. If you like closed games, 3.d5
is your answer. Good luck with that. It’s more likely a 1.e4 player
wants to get an open game and plays 3.dxe5 xe5 . If you want a
Center Game at all costs, then 4. d4 gets you there, as Black has
nothing better than to retreat the knight to c6 at some point. If he
doesn’t, White may play f2-f4, chasing the knight away, and redirect
the queen to the kingside via the usual routes. And if the knight goes
to g6, it will likely be irritated by a timely h2-h4. There is also
4.f4!? and other fourth moves, all of which promise White a
pleasant edge.
Finally, the move 2... f6 .

327
This allows a transposition to the last game in this book after 3. c3,
but White has even better than that: 3.dxe5 xe4 4. e2! leads to a
big plus already, e.g. 4... c5 5. c3 c6 6. e3 e6 7.0-0-0.
Black can only avoid the Center Game at a cost.

328
Epilogue
The moment you doubt whether you can fly, you cease for ever to
be able to do it.
J.M.Barrie, Peter Pan

Before writing this book, I thought I understood the Center Game. I

just needed to organize my preparation files and write down my
thoughts. After all, I already knew quite a bit about its historical
past. I had played it for almost thirty years, I had looked up the
latest theory and I had even analysed it with world-class players.
But less than halfway through, I realized that things were not so
simple. I discovered that many of my beloved, well-scoring lines
were stale and outdated, and countless ideas I had so far neglected
were looking more and more attractive. Some of my older views –
safe, straight and narrow – still proved valid, especially in the
Paulsen Variation:
• For Black, the move ... c8-e6 is usually not a great idea;
• Likewise for White, f1-e2 is typically not a move you want to
play;
• If Black doesn’t go for ...d5 or activity on the queenside, White is
fine;
• White should be careful with the move f2-f4 whenever e4 is weak.

These points were still true, but many others were new to me:
• It’s often not so bad for White to exchange queens;
• The move g1-f3 (with a pawn on f2) should be considered more
often by White;
• Early ... f8-d6 moves are dangerous and not to be underestimated;
• 5.h4 is White’s most promising move against the fianchetto;

329
• The Hall Variation (4. c4) is less dangerous than I thought;
• The ‘Vienna move-order’ is important if White wants to play the
d3 system.

Overall, the opening seemed much more concrete and less ‘logical’
than I’d previously thought. A lot of novelties and lines that looked
hopeful just a few years or decades ago have now, thanks to the
relentless precision of modern engines, faded into the mists of chess
opening history. Many others have appeared, waiting to be tested in
practice.
Studying the opening’s distant and recent past was a true joy for me.
I found that many early pioneers of the Center Game understood the
opening very deeply, while many respected grandmasters who had
expressed strong opinions about it, didn’t know what they were
talking about. Quite a few key ideas that are being employed by the
strongest players today had already been discovered and analysed in
the 19th and early 20th centuries. I have always felt that the old
masters’ understanding of the key principles of opening play was
undervalued. I hope to have demonstrated that some of the lesser-
known players from previous centuries also had an excellent
mastery of the dynamics of the Center Game. Not all players from
the past were swayed by stereotypes of playing their queen too soon.
Not everyone felt ashamed about neglecting their ‘normal’ piece
development.

As a practical player, I am not sure how to deal with my newly

gained knowledge. Many of my preconceptions have been
overturned. All the logic, order and structure seem to be gone.
Rationale, rhyme and reason – pale beside a single engine move. I
can’t just continue to play lines I now know to be dubious, can I?
I’m willing to, but I’m also afraid: if I try out the lines and novelties
I found to be the most promising and sound, I will lose many of the

330
benefits of routine and experience I’ve built up over the years. I
might as well learn a different opening.

While writing the last sentences of this book, I’m listening to a

playlist of The Sisters of Mercy. A fragment: layered synths, a
simple, haunting bass line, a heavy beat like reverberating
clockwork in an immeasurable space. An echo, and Andew Eldritch
sings, almost breathes:
I had a face on the mirror
I had a hand on the gun
I had a place in the sun and a
Ticket to Syria...

As I think of Philipp Stamma, the mysterious nobleman from Syria

who came to Paris and London in the 18th century and was the first
to analyse the sequence 1.e4 e5 2.d4 exd4 3. xd4, I wonder what he
would make of all these new developments in the Center Game, his
own invention. Would he understand the challenges, the
complexities, and the countless new concepts? Would he be a critic
or a fan?
In an article from 1959 titled ‘Prelogical chess’, the Dutch
grandmaster and writer J.H.Donner wrote about Mikhail Tal’s style:
‘There is no system, there is no correct or incorrect, there is only
success.’
Acceptance is part of growing up.

I wish all readers of this book lots of success with the Center Game.

331
Bibliography
Books and magazines
Jimmy Adams (ed.), Mikhail Chigorin – The Creative Genius, New
in Chess 2016
Alexander Alekhine, Alexander Alekhine’s Best Games Algebraic
Edition, Batsford 1996
Johann Berger, Deutsche Schachzeitung Nr.2 Februar 1884
Victor Bologan, Bologan’s Black Weapons, New in Chess 2014
Jeroen Bosch, Secrets of Opening Surprises 12, New in Chess 2010
Jeroen Bosch, Secrets of Opening Surprises 13, New in Chess 2011
David Bronstein, 200 Open Games, B.T. Batsford Ltd 1973
Lewis Carroll, The Annotated Alice (ed. Martin Gardner), Penguin
1970
Nigel Davies, Play 1.e4 e5!, Everyman Chess 2005
Jan Hein Donner, De Koning – Schaakstukken, Bert Bakker 1987
John Emms, Play the Open Games as Black, Gambit Publications
2000
Jakob Estrin, Kleine Schach-Eroeffnungs-Enzyklopaedie, Thomas
Beyer Verlag 1986
Dr. Max Euwe, Theorie der Schaakopeningen No. 12, G.B. van
Goor 1939
Stephen W. Gordon, Samuel Reshevsky: a compendium of 1768
games, McFarland & Company, Inc. Publishers 1997
Andrew Greet, Dangerous Weapons: 1.e4 e5, Everyman Chess 2008
Frank Hoffmeister, Chess Theory from Stamma to Steinitz, 1735-
1894, McFarland & Company, Inc. Publishers 2022
Garry Kasparov, My Great Predecessors, Everyman Chess 2003

332
Garry Kasparov & Raymond Keene, Batsford Chess Openings 2,
B.T. Batsford Ltd 1989
Daniel King, Sultan Khan, New in Chess 2020
Yuriy Krykun, A Complete Repertoire for Black after 1.e4 e5!,
Thinkers Publishing 2020
David Levy & Kevin O’Connell (ed.), Oxford Encyclopedia of
Chess Games, Ishi Press International 1981
William Lewis, Stamma on the Game of Chess, T. and J. Allman
1819
William Lewis, A Treatise on the Game of Chess, A.H. Baily and
Co 1844
Mihail Marin, Beating the Open Games, Quality Chess 2008
Aleksandar Matanovic, Encyclopedia of Chess Openings, 4th
edition, Sahovski Informator, Beograd 2000
Arne Moll, ‘Verstärkungen im Mittelgambit’, Kaissiber 9 1999
Peter J. Monté, The Classical Era of Modern Chess, McFarland &
Company, Inc. Publishers 2014
Harold J.R. Murray, A History of Chess, Skyhorse Publishing Books
2012
New in Chess Magazine 4, New in Chess 1992
John Nunn, Nunn’s Chess Openings, Gambit Publications 1999
Matthew Sadler & Natasha Regan, Chess for Life, Gambit
Publications 2016
Matthew Sadler & Natasha Regan, Game Changer, New in Chess
2019
Philip W. Sergeant, Morphy’s Games of Chess, Dover Publications
1957
Sergei Soloviev, Boris Spassky’s 400 Selected Games, Chess Stars
Ltd. 2003
Marek Soszynski, The Centre Game Re-examined, MarekMedia
2020

333
Alexander Suetin, Russisch bis Köningsgambit, Sportverlag Berlin
1988
Mikhail Tal, The Life and Games of Mikhail Tal, Everyman Chess
Classics 1997
Dr. Siegbert Tarrasch, Dreihundert Schachpartien, Verlag von Veit
& Comp. 1909
Paul van der Sterren, New in Chess Yearbook 34, New in Chess
1994
Paul Rudolf von Bilguer (v.d. Lasa), Handbuch des Schachspiels,
Verlag von Veit & Comp. 1916
Carl Friedrich von Jaenisch, Jaenisch’s Chess Preceptor, Longman,
Bow, Green and Longmans 1847
John Watson, Secrets of Modern Chess Strategy, Gambit
Publications 1998
John Watson, Chess Strategy in Action, Gambit Publications 2003
Mario Ziegler, Paulsen-Eröffnung, Verlag ChessCoach 2010
Chess Informants 14 (1972), 61 (1994), 62 (1994), 63 (1995), 70
(1997), 83 (2002), 100 (2008), 105 (2009) and 112 (2011).

Internet
Michael Gorny, The Daring Center Game, Chessable 2023
Arne Moll, Finding Nepo (on an old laptop), Chessvibes.com 2009
Arne Moll, Morphy and Me, Chess.com website 2017
Arne Moll, The Indian Roots of Modern Chess, Chess.com website
2018
Arne Moll, Why you should move your Queen as early as possible!
Chess.com 2019
Georgios Souleidis, Moderne Wiener Partie, Chessemy (online)
2024
The Sisters of Mercy, www.the-sisters-of-
mercy.com/lyrics/lyrindex.htm
Wikipedia
334
Programs
ChessBase, chessbase.com website
Chess Cloud Database Query Interface, chessdb.cn website
Leela Chess Zero v0.30, lczero.org website
Stockfish 15 & 16, stockfishchess.org website

335