Definition of Game Theory

Game theory is a mathematical framework used to model and analyze

situations in which multiple individuals or groups (called "players") make
decisions that affect the outcomes of the game. The theory is particularly
focused on interactions where the results depend not only on a player's own
decisions but also on the choices made by others. These decisions may involve
competition, cooperation, or a mix of both. The goal of game theory is to
understand how players will act or should act in various strategic settings, and
to predict the outcome of these interactions.
Key components of game theory include:

 Players: The decision-makers in the game.

 Strategies: The possible actions each player can take.
 Payoffs: The rewards or penalties players receive based on their actions
and the actions of others.
 Information: What players know when making decisions, which can
range from complete knowledge to imperfect information.
 Equilibrium: A stable state in which no player has an incentive to change
their strategy unilaterally (e.g., Nash Equilibrium).
Game theory can be applied to a variety of fields including economics, political
science, biology, and psychology, providing insights into how rational actors
make decisions in interactive environments.

History of Game Theory

The development of game theory can be traced back to several key historical
milestones and thinkers:
1. Early Foundations:
o Ancient Origins: The earliest traces of game theory concepts can
be found in ancient philosophy, where thinkers like Plato and
Aristotle explored strategic reasoning, especially in the context of
warfare and political theory. However, these early ideas were not
formalized into a mathematical framework.
2. John von Neumann and Oskar Morgenstern (1944):
o "Theory of Games and Economic Behavior": The formal
foundation of game theory is often credited to Hungarian
mathematician John von Neumann and economist Oskar
Morgenstern. In 1944, they published their landmark book
Theory of Games and Economic Behavior, which established the
field of game theory as a formal discipline. The book introduced
the concept of "games of strategy," where players' actions depend
on the decisions of others. It applied game theory to economics,
particularly in understanding competitive behavior in markets and
the theory of bargaining.
3. Nash Equilibrium (1950s):
o The most significant advancement in game theory came with John
Nash, an American mathematician, in the early 1950s. Nash
introduced the concept of the Nash Equilibrium, a solution
concept in non-cooperative games. A Nash Equilibrium occurs
when no player can improve their payoff by unilaterally changing
their strategy, given the strategies of the other players. Nash's
groundbreaking work earned him the Nobel Prize in Economic
Sciences in 1994.

Key Historical Milestones in Game Theory:

1. 1944: Publication of Theory of Games and Economic Behavior by von
Neumann and Morgenstern, establishing the foundation of game theory.
2. 1950s: John Nash develops the concept of Nash Equilibrium,
revolutionizing non-cooperative game theory.
3. 1970s-1980s: Emergence of evolutionary game theory, applying game
theory to biology and evolutionary processes.
4. 1990s-2000s: Expansion of game theory into computer science,
economics, political science, and the study of auctions and bargaining .
1.2 Basic concepts
Game theory is built on several fundamental concepts that help in analyzing
strategic interactions between rational decision-makers (players). Here are the
basic concepts of game theory:
1. Players
 Players are the decision-makers in the game. These can be individuals,
organizations, firms, countries, or any entities that participate in the
game. Each player is assumed to act rationally, making decisions to
maximize their own payoff or utility.
2. Strategies
A strategy is a plan of action that a player chooses to follow in the game. The
strategy may be a single action or a sequence of actions that a player may take
throughout the game. The choice of strategy can depend on various factors,
such as the strategies chosen by other players.
 Pure Strategy: A player chooses a single action with certainty.
 Mixed Strategy: A player randomizes between different actions, each
with a certain probability.
3. Payoffs
The payoff is the reward or outcome a player receives as a result of the
strategies chosen by all players involved in the game. Payoffs can be
represented in a payoff matrix or a function that depends on the strategies
chosen. The payoff can include tangible rewards like money or abstract values
like satisfaction or utility.
4. Games
A "game" in game theory refers to the structured scenario where players make
decisions that impact each other's outcomes. Games can be classified into
different types:
 Cooperative Games: Players can form binding agreements and work
together to achieve mutually beneficial outcomes.
 Non-Cooperative Games: Players cannot form binding agreements and
must act in their self-interest.
  Zero-Sum Games: One player's gain is exactly balanced by the other
player's loss (e.g., chess, poker).
 Non-Zero-Sum Games: The total payoff is not constant; both players may
gain or lose together (e.g., trade negotiations, environmental
cooperation).
5. Nash Equilibrium
 The Nash Equilibrium is one of the central concepts in game theory,
developed by John Nash. It occurs when each player in the game has
chosen a strategy, and no player can benefit by unilaterally changing
their own strategy, assuming the strategies of the other players remain
the same. In other words, every player's strategy is optimal given the
strategies of the others. It is a state where players' strategies are
mutually best responses to each other.
6. Dominant Strategy
 A dominant strategy is a strategy that provides a player with the highest
payoff, no matter what the other players do. If a player has a dominant
strategy, they will always choose it because it guarantees the best
possible outcome, regardless of the actions of others.
8. Pareto Efficiency
 A situation is Pareto efficient if no player can be made better off without
making someone else worse off. In other words, it’s a state where
resources or payoffs are allocated in such a way that no further
improvements are possible for any player without harming another.
In game theory, games can be classified based on various factors such as the number of players, the
type of strategies available, the availability of information, and the nature of the interactions
between players. Here are the main types of games in game theory:

1.3 Types of games

1.3.1. Cooperative vs. Non-Cooperative Games
 Cooperative Games: In these games, players can form binding
agreements, coalitions, or partnerships to improve their collective
outcomes. Players are allowed to negotiate and coordinate strategies for
mutual benefit. The focus is on how to distribute the collective payoff
 fairly among players. Examples include joint ventures or collective
bargaining scenarios.
o Example: A group of firms collaborating to set prices in a market.
 Non-Cooperative Games: In these games, each player acts
independently and cannot form binding agreements with others. Players
are assumed to act in their own self-interest to maximize their individual
payoff, with no external coordination. This type of game is more
common in game theory.
o Example: Two companies competing in an oligopoly, each acting in
their own self-interest.
1.3.2. Symmetric vs. Asymmetric Games
 Symmetric Games: In these games, all players have identical strategies
and payoffs available to them. The game is symmetric if the identity of
the players does not affect the outcome; all players are in a similar
position.
o Example: Chess, where both players have the same pieces and the
same set of strategies.
 Asymmetric Games: In these games, players have different strategies
and/or payoffs available to them. The players may be in different
positions, and the strategies or payoffs are not identical.
o Example: A market with a monopoly and competing firms, where
the monopoly has more influence over pricing.
1.3.3 Zero-Sum vs. Non-Zero-Sum Games
 Zero-Sum Games: In zero-sum games, the total payoff remains constant,
meaning one player's gain is exactly balanced by another player's loss.
The sum of all players' payoffs equals zero. Zero-sum games are often
associated with competition, where players directly oppose each other.
o Example: Poker, where one player’s winnings are equal to the
losses of others.
 Non-Zero-Sum Games: In non-zero-sum games, the total payoff is not
fixed. It is possible for all players to gain (cooperate) or for all players to
 lose. These games are often used to model situations where cooperation
can lead to better overall outcomes.
o Example: Trade negotiations, where both countries can benefit
from agreeing on trade terms.
1.3.4 Simultaneous vs. Sequential Games
 Simultaneous Games: In simultaneous games, all players make their
decisions at the same time without knowing the choices of the others.
Each player must anticipate the actions of others and choose their
strategy accordingly.
o Example: The Prisoner's Dilemma is a classic simultaneous game
where both players make their decisions (cooperate or defect)
without knowing what the other player will do.
 Sequential Games: In sequential games, players make their decisions
one after another, and each player is aware of the previous players'
choices before making their own decision. These games are typically
represented in game trees to show the sequence of actions.
o Example: Chess, where one player moves first, followed by the
second player, and so on.
1.3.5 Perfect Information vs. Imperfect Information Games
 Perfect Information Games: In perfect information games, all players
know the complete history of the game and have access to all the
information about the game state. Players can make decisions based on
the full knowledge of the game’s progress.
o Example: Chess, where all pieces and moves are visible to both
players.
 Imperfect Information Games: In imperfect information games, players
do not have complete knowledge about the game state or the actions of
other players. Some information is hidden, and players must make
decisions under uncertainty.
o Example: Poker, where players do not know each other's cards.
1.3.6 Continuous vs. Discrete Games
 Continuous Games: In continuous games, players choose strategies from
a continuous set of options, typically involving real numbers. The
decision-making process can be thought of as choosing a point in a
continuous space.
o Example: Auction bidding, where the amount one bids can vary
continuously.
 Discrete Games: In discrete games, players have a finite set of strategies
to choose from. The strategy options are limited to a discrete set of
choices.
o Example: Rock, Paper, Scissors, where players choose from three
distinct options.
Conclusion
These various types of games provide frameworks for analyzing different
strategic situations and interactions. The classification helps in understanding
how different types of decision-making dynamics unfold and which strategies
are most appropriate under different conditions. Game theory is widely
applicable to a range of real-world problems, from economics and politics to
biology and artificial intelligence.
CHAPTER 2- STRATEGIC
FORM GAMES

2.1 NASH EQUILIBRIUM

2.1.1 DIFINITION
A Nash equilibrium is a concept in game theory where no player can improve
their own outcome by changing their strategy, assuming the other players'
strategies remain unchanged. In other words, it represents a situation in which
each player's strategy is optimal given the strategies of all other players.
In a Nash equilibrium, each player's decision is the best response to the
strategies of others, meaning that no player has an incentive to unilaterally
deviate from their chosen strategy. This can apply to both cooperative and non-
cooperative games.
For example, in a two-player game, if both players are playing their best
strategies in response to each other, and neither player can improve their
outcome by changing their strategy alone, the game is in Nash equilibri

2.1.2 Example of Nash equilibrium in different types of games:

1. Prisoner's Dilemma (Non-cooperative game)
Game setup: Two criminals are arrested and offered the following choices:
 Cooperate (stay silent)
 Defect (betray the other)
Payoff matrix:
Prisoner B Cooperates Prisoner B Defects
Prisoner A Cooperates 3 years, 3 years 10 years, 0 years
Prisoner A Defects 0 years, 10 years 5 years, 5 years
Nash equilibrium: Both prisoners choose to defect, as defecting provides a
better payoff regardless of what the other player does. Even though
cooperating would lead to a better overall outcome (both get 3 years), each
prisoner has an incentive to betray the other to get a lighter sentence (0 years
instead of 3).
2.1.3 The Battle of the Sexes (Coordination game)
Game setup: A couple wants to go out but have different preferences for
events. One prefers the opera, while the other prefers a football game. They
both want to spend time together, so they want to coordinate, but each has a
different idea of where to go.
Payoff matrix:
Opera (B) Football (B)
Opera (A) 2, 1 0, 0
Football (A) 0, 0 1, 2
Nash Equilibrium:
 There are two Nash Equilibria in this game:
o (Opera, Opera): Player 1 goes to the opera, and Player 2 goes to
the opera.
o (Football, Football): Player 1 goes to the football game, and Player
2 goes to the football game.
 Both players coordinate on one of the activities, and neither player has
an incentive to deviate once they coordinate on a strategy.

Applications of Nash Equilibrium across different fields and scenarios:

1. Pricing Strategies in Oligopolistic Markets (Economics)
 Scenario: Multiple firms compete in a market by setting prices for their
products. The price set by each firm depends on the pricing decisions of
the other firms in the market.
 Nash Equilibrium Application: In this context, Nash Equilibrium helps
predict the prices that firms will charge when each firm’s pricing strategy
is optimal given the strategies of the other firms. In an oligopoly, firms
tend to set prices where no firm can improve its profit by unilaterally
changing its price.
  Example: Two competing airlines setting ticket prices for similar routes. A
Nash Equilibrium occurs when neither airline can profitably change their
price without being undercut by the other.
2. Traffic Flow and Route Selection (Transportation)
 Scenario: A network of roads where multiple drivers choose their routes
to minimize their travel time. However, the travel time on a route
depends on the number of drivers choosing that route.
 Nash Equilibrium Application: In this case, Nash Equilibrium can predict
the flow of traffic when no driver can reduce their travel time by
changing their route. This is called a Wardrop equilibrium in
transportation theory.
 Example: In a city with multiple routes to a destination, the Nash
Equilibrium would occur when drivers distribute themselves across the
routes such that no one can improve their travel time by switching
routes.
3. Voting and Political Strategy (Political Science)
 Scenario: In an election, political candidates choose their policies (or
strategies), and voters choose their preferred candidate. The candidates'
strategies depend on how they expect voters to behave, and voters base
their decisions on the strategies of the candidates.
 Nash Equilibrium Application: Nash Equilibrium helps in analyzing how
candidates will set their platforms and how voters will vote based on
those platforms. The equilibrium occurs when no candidate can change
their policy to gain more votes, and no voter can switch their vote to
improve the outcome for themselves.
 Example: Two political parties running for office, where each party's
platform is chosen based on what they think will attract the most voters.
The Nash Equilibrium is reached when neither party can adjust their
platform to improve their chances of winning without losing votes to the
other party.
4. Game Theory in Auction Design (Economics/Computing)
 Scenario: Auctions are a common mechanism for selling goods or
services where multiple bidders compete for a good. Bidders must
decide how much to bid based on the strategies of other bidders.
 Nash Equilibrium Application: In sealed-bid or second-price auctions,
Nash Equilibrium helps predict the optimal bidding strategies. In a
second-price auction, a Nash Equilibrium is reached when each bidder
bids their true value for the good.
 Example: In a sealed-bid auction for a piece of art, the Nash Equilibrium
would occur when each bidder bids an amount equal to their true value
of the art, assuming they know the other bidders' strategies.

2.2 Mixed Strategies

2.2.1 Definition:
A mixed strategy is a strategy in which a player randomly chooses among
available actions according to a specific probability distribution. Unlike a
pure strategy (where a player always chooses the same action), a mixed
strategy allows a player to randomize their decisions, assigning
probabilities to each possible action.
In a mixed strategy Nash Equilibrium (MSNE), players choose strategies
in such a way that no player can improve their expected payoff by
unilaterally changing their strategy mix.
Why Mixed Strategies?
 Mixed strategies are used when no pure strategy is optimal or when
players are trying to make their behavior unpredictable to other players.
 They are particularly useful in games where a player’s strategy might be
countered if it’s predictable, so randomization can add a level of
unpredictability.

Example:
 certain probabilities. Let's say:
Player 1 chooses Heads with Let’s consider a simple game called
Matching Pennies between two players, Player 1 and Player 2:
 Player 1 can choose either Heads (H) or Tails (T).
 Player 2 can also choose either Heads (H) or Tails (T).
The payoffs are as follows:
 If both players choose the same side (both Heads or both Tails), Player 1
wins 1 point, and Player 2 loses 1 point.
 If the players choose opposite sides (one chooses Heads, and the other
chooses Tails), Player 2 wins 1 point, and Player 1 loses 1 point.
Payoff Matrix:
Player 2: Heads Player 2: Tails
Player 1: Heads (1, -1) (-1, 1)
Player 1: Tails (-1, 1) (1, -1)
Pure Strategy:
 If Player 1 always chooses Heads, Player 2 can simply always choose Tails
to win.
 Similarly, if Player 2 always chooses Heads, Player 1 can always choose
Tails to win.
Thus, there is no stable outcome using pure strategies alone, as both
players would try to outguess the other.
Mixed Strategy:
o Player 1 and Player 2 will each randomize their choices, choosing
Heads or Tails with probability pp and Tails with probability 1−p1-
p.
o Player 2 chooses Heads with probability qq and Tails with
probability 1−q1-q.
 In this mixed strategy scenario, each player is making their choice in such
a way that the other player is indifferent to choosing either Heads or
Tails. This happens when the expected payoffs from each of the players’
actions are equal, making the opponent's decision effectively random.

Equilibrium:
 The mixed strategy Nash Equilibrium occurs when:
o Player 1 is indifferent between choosing Heads and Tails (the
expected payoff for choosing each is the same).
o Player 2 is indifferent between choosing Heads and Tails.
By solving the equations, we find that:
 Player 1 should choose Heads with probability p=0.5p = 0.5, and Tails
with probability 1−p=0.51-p = 0.5.
 Player 2 should choose Heads with probability q=0.5q = 0.5, and Tails
with probability 1−q=0.51-q = 0.5.
Thus, in this equilibrium, both players randomize their strategies, each
with a 50% chance of choosing Heads or Tails.

2.2.2 Application of Mixed Strategies:

Mixed strategies are widely used in various fields, particularly in
competitive environments or situations where players interact multiple
times. Here are some notable applications:
1. Sports (Penalty Kicks in Soccer)
o In penalty kick situations, both the kicker and the goalkeeper have
multiple choices: the kicker can aim for the left, center, or right of
the goal, and the goalkeeper can choose to dive left, center, or
right.
o If the kicker always aims for the same spot, the goalkeeper can
predict and counter the shot. Using a mixed strategy, the kicker
randomizes their shots (choosing the direction with some
probability), and the goalkeeper randomizes their dives to make
themselves unpredictable.
2. Auction Bidding
o In competitive bidding, such as in second-price auctions, bidders
may use mixed strategies by submitting bids that depend on their
beliefs about the other participants’ behavior. Randomizing bids
may help avoid predictable patterns, which can be exploited by
competitors.
 3. Advertising
o Companies often use mixed strategies in their advertising
campaigns. For example, if two companies are competing for
market share, they may not want to always use the same amount
of advertising spending every period. Randomizing the level of
investment in advertising can make the company less predictable
to competitors, avoiding counteracting strategies.
4. Military Strategy
o Mixed strategies are often used in warfare or conflict scenarios.
For example, a nation might randomize its military deployments or
tactics to avoid predictability. This makes it harder for the
adversary to plan effective counterattacks based on previous
behavior.
5. Pricing Strategy in Oligopolies
o Firms in an oligopoly might use mixed strategies in deciding prices
for their products. If all firms in the market choose their prices
deterministically (using pure strategies), competitors may be able
to anticipate the prices and adjust accordingly. By using mixed
strategies, firms can avoid predictable pricing patterns and
maintain an advan

2.3 Finding Nash Equilibrium with Mixed Strategies

To find the Nash Equilibrium (NE) in a game with mixed strategies, you
need to analyze the players' strategies when they randomize their
actions. A mixed strategy Nash Equilibrium is a strategy combination
where each player's mixed strategy is the best response to the other
players' mixed strategies.
Here is the step-by-step process for finding the Nash Equilibrium using
mixed strategies:
Steps for Finding the Mixed Strategy Nash Equilibrium:
1. Define the Players' Strategies:
 Suppose we have a 2-player game (although the process can be
generalized to more players).
  Each player has several available pure strategies.
 Players will assign probabilities to their strategies. For example:
o Player 1 assigns probability pp to Strategy 1 and 1−p1-p to
Strategy 2.
o Player 2 assigns probability qq to Strategy A and 1−q1-q to
Strategy B.
2. Set Up the Payoff Matrices:
 Write the payoff matrix for both players, detailing the payoffs each
player receives for each combination of pure strategies.
3. Indifference Condition:
 A player is indifferent to choosing between two or more strategies in a
mixed strategy Nash Equilibrium if the expected payoffs for each strategy
are equal.
 Set the expected payoffs equal to each other for the player to be
indifferent between strategies.
5. Solve the System of Equations:
 By setting up equations based on the indifference condition, you can
solve for the probabilities (e.g., pp for Player 1 and qq for Player 2) that
make each player indifferent to choosing their strategies.
Example of Finding Nash Equilibrium in a Mixed Strategy Game:
Consider a simplified 2-player game with the following payoff matrix:
Player 2: A Player 2: B
Player 1: X (3, 2) (0, 1)
Player 1: Y (1, 0) (2, 3)
 Player 1 has two strategies: X and Y.
 Player 2 has two strategies: A and B.
Step 1: Assign Probabilities to Each Player's Strategies
 Let Player 1 randomize their strategies:
o Player 1 plays X with probability pp and Y with probability 1−p1-p.
 Let Player 2 randomize their strategies:
o Player 2 plays A with probability qq and B with probability 1−q1-q.
Step 2: Calculate the Expected Payoffs for Each Player
For Player 1:
 Expected payoff for Player 1 choosing X:
Payoff(X)=(q⋅3)+((1−q)⋅0)=3q\text{Payoff(X)} = (q \cdot 3) + ((1-q) \cdot
0) = 3q
 Expected payoff for Player 1 choosing Y:
Payoff(Y)=(q⋅1)+((1−q)⋅2)=q+2(1−q)=2−q\text{Payoff(Y)} = (q \cdot 1) +
((1-q) \cdot 2) = q + 2(1 - q) = 2 - q
For Player 2:
 Expected payoff for Player 2 choosing A:
Payoff(A)=(p⋅2)+((1−p)⋅0)=2p\text{Payoff(A)} = (p \cdot 2) + ((1-p) \cdot
0) = 2p
 Expected payoff for Player 2 choosing B:
Payoff(B)=(p⋅1)+((1−p)⋅3)=p+3(1−p)=3−2p\text{Payoff(B)} = (p \cdot 1) +
((1-p) \cdot 3) = p + 3(1 - p) = 3 - 2p
Step 3: Set the Indifference Conditions
In a Nash Equilibrium, each player must be indifferent between their
available strategies. This means the expected payoffs from each strategy
must be equal for each player.
For Player 1 to be indifferent between X and Y:
Set the expected payoffs equal:
3q=2−q3q = 2 - q
Solving for qq:
3q+q=23q + q = 2 4q=24q = 2 q=12q = \frac{1}{2}
For Player 2 to be indifferent between A and B:
Set the expected payoffs equal:
 2p=3−2p2p = 3 - 2p
Solving for pp:
2p+2p=32p + 2p = 3 4p=34p = 3 p=34p = \frac{3}{4}
Step 4: Conclusion (Nash Equilibrium)
 The mixed strategy Nash Equilibrium occurs when:
o Player 1 plays X with probability 34\frac{3}{4} and Y with
probability 14\frac{1}{4}.
o Player 2 plays A with probability 12\frac{1}{2} and B with
probability 12\frac{1}{2}.
Thus, the mixed strategy Nash Equilibrium for this game is:
 Player 1: (X: ¾, Y: ¼)
 Player 2: (A: ½, B: ½)