I(ane Trading on:

Entry
Techniques

by Jim I(ane
Tips� ideas and techniques for market traders
Kane Trading on:
Entry Techniques

By Jim Kane

KaneTrading.com
Kane Trading on: Entry Techniques

Copyright © 2003 by James J. Kane

Published by Kane Trading

ALL RIGHTS RESERVED. No part of this publication may be reproduced,

stored in a retrieval system or transmitted in any form or by any means,
electronic, mechanical, photocopying, recording, scanning or otherwise,
without prior written permission of the publisher and the author.

This publication is designed to provide accurate and authoritative

information in regard to the subj ect matter covered. It is sold with the
understanding that the publisher and the author are not engaged in rendering
legal, accounting or other professional services. If professional advice or
other expert assistance is required, the services of a competent professional
person should be sought.

Printed in the United States of America

Disclaimer

No claim is made by James J. Kane, or Kane Trading, that the trading

methods shown in this book will result in profits, or will not result in losses.
There is a substantial risk of loss in trading securities, options on securities,
futures, options on futures or any other trading vehicle. Past performance is
not indicative of future results. Trading securities, options on securities,
futures, options on futures or any other trading vehicle may not be suitable
for all recipients of this book. Always seek competent professional advice
when considering any trade. All examples in this book are for educational
purposes only. All material and examples in this book are based on
information obtained from sources that are believed to be reliable, but which
are not guaranteed as to their accuracy or completeness. Nothing in this book
should be construed, in any way, shape or form, as a solicitation of any offer
to buy or sell any trading instrument. James J. Kane, his fami ly and friends,
and associates of Kane Trading have at times in the past and may now or at
times in the future, trade or have traded any or all of the issues used as
educational examples in the book. Any thoughts or opinions expressed in
this book are subj ect to change without notice. No information provided in
this book should be construed in any way as an encouragement by the
author, publisher or distributors to trade. Each trader must make his or her
own decisions with regard to trading. Each trader must be responsible for his
or her own decisions and his or her own actions, if any. Purchasing or
reading this book or parts thereof constitutes acceptance of and agreement to
this disclaimer and exempts the author, publisher and distributors from any
and all liability and l itigation.

v
Table of Contents

Acknowledgements
.

IX

Introduction 1

Chapter 1 Entry Technique Overview 7

Chapter 2 Moving Average Cross 11

Chapter 3 Moving Average Crossover 35

Chapter 4 Trendline Violations 59

Chapter 5 Regression Channels 87

Chapter 6 Swing-High and Low Violations 99

Chapter 7 Pullbacks 117

Chapter 8 Additional Swing-Point Entries 135

Chapter 9 A 'Cool Trick' 169

Conclusion Conclusion 193

Vll
 Acknowledgements

In my evolution as a trader I 've read more material than I can even recall .
The majority o f this material has contributed very l ittle to my knowledge
base. That lack of value, for me, in the material, in and of itsel f, i s important
information. It' s shown me, by a process of elimination, the things that don't
help me, and I can use that information when formulating a trading p lan.

In developing material related to Fibonacci trading, two sources have been

of great help. I woul d l ike to acknowledge these sources, and recommend
that readers look into their materials. See if they might be of as much help to
your own trading, as they were to mine.

I ' d like to acknowledge Scott Carney over at Harmonic Trader. Scott's book,
The Harmonic Trader, and the material on his website
(www .HarmonicTrader.com). opened my eyes to another way to view the
markets. This was my first substantial introduction to the concepts of
Fibonacci and harmonics in trading the markets. Scott has quite extensive
information on harmonic patterns on his website and has developed several
patterns of his own. Scott and I have since spent endless hours discussing
harmonics and the markets. His historical knowledge of the markets is
extraordinary and has contributed greatly to my own knowledge base.

1 ' d also l ike to acknowl edge Robert Miner at Dynamic Traders Group, Inc.
(www .DynamicTraders.com). Robert's book, Dynamic Trading™, was my
next serious excursion into Fibonaccis and trading. This book is extensive
beyond belief. There is so much material in Dynamic Trading™ that I would
have to consider it must reading for anyone interested in increasing their
knowledge of Fibonacci in trading and in E l liot wave analysis. Robert's use
of the time factor is also extensive and will open one ' s eyes to factors
outside of j ust price. Robert also has Dynamic Trader software available,
which I use for creating charts l abeled with various Fibonacci, harmonic and
time factors. It is the software that I used to create the charts for this book. I
would l ike to extend an additional thanks and acknowledgement to Robert
for allowing me to use these charts in my works. Information on his
products is available on his website. I recommend checking it out to see if
you feel that it has information that you can use to help your trading. His
contributions in the field are practically immeasurable.

IX
 Introduction

This purpose of this work is to present an assortment of specific entry

techniques that a trader may use to initiate a position in a market. But what
is an 'entry technique' ? And where does an ' entry technique' fit into an
overall 'Trading Plan'?

Let me first lay out what I think are the critical elements of a trade, what I
sometimes call the 'Plan for a Trade' . Once I have laid this out, the specific
place of the entry technique in the hierarchy of the plan will be quite clear. I
feel this is important to understand, because far too many traders trade
without a plan, or with a poorly conceived plan. If you approach trading as a
business, why would you expect to run your business without a plan?

When I consider a trade, I go through a process to determine exactly what it

is that I am trying to do, and how I wil l react, regardless of what I may
encounter as the process unfolds. Understand, you must develop your own
'Trading Plan', based on your own business plan, risk tolerance, goals,
financial situation, and whatever other factors come into play in your
individual situation.

What I will present here is what I have developed for myself. Hopefully, you
can look this material over and find things that you can use, or things that
you can modify for your own use. I am in no way implying that you blindly
follow anything I present here, j ust because I ' ve presented it. Evaluate
everything you encounter (from me or any source), and decide if you feel it
will help you. I f not, rej ect it. With that said, let's look at what I do.

I look at the initiation of a potential trade in four ' sections' , the critical
elements of a trade:

1 . Potential trade 'area' (PTA)

2. Entry technique
3. Trade size determination
4. Trade management

1
I can't emphasize enough, though; these are not the elements of the 'so
called' 'Trading (or Game) Plan . (Please see my very brief free article
'

entitled ' Kane Trading on: The Difference B etween a 'Plan for a Trade' and
the 'Trading Plan' .) These are the four elements that I feel are critical to
initiating a specific trade.

These four elements say nothing about how much capital you would use on
the trade, how much capital you woul d be willing to put at risk, what trading
vehicle(s) you might use (futures, options and/or option combinations, single
stock futures, stock, etc.), and so on. These things are part of the 'Trading
Plan', or ' Game Plan', as it is sometimes called. Here we are looking at j ust
one piece of the larger scope 'Trading Plan', the immediate and critical
elements that get you into a specific trade. With all that said, let's look, now,
at the four elements in a bit more detail.

I will briefly explain each of the four sections, but first notice something that
should be fairly obvious: the title of this work only encompasses section
two. After reading this work you will only have one part in four of the 'Plan
for a Trade' 'puzzle ' .

I have purposefully done this i n all o f my works, that is, focused on

individual concepts. I feel this separation allows the ideas to be explained
much better. It also allows those readers looking to fill very specific gaps in
their trading plans a much easier time finding the material they are seeking.
Let's move on to brief explanations of the four sections.

Section one of the plan is 'potential trade area' (PTA). When I think about a
trade, the first thing I decide is where would I consider a trade. Just because
I want to trade an issue doesn't mean I j ust open a trade at that instant.

For example, say an issue is trending up fairly strongly, like a typical tech
issue in the late 90' s. Perhaps I want to get into a long position, and attempt
to ride the trend up for awhile. I have found that if I j ust buy right then, it
seems l ike that ' s the near exact instant that the issue pulls back a bit. All of
the sudden I ' m in a losing trade right off. S ince I always use a protective
stop loss on all trades (and you should, too), I ' m frequently stopped out on
that type of trade very quickly. Ever have that happen to you?

And if that wasn't bad enough, what frequently happens right after you get
stopped out? You guessed it; the issue turns and does exactly what you

2
expected it to do, only without you on board. This is an example of having
the right idea, but not the right area for the trade. One possible solution, for
example, might be to enter on a pullback. The area you might focus on for
the possible termination of the pullback could be determined with various
technical analysis tools, according to your specific trading plan.

In my book Kane Trading on: Advanced Fibonacci Trading Concepts, I

present methods for finding potential areas of support and resistance, against
which a trader might consider taking a trade. That book focuses entirely on
section one of the ' Plan for a Trade' , finding areas where a trade may be
considered. There are many other ways to look for potential trading areas.
The point is, step one is to define an area where you would be comfortable
attempting to enter the market. Once I ' ve decided on the area in which I am
considering a trade, I move on to section two, entry techniques.

Entry techniques are the subj ect of this work. I ' m watching an area, but I am
not going to take a trade until a specific entry technique 'triggers' me into
the trade. The main idea, for me, is that I need some confirmation (once the
issue enters a potential trade area), that the trade is ' acting' like it may work
out as I hope it will. The determination of my specific entry signal will also
be used in my calculations in section three, on how many shares or contracts
I will trade.

Once I have my entry signal, I will be sitting back and waiting to see if it
actually triggers. If I don't get a trigger, I simply don't enter the trade, and
move on, looking for another trade. I have a saying, "No trigger, no trade."

Element number three is the determination of the size of the trade. To make
this calculation I need three things. I need the dollar amount I am willing to
risk, the protective stop loss point, and the entry point. Some simple
arithmetic then gives me the trade size. Understand this section is a simple
calculation of the trade size, and no more.

This section does not cover calculating or deciding the amount to risk,
maximum allowable trade size, or anything of that nature. Those things
come from the master 'Trading Plan', and are ' imported' into this section for
use in the calculations of trade size.

There are a host of books out there dealing with the topic of capital at risk,
and since I have my own ideas, I will soon be producing a work entitled

3
Kane Trading on: Capital at Risk and RewardlRisk Ratios, dedicated to this
topic. Some sources say never risk more than two percent of your total
capital on any given trade. Other sources say never have more than ten
percent of your capital in a given play. There are as many ideas about initial
risk as there are books out there on the topic. The most important thing, I
think, is that you quantify your risk before you enter a trade, according to
your fully laid out business plan.

In most of my trading, I decide, first, what dollar amount I am willing to risk

on a particular trade. Then I make a decision, based on my technical
analysis, where I should set my protective stop loss. In overly simplified
terms, I usually pick an area just under support (for long trades) or over
resistance (for short trades), for my stop loss. I then determine how far that
stop is from my entry point trigger, and by doing some simple math, I can
determine how many shares or contracts I can trade and stay within my
maximum acceptable loss for the trade.

There is one caveat I must include here. Understand, an issue can gap
through a stop, or can be halted and open trading far above or below your
stop, it can stop trading altogether, it can go limit up or down for days on
end, and a host of other occurrences that can increase your loss far above
your desired limit. This is a part of trading, and must be factored into your
business plan. Don' t expect that just because you have a protective stop loss
set, that your loss will be guaranteed to be l imited to that amount.

As an aside, another aspect I look at in the master 'Trading Plan' that can
affect whether or not the trade is acceptable is the reward/risk ratio.
Although I will leave the details of this to my upcoming work on capital risk
and reward/risk ratio, suffice it to say I also want a favorable reward to risk
ratio on a given trade. For me, there is no hard and fast rule, like it must be 2
to 1 , or 3 to 1, or whatever. I believe that you must take into account both
the reward to risk ratio and the probability of the success of the trade.
Taking both of these factors into account is known from game theory as the
'expected value ' . In trading I sometimes see this referred to as the ' expected
outcome' . In my work on capital risk and reward/risk ratio, I ' ll go into
extensive detail on my outlook on all this.

This section on trade size determination is only included as a critical element

of a trade because in order to initiate any trade you must say what size you

4
want the trade to be. The scope of this element, in the context of the 'Plan
for a Trade' , is strictly limited to this one simple calculation.

If the trigger does initiate a trade, I then move on to section four, which is
trade management.

Trade management tells me what to do, as a variety of potential scenarios

unfold. I believe it is an important part of a sound trading plan to have a pre­
determined idea of how you will manage your trade, once you have entered
it. This includes pre-determined reactions to every possible scenario that can
come into play. When I ' m in a trade, you could ask me ' What would you do
if this happened?' or ' What would you do if that happened?' and I would be
able to spit out an answer instantaneously.

If it' s something that could realistically happen (even if the odds of it

happening are slim), I know, without even thinking, how I will react. In my
opinion, if I don't know without thinking how I will react, I have no
business being in the trade. Pre-determined trade management allows me to
have the answers, in advance, for every possible scenario I may encounter.
My work Kane Trading on: Trade Management lays out my views, in detail,
on the topic of trade management. Another of my works, Kane Trading on:
Trailing Stops, provides information on trade management for trades that are
moving in the trader's favor.

There you have the four sections or elements of the 'Plan for a Trade' . You
can now see how entry techniques are a mere part of a complete 'Trading
Plan', fitting in quite logically in section two of the 'Plan for a Trade' . And
the 'Plan for a Trade' is only a part of the all-encompassing 'Trading ( =

Game, Business) Plan' . You can also see that an entry technique is nothing
=

more than a 'trigger' to get you into the trade. It is, in and of itself, not
anywhere near a sufficient reason to initiate a trade. It is a crucial part of the
'Plan for a Trade' , but it is just that: a part of a plan. As we proceed now to
some specific entry techniques, never forget the context of an entry
technique, within the overall 'Trading Plan' .

5
 Chapter 1

Entry Technique Overview

Before I proceed ahead and start presenting specific entry techniques, I want
to delve into the concept of ' context' a little bit. What do I mean by
'context'? If a 'Plan for a Trade' has four sections, as described in the
introduction, then any look at a section by itself, or a part of a section, has
very little meaning outside of the context of the entire ' Plan for a Trade' .
The entry techniques I will present should only be viewed, and used, in the
context of a well thought out, complete trading plan.

This means that an entry technique is to be viewed as a choice you make, to

fulfill section two of your own personal 'Plan for a Trade' . The 'trigger' of a
specific entry technique is next to valueless if it occurs outside of a
'potential trade area' (an area that satisfies your section one criteria). Why is
that? Simple. The potential trade area is that area where you have
determined you have a possible edge, hence the desire to trade in that area. If
the area didn't appear to give you an edge, you wouldn't consider a trade
there, i.e. it wouldn't meet your section one criteria as a potential trade area.

It follows that the entry technique wil l also factor into your calculations on
potential capital at risk, as discussed in the introduction. Now you can
simply adjust your shares or contracts traded to not exceed your total capital
at risk limit, as outlined in section three, trade size determination. This, of
course, doesn't address the reward to risk ratio or other factors that are laid
out in the larger scope 'Trading Plan' . I won't go into details in this work on
the capital at risk aspects of the 'Trading Plan', but it should be obvious that
the entry technique is directly involved with the amount of capital at risk.
Again, both have little meaning without the context of the other.

Finally, the entry technique gets you into a trade, and then trade
management (section four) takes over. But entry point(s) will play into
calculations you may do to determine exit points and trade management.
Your management plan is not set in stone, devoid of the context of where
you entered the trade, and why. The stops you choose, and how you move
them, will be decided based on all three other areas of the trading plan.

7
This is pretty easy to see from a common sense standpoint. Good choices on
what you do as a trade unfolds will surely take into account why you chose
the area for the trade in the first place (section one). It would surely take into
account how much capital you are willing to risk. And it' s obvious where
you initiated the trade factors into anything you would do during your
management phase. All of these sections form a web, with each section
intricately involved with all the others.

Why go into all this detail explaining this? First, I think the importance of
having a complete trading plan, a plan where the relationship between the
individual pieces is clear, can't be overstated. Second, I ' m presenting j ust a
piece of the whole in this work, and I want it to be crystal clear that this is
not to be viewed as a stand-alone technique. The entry techniques have little
value outside of the context of a complete plan, and this can't be overstated.
Finally, the techniques themselves are a part of my trading plan. I want to
give the reader a glimpse, at least, into the context in which 1 use the
techniques.

The techniques may have some value, in many applications, outside the
context in which I use them. I have found them to be of value, to me, in the
context of my own trading plan. This is why I encourage the reader to
experiment with the techniques and see, without risking any real capital, if
they appear to be of some help. Test them extensively, and reject what you
don't like.

A significant part of my trading plan revolves around attempts to get on

board existing trends. (The main process by which I do this is outlined in my
book Kane Trading on: Advanced Fibonacci Trading Concepts.) The trends
may be on essentially any timeframe. I use three timeframes to manage a
trade. I choose a timeframe on which I will trade, one that shows me a trend
that looks good to me on that timeframe. I then look at a higher timeframe
(usually three to five times as high) for context, to see how this trend fits in
the bigger picture. I frequently look at even higher timeframes to get even
more context.

Once I decide I l ike the trend and I want to attempt a trade, I look for a
pullback. Once the pullback is in progress, I drop down to a lower time frame
(usually three to five times lower) and decide on an entry signal. This entry
signal is what actually triggers me into the trade. It' s this entry signal that we
will delve into in this work. The entry signals I will present here are signals I

8
use to attempt to board an existing trend on a pullback, where the pullback is
viewed in a lower timeframe.

These signals may have some use in attempting to enter many other different
types of trades (such as a channeling issue or a reversal trade entry off of a
pattern). My use of the signals is mostly l imited to the above 'pullback in a
trend' context, although I have had some positive results in other contexts. I
j ust want to point out how I usually use the signals. It' s up to you, as the
reader, to test the signals for yourself, in the context of your trading style
and trading plan, and make your own decision if they are of any use to you.

9
 Chapter 2

Moving Average Cross

Before we proceed on to the details of the ' moving average cross' entry
technique, let ' s reiterate a few assumptions first. These assumptions apply
not only to this chapter, but also to all the techniques in the following
chapters.

First off, it' s assumed one wouldn' t even be looking at an entry technique
unless criterion one, the potential trade area, has already been addressed.
Next, once an entry technique is decided on and trade size determined, it' s
assumed one will have addressed criterion four, trade management, before
the entry is actually taken.

This should be obvious, since if you initiate the trade and then try to figure
out how you will manage it, you will be trying to figure out critical details
' in the heat of battle' . That is something you want to avoid like the plague. I
want to make all critical decisions in the cool, calm of the off-market hours.
I want to have one sole purpose when the market is ' l ive' , and that' s to be an
execution robot.

If I 've laid out all the possible scenarios that can occur, and pre-decided on
what my action(s) will be when they occur, I can simply be an ' execution
robot' when the pressure is on. I can't over-emphasize how important I feel
it is to not make decisions on the fly. This doesn' t mean that you don't take
action on the fly, but means that you don't have to think out different
courses of action, decide on the possible outcomes of these courses of
action, and weigh up the pros and cons on the fly. That ' s a recipe for
disaster, in my opinion. I make all my decisions before the market opens.

Now given all these assumptions, and the context of the entry technique
within the plan, let's move on to the techniques themselves. I will present
quite a few techniques in the chapters that follow, and each has its own
'personality' . I use different techniques for different situations. A lot of the
basis for my decisions on which technique to use in a given situation is
experience, and that's something I can't give you. I will give you hints and
tips on my thinking process as often as I can, but to some extent you will
have to experiment on your own and test the ideas with your own style, in

11
order to make your own decisions on which technique you might use in a
given situation.

The moving average cross technique is an extremely simple technique to set

up and use. Entry is simply triggered when the moving average is crossed, as
the name implies. B ut does that mean as soon as the price crosses the
average? Or when it closes for the first time above (for long trades) or below
(for short trades) the average? And what moving average? Simple or
exponential moving average? What period for the moving average?

These are some of the questions that should be coming to mind. Get ready,
though, the answers that I will give to these questions will not only surprise
you, but unless you are really familiar with the Kane Trading methodology,
probably will also disappoint you. I expect that disappointment will wear off
long before the end of this book, though.

So what are the answers? The answer to that is: there aren't any! I can hear
the disappointment. I ' m hoping that anyone who buys this book already
knows that the Kane Trading style is not to promise there are answers, or
techniques (read that: tricks), that one can j ust tell you, and off you go, an
instant success. The moving average cross entry may work for you, but only
if you tailor it to your style and needs.

As soon as you think about tailoring something to fit your needs, you see
that there would, then, be no single way to configure it so that it would work
well for everyone, in varied situations. This is perhaps the single-most
important thing you should take from this book. A technique is only as good
as how well you are able to utilize it to suit your own personal needs.

By this, I don't mean optimizing. I don't mean to tweak it until it maximizes

the results in backtesting. I mean pick techniques to meet a specific need you
have, techniques that are suited to the unfilled part of your plan. Let's get
more specific, looking at the moving average cross technique.

This technique, as I mentioned, is extremely simple and easy. Right off, that
tells you that you shouldn't expect too much from it. It would be a technique
I would use, then, when I don't need a lot of confirmation. That's the first
screening tool I use when trying to decide on which entry technique I might
use. How much confirmation do 1 want? Each entry technique could be
roughly assigned to a level of confirmation that it provides.

12
It only makes common sense, then, that the less confirmation a technique
provides, the more signals you will get, and that more of those signals will
be false. If you want a lot of confirmation, you ' l l get a lot less signals, but
usually a lot less false starts. Another thing I frequently see is that the less
confirmation the signal provides, the closer your stop can usually be.

This puts you into a decision making process where you see that you can
frequently stop out with a smaller loss on the lesser confirmation techniques,
but you will take a larger number of those losses. On the other hand, as you
go up the spectrum towards larger and l arger degrees of confirmation, the
losses will be larger and larger, but less and less frequent. In my experience
it' s j ust about a wash, with all the techniques being roughly the same over
the long run.

So why have multiple techniques? My next screening tool is to screen for a

technique that is better suited to the way that particular issue is trading at
that time. When I say that I think that the net variance between each
technique is a wash over time, I say that given the assumption that the
techniques are applied only when the conditions are suitable for that
particular technique. I wil l do my best to state how I best l ike to use each
technique, and give examples, as we go along.

Looking at the moving average cross technique, we know it is simple and

has very little confirmation. That begs the question, ' When would I want just
a small amount of confirmation?' For me, step one is to choose my
'potential trade area' , as discussed earlier. But not all setups are equal. You
can make a j udgment as to how confident you feel about that setup.

Let' s make one thing clear, though, about your judgment call on the setup.
Be wary of relating your emotional feelings about the trade to your level of
confidence. Frequently the best trades can be the scariest. If you j udge the
confidence you have in a trade based on this emotional fear level , your
j udgment could frequently be exactly opposite of what it should be.

I'm talking about a technical j udgment based on comparing the setup to a

group of similar setups from your experience base. You might be saying
' Given this type of setup, I rate this one a seven out of ten, compared with
the large group of these I have seen and traded in the past', or something to
that effect. Perhaps you decide you wil l only trade ones that rate from seven
to ten, and that a seven requires a maximum confirmation technique and then

13
a gradient applies, and the setups that rate a ten match up with the techniques
that have the least amount of confirmation.

But why do it that way? Why not j ust go with the maximum confirmation in
every case? Remember, the more confirmation you desire, the smaller the
number of trade signals you will get, and the greater the distance your entry
point will l ikely be from a logical stop loss point. By not using the ' least
restrictive' technique that fits the scenario, you may miss trades that perhaps
you don't want to miss, and pay more for that entry than you otherwise
would have to.

I, personally, try to use the least confirmation technique I am comfortable

with for that given trade. I feel this is one of my better ' secrets' (well, it' s
not a secret now ! ), and something I can't recall ever having heard anyone
talk about before.

I also evaluate the potential trade area in terms of what type of trade it is, to
determine how much confirmation I want. Although I have no hard and fast
rules that I can quantify for you, I do apply a fairly uniform standard to all
my potential trades. Some trades are lower probability than others, j ust by
their nature.

This doesn't mean that these trades may yield less money in the long run
(they may pay out more per trade), j ust that they work out less frequently.
F or trades like that I prefer a larger degree of confirmation. There is a
gradient of trades and a gradient of triggers, and I try to match them up as
best as I can.

Let's assume we have a potential trade area we like, and if we get a trigger
in this area, we want to take a trade. Let ' s also assume that we have decided
that the trade is a fairly high probability trade, and that the setup is looking
real good so far. We are willing to initiate a trade with just a smal l amount of
confirmation. First, l et's look at an example of a potential trade area, based
on a Fibonacci grouping (see Kane Trading on: Advanced Fibonacci Trading
Concepts for more information on constructing and trading Fibonacci
groupings). See figure 2 .l .

14
Figure 2 . 1

jj) �
?. AOl D-D 1!!113 £J

1 4 . 000

1 3 . 500

tfl 1
1 3 . 000

II
11,1"_••
1 2 . 500

1 2 . 000

I U JI I

I1hlJ
1 1 . 500

1 1 .000

ttp! 1 0 . 500

1 0 . 000
Feb 14 21 28 Mar 14 21 28 Apr 11 17 25
Chart created by Dynamic Trader (c) 1 996-2001

A simple choice for a moving average cross trigger would be a close above a
1 0-period simple moving average. This is a middle of the road average for
short-term trading, and the requirement of just one close above the average
is also middle of the road. This is a good starting point to look at.

Let's first look at the same issue as above, but with the 1 0-period simple
moving average on the chart. Before we do this, though, let' s give this a
little thought. Do we want to look at the issue on the same timeframe? The
answer is no, we want to go down one time frame.

Remember, what I l ike to do is use three timeframes, with the ' middle '
time frame being the 'traded' timeframe. The higher timeframe is for context
and the lower timeframe is for entry. I usually use a timeframe 'j ump' of
approximately three to five times, but when going down from a daily, I go a
little more of a j ump, in order to use a very common bar.

Hence, with figure 2 . 1 being a daily chart, I ' l l choose to use a 60-minute for
entry. So we' l l look at the issue, now with a 1 0-period simple moving

15
average, on a 60-minute chart. We' l l also move a l ittle bit ahead in time, as
we approach the potential trading area. See figure 2.2.

Figure 2 .2

?: AOt. 60-1 I!! EH!I

1 4 , 200

1 4 , 000

1 3 , 800

1 3 , 600

1 3 , 400

1 3 , 200

1 3 , 000

1 2 , 800

1 2 , 600

Apr 22t 23w 24t 2sf Apr 29t 30w 1t 2f May 6t 7w

Chart created by Dynamic Trader (c) 1 996-200 1

The first thing I notice is how ' choppy' AOL is with respect to its l O-period
moving average. My first inclination woul d be to consider another entry
technique here, one that perhaps is better suited to such ' chop ' . I would be
thinking about a trendl ine violation technique or a regression channel
violation technique, or perhaps even a swing-high or low technique.

I wil l cover all these in detail in l ater chapters, and I wil l use this exampl e
again for comparative purposes. S o why even continue here with the moving
average cross technique? I already stated I felt very confident with the setup
and so I preferred to have a lower confirmation entry technique.

Remember, I l ike to use the l east confirmation technique that I fee l is

warranted, so as to maximize my number of potential trades that trigger.
Despite the ' chop ' , I stil l want to see how this one looks as it gets closer to

16
the potential trade area. Let's advance ahead in time as AOL approaches the
area we might consider trading. See figure 2.3 .

Figure 2.3

?. AO L 60-1 1!!!I[;l r!i

1 4 . 200

1 4 . 000

1 3 . 800

1 3 . 600

1 3 . 400

1 3 . 200

1 3 . 000

1 2 . 800

1 2 . 600

Apr 22t 23w 241 25f Apr 29t 30w 1t 2f May Gt 7w 8t

Chart created by Dynamic Trader (c) 1 996-2001

The first thing I notice here as time has moved on a bit, is that at least AOL
is behaving in a manner where I would stil l consider using the moving
average cross technique. I am looking for a somewhat smooth decent into
the area, mostly trending down and staying below the moving average.

I don't want AOL to be going above and below the average multiple times.
This is obvious, since each close above would be a signal, and most would
prove to be false signals. Keep in mind here what my obj ective is. I think
this grouping (or more precisely, two very tight groupings) wil l reverse AOL
and the larger daily timeframe trend wil l reassert itself.

Some traders would simply ' fade' the trade and take it as it enters the zone.
This is a valid entry technique. It is on the scale on the end that says 'no
confirmation' . You could move it on the scale a tiny bit, to ' a tiny bit' of
confirmation, by simply hesitating for a l ittl e while and watching to see if

17
the issue slows down or appears to stop, as opposed to j ust plowing right
through the area.

For me, I don't like either of these two techniques, although I do use them
under certain circumstances. If I ' m very confident in the trade' s potential, I
want just a bit more than the ' fade', or the ' fade with a pause' . Hence I
might choose the moving average cross technique. This is why I am hanging
with this AOL example up to this point. Let's watch as AOL enters the
potential trading area. See figure 2.4.

Figure 2 . 4

�:. AOl 60-1 1!!113 £J

1 4 , 200

1 4 , 000

1 3 , 800

1 3 , 600

1 3 , 400

1 3 , 200

1 3 , 000

1 2 , 800

1 2 , 600

Apr 22t 23w 24l 25f Apr 2'lt 30w 1t 2f May E.t 7w at
Chart created by Dynamic Trader (c) 1 996-2001

What a difference a few bars can make. The last bar is a very significant
plunge into the top of the groupings, and is a little disconcerting. I have
noticed, though, that this is a fairly common occurrence when trading
Fibonacci groupings, and if AOL ' straightens out' and doesn't simply
continue to plunge right through the zone, I ' m still willing to take a trigger if
it comes off the grouping. Let's continue forward a few more bars and re­
evaluate. See figure 2 . 5 .

18
Figure 2 .5

:::.: AOl 60-1 !!9 l3

1 4 , 200

1 4 , 000

1 3 , 800

1 3 , 600

1 3 , 400

1 3 , 200

1 3 , 000

1 2 , 800

1 2 , 600

Apr 22t 23w 2<k 25f Apr 29t 30w it 2f May 6t 7w St 9f

Chart created by Dynamic Trader (c) 1 996-200 1

Now we are right at the critical choice point. AOL has bounced right off the
top of the first grouping and hasn't c losed above the moving average for
eighteen bars. It came off the grouping and went up a bit, rolled down a bit,
and is now moving up a l ittle.

This might j ust be some congestion, or that may have been it, a 'test' of the
bottom and up it goes. I would prefer to get deeper into the groupings, but
we have 'officially' hit them, so a cross and close above the moving average
now would be considered a trigger that I would take. Let's see what
happens. See figure 2.6.

19
Figure 2 .6

?. AO L 60-1 I!!!IS £I
1 '\ , 200

1 4 , 000

1 3 , 800

1 3 , 600

1 3 . 400

1 3 , 200

1 3 , 000

1 2 , 800

1 2 , 600
22t 23w 24t 25F Apr 29t 30w it 2F May Gt 7w St 9F r
Chart created by Dynamic Trader (c) 1 996-2001

Okay, I couldn't ask for more on this one. AOL gapped down on the next
bar (which is also the open of a new trading day), went right down to the top
of the groupings again, and then rocketed up, closing wel l above the moving
average. In fact, it hit the low from the previous day to the exact penny, and
then buyers came i n and it went up strong with an expansion bar and a
strong close.

At this point I would be triggered into the trade. As an aside, I would be

using the area of the tight groupings to choose my stop placement, and I,
personally, would be placing a stop just below the lower end of the
groupings. This is one of the main reasons why I l ike the grouping technique
so much, because it gives me some real nice choices for stop losses.

Let's finish up this example with a look at how AOL reacted to the grouping
as more time passed. First let's look at the ' shorter-term' trade on 60-minute
chart. See figure 2 . 7 .

20
Figure 2. 7

?: A O l 60-1 I!UiI F3

1 4 . 200

1 4 . 000

1 3 . 800

1 3 . 600

1 3 . 400

1 3 . 200

1 3 . 000

1 2 . 800

1 2 . 600
25 May2 '3 16
Chart created by Dynamic Trader (c) 1 996-2001

This trade really came off that zone, and the moving average cross technique
did, in fact, work extremely wel l in triggering the trade. But remember, the
60-minute chart was only used as the trigger chart. The trade was on the
daily. Don't change your trade parameters and stay with the trigger
timeframe once you are in the trade. I f you are trading the daily, trade the
daily. So let's look at how AOL responded on the daily. See figure 2.8.

21
Figure 2 .8

?. AOL 0-0 !lEI£'{

1 7 . 000

1 6 . 000

1 5 . 000

1 4 . 000

1 3 . 000

1 2 . 000

1 1 . 000

1 0 . 000
� � � b �
Chart created by Dynamic Trader (c) 1 996-2001

As can be clearly seen, this trade had quite a ways to go on the daily
timeframe. If one would have stayed on the 60-minute timeframe to follow
the trade as it progressed, it is very l ikely that the perspective would have
induced the trader to take a quick profit, and miss most of what was to come.
It' s important not to forget the role of each of the timeframes within the
context of the trade, as wel l as the context of the entry technique.

Let's do one more example, this time with a longer moving average, a 20-
period. What effect, in general, will this have, as compared to, say, the 1 0-
period we just used? The 20-period will usually take more movement on the
part of the issue to overcome it.

This means you wil l l ikely get fewer false signals. And you will l ikely be
farther away from a logical, technical stop loss point. You will have more
confirmation, but you pay for it. I would consider using the 20-period when I
don't have the amount of confidence in the potential trade area as I have
when I consider the 1 0-period.

22
I understand this may all sounds pretty abstract or nebulous, but that' s not
the intention. It' s just that I don't have any hard and fast rules for this. I
make decisions based on experience. My advice would be to look at
examples of trade areas that you trade, based on your style, and then look at
the various entry techniques I have presented. See which ones work best
with each different type of trade, and try to see which ones work best for
you, and under what circumstances. Try to match the technique to the trade.

In essence, that is what I have done. There is no perfect answer, it' s just a
matter of finding what works best for your style and your types of trades. It
takes a lot of off-market hours chart work to figure that out, but that's
trading.

Let ' s look at C SCO. I noticed CSCO had been in a fairly strong downtrend
on a daily chart. I had also noticed that it had had a fairly good-sized
correction to the downtrend in the end of February and beginning of March,
and then it resumed its downtrend. I am l ooking for a chance to get on board
the downtrend. See figure 2 .9.

Figure 2 .9

;'-1. CSCO D-D l!!Il3! £I

2 1 .000

20 , 000

1 9 , 000

1 8 , 000

1 7 , 000

16,000

1 5 , 000

11 18 25 Feb 8 15 22 Mar 8 15 22 28 Apr

Chart created by Dynamic Trader (c) 1996-2001

23
I also realize that, perhaps, we would be looking at a short trade right in an
area of potential support. I, myself, would want a little bit more confirmation
on a trade like this, and I would only consider it if the groupings looked
pretty good and the larger scale picture warranted it. Of course, the
techniques presented are not limited to use with Fibonacci groupings, I only
use the technique here because it is one of my main ways of trading. Let's
look at a potential trade area for this chart. See figure 2.10.

Figure 2.10

?: CS co 0-0 I!!IiI £J

2 1 . 000

20, 000

1 9 , 000

1 8 , 000

17,000

1 6 , 000

1 5 , 000

11 18 25 Feb 8 15 22 Mar 8 15 22 28 Apr 12

Chart created by Dynamic Trader (c) 1 996-2001

The next step would be to dial down to a smaller timeframe; again in this
case we' l l go to a 60-minute chart. I ' l l also add in the 20-period simple
moving average. See figure 2 . 1 1 .

24
Figure 2 . 1 1

1�!IH Mf�!I
?. CSCO 60-1 1!113 £J
1 6 . 200

1 6 . 000

1 5 . 800

t
1 5 . 600

t tJ
{I
�
1 5 . 400

1 5 . 200

1 5 . 000

1 4 . 800

1 1t 12f AprlSm lGt 17w

Chart created by Dynamic Trader (c) 1 996-2001

C SCO is approaching the potential trade area and is trending up quite

smoothly. It is also getting stretched quite a bit away from the moving
average. This can be good, in the sense that issues tend to be ' mean­
reverting' , that is, they frequently return to the average.

Having the issue stretched right at the point when it encounters the
groupings can stack the odds more in the favor of the trader. On the other
hand, the entry can be quite far down in this case, with the stop loss, perhaps
set j ust above the groupings, very far away. That can be the price for
wanting that much confirmation.

Let's move ahead one bar and see how C SCO is acting. See figure 2 . 1 2.

25
Figure 2 . 1 2

?. CSCO 60-1 1!![3 F3

i�!lH ftf'�
16. 200

j l ll!{
ltt l)
16. 000

Pi
15. 800

jl
15.600

15. 400

15 . 200

15 . 000

14 . 800

l lt 12f AprlSm lGt 17w lf

Chart created by D'ynamic Trader (c) 1996-2001

Wow! CSCO gapped down, ran hard straight up to the groupings, and
reversed on a dime and closed the bar below the opening, and right near the
bottom of the bar. It' s plain to see that it didn't l ike that area one bit. It
seems that it' s now trading quite far away from the area, and it stil l hasn't
closed below the moving average.

If you want more confirmation, this is something you will have to accept. If
you had chosen a faster moving average you' d l ikely be in the trade at this
point. On the other hand, the amount CSCO has dropped in fairly smal l in
the context of the daily chart, and that is the traded timeframe for this trade.
Let's look at CSCO after another bar has unfolded. See figure 2 . 13 .

26
Figure 2 . 1 3

?: CSCO 60-1 BS £I

t�:tH tfmll
16.200

fl w 1 16 . 000

llj )l�
15 . 800

} 1)
15 .600

1 5 . 400

/ 1 5 . 200

1 5 . 000

1 4 . 800

l it 12f Apr15m lGt 17w 1St

Chart created by Dynamic Trader (c) 1 996-2001

CSCO is still coming off the grouping nicely, and has dropped j ust about to
the moving average. A close below the moving average is the trigger. Let' s
advance one more bar. See figure 2 . 1 4.

27
Figure 2.14

?: C S CO 60-1 I!!JI3 I3

t�!tH If�rl
1 6 . 200

lj w 1
16.000

i
lPl t i 1 l
1 5 . 800

1)
1 5 . 600

1 5 . 400

1 5 . 200
�/
1 5 . 000

1 4 .800

l lt 12f AprlSm IGt 17w 1St

Chart created by Dynamic Trader (c) 1 996-2001

That' s it; the trade would now be triggered at this point. The trigger bar is
quite interesting, and worthy of some discussion. The bar opened pretty
close to the previous bar' s close, and it closed about midway on the bar, with
the close below the open. But it was quite a large expansion bar, with a large
lower tail .

I f one was strict with the rules, the bar met the criterion for an entry. This
brings up the question of ' second guessing' . Thi s bar had a lot of lower level
action, and then it reversed and went up quite a bit before it closed. Given
how far CSCO had already dropped, I might be thinking it was time for
some corrective action to the decl ine.

If I followed this l ine of reasoning, I might decide to abort the trade entry, or
sit tight and see how it plays out. The danger is, the trade may play out
exactly as anticipated, and the trade may be missed. I have found that,
although I sometimes 'guess' correctly and save myself a stop out, I
frequently watch the trades I skip do exactly what I originally expected they
do.

28
My point is, for me, I rarely second guess them. Unless I see something
really blatant (and that' s a subj ective, experience based call), I take the trade
if it meets my pre-defined trigger conditions. Let's look at how CSCO
played out shorter term, on the 60-minute chart. See figure 2. 1 5 .

Figure 2 . 1 5

�-:. CSCO 60-1 I!I3 £I

1 6 . 000

1 5 . 500

1 5 . 000

1 4 . 500

1 4 . 000

1 1t 12F AprlSm 16t 17w 18t 1 ':IF Apr22m 23t

Chart created by Dynamic Trader (c) 1996-2001

As it turns out, the entry signal woul d have worked quite well . There was
one bar where it looked l ike CSCO may have been reversing, but that bar
barely exceeded the expansion bar' s high, only going one cent higher. Any
technical stop you l ikely would have been using at this point wouldn't have
even been approached. But remember, the traded timeframe is the daily.
This little blip is hardly noticeable on the traded timeframe. See figure 2. 1 6.

29
Figure 2 . 1 6

?: eseQ 0-0 I!IEf £i

21 .000

20 ,000

19,000

1 8 . 000

1 7 . 000

1 6 . 000

1 5 . 000

1 4 , 000

1 3 . 000

11 18 25 Feb 8 15 22 Mar 8 15 22 28 Apr 12 19 2& Ma

Chart created by Dynamic Trader (c) 1 996-2001

The trade offered quite a bit of a chance for taking profits and letting some
part of the trade ride. When I trade a trend, I usually scale out and take most
of my profits by the time the previous significant swing-low is exceeded and
reaches the area of the 1 .272 external retracement from that swing-low. (The
swing-low I am looking at was on February 27, and the swing-high was on
March 8).

Although many trades exceed this number, few go beyond the 1 .6 1 8 external
retracement. Hence I always put these on my chart, to help guide my trade
management. As I explain in detail in Kane Trading on: Trailing Stops, I
mostly use trailing stops for the majority of my profit taking.

But trailing stops can vary, and how tight you trail them can depend on
where you see potential reversal areas, and areas of potential support and
resistance, coming in. In a nice, reasonably smooth trend, I ' m always
watching the area between the 1 .272 and 1 .6 1 8 external retracement area
very closely, and usually have most of my trade off by the time that area is
reached.

30
Let's look at the same daily chart as in figure 2. 1 6, but with the 1 .272 and
1 .6 1 8 external retracements added on to the chart. See figure 2 . 1 7.

Figure 2 . 1 7

?: CSCO D-D I!! EU!J

22 .000

2 1 . 000

20.000

1 9 .000

1 8 . 000

1 7 . 000

1 6 . 000

1 5 . 000

1 4 . 000

1 3 . 000

------- 1 1 .814 Ro:t 1 . 6 18 1 2 . 000

11 18 25 Feb 8 15 22 Mar :3 15 22 28 Apr 12 1� 26 Ma

Chart created by Dynamic Trader (c) 1 996-2001

It can clearly be seen that CSCO is right into that zone between the
retracements. In a case l ike this I would l ikely have scaled out of most of the
position, using a variety of techniques, with heavy focus on trail ing stops. I
will refrain from any further discussion on trade management here, so as to
stay on topic for this book, but I am pointing this out here for a reason.

I chose this example from the beginning because, although it did work out
quite well, it didn't work out as well as it may have. And it is about to throw
a curve ball at any trader who is stil l with the trade. Let' s move forward one
more bar and see what I am talking about. I wil l remove the anchor l ine from
the swing-low on February 27, to help with clarity. See figure 2. 1 8.

31
Figure 2 . 1 8

?: eseQ D-D !IS F!I

22. 000

2 1 . 000

20. 000

1 9 . 000

1 8 . 000

1 7 . 000

1 6 . 000

1 5 . 000

1 4 . 000

1 3 . 000

------- 1 1 .8 14 Ro:t 1 .& 1:3 1 2 . 000

11 13 25 Feb 3 15 22 Mar 3 15 22 28 Apr 12 19 2& Ma�

Chart created by Dynamic Trader (c) 1 996-2001

BAMM. A huge reversal, with a massive gap up opening. Low to high, this
is over a 32% move in one bar! The spinning top bar in the l .272/l .6 1 8
external retracement area was a big time red flag. But even if you decided to
hold some of the trade at that point, all would not be lost.

The original entry on the trade was at the close of the first 60-minute bar
below the 20-period moving average, and that was at 1 5 .46. Assuming you
would sell on the gap up opening, or thereabouts, the opening price was
1 5 .23 . (Keep in mind that the prices I mention are the prices when you might
be looking to get in and out of the trade. I ' m not suggesting that anyone
could actually have gotten these exact fill prices. I use these prices only as
examples, to get a general idea of what's going on here.) Given this, the
remaining part of the trade would be a scratch.

All in all, 1'd consider that a pretty decent trade. But it still serves as a good
example of one of the many things that can happen in a trade, and how
important good trade management can be.

32
In this chapter we've looked at two different moving average cross entries.
One was entry when the first bar closed above a 1 0-period simple moving
average (this was a long trade), and the other was when the first bar closed
below a 20-period simple moving average (this was a short trade). There are
many other variations, and I encourage you to experiment and find which, if
any, help your trading.

I, mysel f, l ike to use a 5-period simple moving average when the issue is
trading smoothly into the zone and I have a high degree of confidence in the
trade. As you may suspect, this is a very fast setting and is extremely prone
to whipsaws. I use it when I ' m only one step away from j ust wanting to fade
into the trade if it hits the zone and stalls.

When I use this technique and want j ust a bit more confirmation, I use the
1 0-period simple moving average. I rarely use any other variation than the 5
or 1 0-period simple moving averages. I find that I prefer other techniques to
using a longer period average than the ten. You may find, though, that your
preferences may be different.

Keep in mind that you can also experiment with exponential moving
average, weighted moving averages, and other newer generation moving
averages. To simplify this discussion, the following appl ies to long trades;
simply reverse for short entries.

You can look at triggers other than the first bar close above the moving
average. You can look at when the issue first trades above the average, that
is, when it breaks the average with j ust one trade. You can also look at
taking out the high of the first bar that closes above the average, or taking
out the high of the first bar that trades above the average.

The point is, there are a lot of possibilities here, and it' s a great idea to
experiment, on paper, with as many possibilities as you can think up. Find
what works the best for you. Examine and experiment with not only all the
techniques I present in this book, but also with variations on all the
techniques. Try to find the techniques and the variations that suit yours
needs as best as possible.

33
 Chapter 3

Moving Average Crossover

I like to use the moving average crossover entry technique quite a bit. But I
want to point out right up front that many writers malign the technique. I ' d
like to explain why I think that i s before we proceed to the technique itself.

One of the things you most often hear is that moving average crossover
' systems' don't 'test out' . What that means is, in backtesting, they don't
make money overall . They may make money in the trending phases of the
market, but I agree that they don't make money overall .

But I ' m not talking about using a moving average crossover to trigger you
into a trade, and then to take you out of the trade, as these simple moving
. average crossover ' systems' describe. No, far from it. I ' m talking about
using a moving average crossover to trigger you into a trade, as j ust one
small part of a comprehensive 'Plan for a Trade' , as I outline in my free
article Kane Trading on: The Critical E lements of a Trade.

In a simple moving average crossover ' system' , the crossover itself is the
sole criterion for taking the trade. But here we are discussing an entry
technique that will trigger us into a trade; a trade that we have already
decided is in a potential trade area. This trade area is part of our 'Plan for a
Trade' , and is completely independent of the entry trigger, in this case a
movmg average crossover.

It' s important to understand that the crossover is in no way a stand-alone

criterion to trade. It' s j ust a simple trigger to initiate a trade in an area we
have already decided we want to initiate a trade. That's one maj or
difference, and here's another one. In a moving average crossover ' system' ,
you use the moving averages on the traded timeframe.

Remember, we first decide on a potential trade area, and then we drop down
to a lower timeframe and look for a trigger. If you want to use a moving
average crossover trigger, you will be doing it on a lower timeframe than
your traded timeframe. This completely breaks this technique away from any
relationship with a moving average crossover ' system' .

35
Lastly, since we look at trade management as an entirely separate critical
element of a trade, we are free to choose any management technique that we
want. We can choose to manage this trade without any moving averages of
any kind, if that's how we want to go. So you can clearly see that although
you may use a moving average crossover (on a lower time frame ) as an entry
trigger, that use of the phrase is, pretty much, the only commonality with a
moving average crossover ' system' .

Hopefully, this has lain to rest any negative thoughts that reference past
thinking about such ill-conceived ' systems ' as the moving average crossover
' system' . And hopefully, you can see what I am presenting for what it is, and
no more: j ust one more technique that you should experiment with, and
possibly consider, to help trigger you into a trade, in the context of your
entire 'Plan for a Trade'.

I explained all this in such detail to clarify what I am presenting here. I

always try my best to explain what it is I am trying to accomplish in my
presentation, and in what context I am suggesting you experiment with it and
see if it may help you. This technique, l ike everything I present, is j ust one
small part of a much larger overall plan and context. It is important that you
don 't get confused, in any way, about the role of this or any technique I
present.

With all that said, let' s look at the technique. What is a moving average
crossover, and how can I use it? A moving average crossover is simply the
plotting of two different moving averages on a chart, and having the ' faster'
average cross above or below the ' slower' average. This indicates that the
shorter-term trend has now taken precedence, at least for the time being.

It is for this reason that we may choose to use this crossover as an indication
that the trend has now started in the direction we are anticipating. Bear in
mind, though, this is in no way a guarantee that this start of the trend will
continue at all . You simply have to let nature take its course, so to speak.
You chose an entry based on a potential trade area, and now you let it play
out as dictated by your trade management plan.

As many who have studied moving average crossover ' systems' have noted,
crossovers are prone to massive whipsaws. I feel this is due, in part, to the
fact that the market spends a lot of its time see-sawing back and forth and all
around. As it does this, the moving averages will cross and re-cross

36
repeatedly, causing these horrendous whipsaws. But the bottom line is, many
of these areas woul d not be areas where you would even consider taking a
trade. So why would you take the trade there because the seesaw action
caused the moving average to crossover?

The stock answer is that you must take all the signals you get, or something
to that effect. What I ' m pointing out is: there was no reason to believe you
had any edge for a trade at that point. Hence, any entry signal would be
wisely ignored. Now l et' s j ump ahead to what we are looking at.

We have an area where we have decided that we want to trade. Any hint that
the trade is setting up and triggering would be all we want. We see a
crossover, so we initiate the trade. This is not an area we expect seesaw
action. We expect something to happen here.

There is no guarantee anything will happen in this potential trade area. There
is also no guarantee that the market won't begin to seesaw the exact second
that we initiate the trade. What we do know is that our work and research
and due diligence have led us to believe we have an edge in this potential
trade area, and that we want to attempt to capitalize on that edge.

That is where the difference lies. We use the crossover when we have
decided this is an area we can expect something to happen. When you see
people use moving average crossovers to whipsaw themselves into oblivion,
you will usually find these are areas where, if you j ust looked at the chart
action without the moving averages on it, you would expect the market to
seesaw. It is rarely a surprise.

Let's move on to some examples. I am going to apply two favorite moving

average crossovers to the two examples from the last chapter. These two
' settings' , if you can call them that, are by no means anything special . I like
to use these two the best, but that is because they fit my style and the
markets I trade the best. As I outlined in the last chapter, you should
experiment with the concept and find what works best for you, if anything.

My most favored setting is using a 5-period and a I 5-period simple moving

average. My next favorite is a I O-period exponential and a 20-period simple
moving average. The latter setting allows me to use Bollinger Bands around
the longer period moving average, the 20-period, as that is the default setting
for the bands, and my most commonly used setting for them.

37
In the last chapter, our first example was AOL on a daily chart. Let's look,
again, at the trade as it enters the potential trade area. See figure 3 . 1 .

f�
Figure 3 . 1

?. AOL 0-0 I!GU3

1 4 . 000

jjHti) 1
1 3 . 500

j) II,IUI,MIIII 1 3 . 000

1 2 . 500

III
1 2 . 000

t 11 j

) lhlJ tjpl
1 1 . 500

1 1 . 000

1 0 . 500

1 0 . 000
Feb 14 21 28 Mar 14 21 28 Apr 11 17 25
Chart created by Dynamic Trader (c) 1 996-2001

Now let's look, again, as it enters the potential trade zone on the 60-minute
chart. I ' ll add on the 5-period and I 5-period simple moving averages, j ust as
I added the l O-period simple moving average in the l ast chapter. See figure
3 .2 .

38
Figure 3.2

?. AOl 60-1 !lEi F!I

1 4 . 200

1 4 . 000

1 3 . 800

1 3 . 600

1 3 . 400

1 3 . 200

1 3 . 000

1 2 . 800

1 2 . 600

Apr 22t 23w 24( 25f Apr 29t 30w 1t 2f May 6t 7w

Chart created by Dynamic Trader (c) 1 996-200 1

Notice how the 5-period simple moving average never crossed the I 5-period
simple moving average during all that ' chop' that occurred around May 1 -3 ,
despite multiple bar closings above the I 5-period simple moving average.
You can see how this technique, in general, will be less sensitive, and hence
less prone to false signals. The trade-off? (You know there will always be
one ! ) Most of the time you will get in later and be farther from your stop.
You pay a price for more confirmation. Let's move ahead in time. See figure
3.3.

39
Figure 3 .3

?: :AOt 60-1 !IS F.3

1 4 . 200

1 4 . 000

1 3 . 800

1 3 . 600

1 3 . 400

1 3 . 200

1 3 . 000

1 2 . 800

1 2 . 600

Apr 22t 23w 241 25F Apr 2'lt 30w 1t 2F May 6t 7w 8t

Chart created by Dynamic Trader (c) 1 996-200 1

As noted before, AOL is trending smoothly into the potential trade area, and
now with the two moving averages as a guide, the behavior again looks very
good. Let's move forward, j ust a bit more. See figure 3 .4.

40
Figure 3.4

?. AO L 60-1 1!!1 13 £I
1 4 , 200

1 4 , 000

1 3 , 800

1 3 , 600

1 3 . 400

1 3 , 200

1 3 , 000

1 2 , 800

1 2 , 600

Apr 22t 23w 24t 25f Apr 2�t 30w 1t 2f M.ay &t 7w 8t
Chart created by Dynamic Trader (c) 1 996-2001

AOL is trending down nicely into the potential trade area, with the moving
averages trending down nicely j ust above the price action. So far the trade is
behaving excellently with respect to the moving averages. Let's go forward
a few more bars, again. See figure 3 .5 .

41
Figure 3 . 5

?: A O t 60-1 !lEI £J
1 4 . 200

1 4 . 000

1 3 . 800

1 3 . 600

1 3 . 400

1 3 . 200

1 3 . 000

1 2 . 800

1 2 . 600

Apr 22t 23w 24t 25f Apr 2<jt 30w 1t 2F May Gt 7w St "If
Chart created by Dynamic Trader (c) 1996-2001

AOL comes off the area and starts to move sideways. It is now moving
above and below the very fast 5-period moving average, which is now
turning up slightly. Let's add one more bar to the chart. Figure 3 .6 .

42
Figure 3.6

?. AO l 60-1 I!!l IiI £I

1 4 . 200

1 4 . 000

1 3 . 800

1 3 . 600

1 3 . 400

1 3 . 200

1 3 . 000

1 2 . 800

1 2 . 600
221 23w 241 25f Apr 2'31 30w 11 2f M.ay 61 7w 131 '3f r
Chart created by Dynamic Trader (c) 1 996-2001

We now have a bar close above both moving averages, but nothing close to a
cross of the 5-period average above the I 5-period average. So although AOL
is moving up here, we don't have a trigger. Let's look at the same chart from
the last chapter, with the I O-period simple moving average on the chart.
Recall that the trigger was the first close above the average. See figure 3 .7 .

43
Figure 3.7

?: AD L 60-1 !!I8 D
1 4 , 200

1 4 , 000

1 3 , 800

1 3 , 600

1 3 . 400

1 3 . 200

1 3 . 000

1 2 . 800

1 2 . 600
22t 23w 24l 25f Apr 29t 30w It 2f May Gt 7w 8t 9f
Chart created by Dynamic Trader (c) 1 996-2001

Using the I O-period simple moving average cross trigger, the trade was
triggered at this point. But using the 511 5-period simple moving average
crossover trigger, the trade is stil l in the wait and see mode. It general ly
takes more to get a trade to trigger with the crossover technique. Let's add
one more bar and see what unfolds. See figure 3 .8 .

44
Figure 3.8

::::: AOl 60-1 1!19£J

1 4 . 200

1 4 . 000

1 3 . 800

1 3 . 600

1 3 . 400

1 3 . 200

1 3 . 000

1 2 . 800

1 2 . 600

Apr 22t 23w 24t 25f Apr 2'3t 30w 1t 2f May &t 7w 8t '3f '"
Chart created by Dynamic Trader (c) 1 996-2001

We now have two bar closes above both averages, but still no crossover. As
I ' m wont to say: "No trigger, no trade". You can see that a crossover is very
near at hand, and I would suspect that any more upward movement will
bring the crossover, but I don't recommend j umping the gun on triggers.
Let's add one more bar and see what the moving averages do. See figure 3 .9.

45
Figure 3.9

?: AOL 60-1 1!!r;J £!

1 4 . 200

1 4 . 000

1 3 . 800

1 3 . 600

1 3 . 400

1 3 . 200

1 3 . 000

1 2 . 800

1 2 . 600
22t 23w 24t 25F Apr 29t 30w it 2F May f>t 7w 8t 9F M.
Chart created by Dynamic Trader (c) 1 996-200 1

Well, finally, that's it. The trade is now triggered; we have a moving average
crossover. The closing price on this last bar is $ 1 3 . 1 7. Compare this with the
previous example using the 1 0-period simple moving average cross entry,
which had the trigger bar's close at $ 1 3 . 1 0. You paid more, but not very
much, for the added confirmation.

Now, the questions that undoubtedly come to mind are : ' So, what is the
actual trigger, when the moving averages j ust crossover the first time, or
when they crossover and the bar closes? And how about the price action?
Does it matter if the price is above the averages, or not?' These are great
questions.

Hopeful ly, you have an idea what I ' m going to say here. What I ' m going to
say is that you must experiment and decide what works best for you. I can
(and will) tell you what I like best, but that is for my trading. I can't say it' s
going to be the best for you. I , personal ly, usually like to wait for the first
bar close after the crossover, and I like the price above the averages for long
trades (below for shorts, of course).

46
But this is not a hard and fast rule. One thing I frequently see is that many
trades will do a small pullback j ust after the crossover occurs. This will
frequently bring the price action back down below the averages. It can even
re-cross the averages back down; only to have them cross back up and off
the trade goes.

If you choose a stop, for example, j ust below the groupings you used to
determine the potential trade area, this small pullback will not come
anywhere near your stop. Hence, the crossover gets you in, and then you
forget all about the moving averages. You don't use them in any way for
your stop action.

What's the point to all this? I have a few techniques where I initiate the trade
on this small pullback, which will most frequently have the averages crossed
up as discussed, but the price dropping back below them. I will discuss this
variation further in a later chapter. Suffice it to say, though, j ust as with the
variations we discussed in the last chapter, you have to tinker around with
them and decide what, if anything, works for your trading.

I won't repeat chart examples showing how the AOL trade played out, since
we have already covered that in detail in the last chapter. The reader is
referred to the previous charts in chapter two for review, if needed. Instead,
let's look at this example again, this time using the I O-period exponential
and 20-period simple moving average crossover trigger.

47
Let' s j ust j ump right in and show the same potential trade, right before the
moving averages might be getting ready to crossover. This is right at the
same point where the 511 5-period set were j ust getting ready to crossover.
See figure 3 . 1 0.

Figure 3 . 1 0

1 4 . 200

1 4 . 000

1 3 . 800

1 3 . 600

1 3 . 400

1 3 . 200

1 3 . 000

1 2 . 800

1 2 . 600
22t 23w 2<k 25f Apr 2'3t 30w 1t 2f May f>t 7w 8t '3f W
Chart created by Dynamic Trader (c) 1 996-2001

So here, too, the averages are l ooking l ike they are about to crossover, but
haven't yet. Recall that with the 5/ 1 5 moving average combination shown in
figure 3 .9, they crossed on the next bar. Remember that the 1 0/20 moving
average combination shoul d provide even more confirmation (at a price, as
always), and hence generally will be even later. Let' s see what happens. See
figure 3 . 1 1 .

48
Figure 3 . 1 1

?. AOl 60-1 1!!8 £.1

1 4 , 200

1 4 , 000

1 3 , 800

1 3 , 600

1 3 . 400

1 3 , 200

1 3 , 000

1 2 , 800

1 2 , 600

Apr 22t 23w 241 25f Apr 2'lt 30w it 2f May 6t 7w 8t 'If M.
Chart created by Dynamic Trader (c) 1 996-2001

Interesting, the averages haven't crossed, like they did with the 511 5
combination. This is what one would suspect, but it is still interesting to see
it play out. If AOL keeps going up a crossover would be expected on the
next bar. Let' s move ahead one more bar and see if we get the crossover. See
figure 3 . 1 2.

49
Figure 3 . 1 2

?, A O L 60-1 1!!!18 E1
1 4 , 200

1 4 , 000

1 3 , 800

1 3 , 600

1 3 . 400

1 3 , 200

1 3 . 000

1 2 , 800

1 2 , 600
22t 2)w 24t 2sf Apr 29t 30w 1t 2f May 6t 7w 8t 9f Ma
Chart created by Dynamic Trader (c) 1 996-2001

Okay, there it is. It's j ust barely a crossover, but it is enough. The closing bar
price is $ 1 3 . 1 9. Recall that the closing price for the 511 5 combination was
$ 1 3 . 1 7, and for the 1 0-period by itself it was $ 1 3 . 1 0.

So more confirmation was obtained by going to a crossover technique and

going to longer period averages, but at the cost of a higher entry price. Also
note, as I ' m sure you have, at j ust how little the actual price difference is.
This will not always be the case, but it should be clear that the variance in
these techniques is a lot less than one might expect at first glance.

I would also l ike to take a second here to point out an observation on the
actual crossover itself. As you watch a crossover develop in real time, you
will frequently see the crossover form and then ' uncross' as the price
changes. I generally use c losing bar prices to generate indicators, and most
charting software will form the c losing price hash mark on the last bar from
the last trade. But until this bar has completed and the next bar has started,
this value will change as the last traded price changes.

50
Since this 'closing' price is used, in this case, to calculate the moving
averages, the last value of these averages wil l change repeatedly, until the
final value is in. For this reason, I don't consider a crossover a crossover that
I woul d trade until the bar that creates the crossover has completed.
Otherwise it may be a ' headfake' , and the crossover wil l be j ust a 'phantom
crossover' .

That's not to say that you woul d not want to explore such pre-mature
signals. I j ust want to point out what can happen, and leave it up to you to
decide how you want to design your entry signal. Just l ike there were many
options to how one might handl e the moving average cross entry technique,
there are many potentially useful ways to use the moving average crossover
technique, too.

At this point, l et's take a brief look at how these two crossover entries
compare to the two moving average cross entries presented in the last
chapter, using our second example with CSCO. F irst, for review, let's look
again at the 60-minute trigger chart of C SCO, using the 20-period simple
moving average cross entry. This is the same chart shown as figure 2 . 1 4. See
figure 3 . 1 3 ,

51
Figure 3 . 1 3

?. eseQ 60-1 1!!l8 £'J

1�!IH h!�.tI
1 6 . 200

1 6 . 000

J i+ t l
1 5 . 800

1) p
1 5 . 600

1 5 . 400

1 11
1 5 . 200

1 5 . 000

1 4 . 800

l lt 12f Apr1Sm 16t 17w 1St

Chart created by Dynamic Trader (c) 1 996-2001

As discussed in the last chapter, the trade would now be triggered at this
point, using the 20-period simple moving average cross entry. For the sake
of the example, and so we can do some comparisons, the c losing price on the
trigger bar is $ 1 5 .46. This is, of course, not to suggest that anyone could or
would have gotten that exact price in a trade, but it helps us compare
techniques to each other.

Let's now look at the three other techniques that we want to compare to this
example. First, let's look at the same technique, the simple moving average
cross, but with a shorter period average, the 1 0-period. This is the trigger we
used in the first AOL example in the last chapter. See figure 3 . 1 4.

52
Figure 3 . 1 4

?: CSCO 60-1 1!13 F!'I

1 6 , 200

1 6 , 000

1 5 , 800

1 5 , 600

1 5 . 400

1 5 , 200

1 5 , 000

1 4 , 800

1 11 12f AprlSm lGI 17w 1€

Chart created by Dynamic Trader (c) 1 996-2001

Notice how the shorter 1 0-period simple moving average triggered the trade
two bars earlier than the 20-period, as shown in figure 3 . 1 3 . The closing
price of the trigger bar in this case, is $ 1 5 .65 . You would have gotten a fairly
higher price using the ' quicker' trigger, as you might expect. Let's now see
how these two compare to the two crossover triggers. F irst, l et' s look at the
'quicker' 511 5 crossover. See figure 3 . 1 5 .

53
Figure 3 . 1 5

?: CSCO 60-1 .
1!13 £J

1 6 . 200

1 6 . 000

1 5 . 800

1 5 . 600

1 5 . 400

1 5 . 200

1 5 . 000

1 4 . 800

l lt 12f Apr1Sm 16t 17w 1St

Chart created by Dynamic Trader (c) 1 996-2001

This entry triggered exactly as expected, after the two cross triggers. I t
triggered one bar after the ' slower' 20-period cross trigger. The closing bar
price was $ 1 5 .3 3 , l ower than either of the cross triggers. Let's now look at
the 1 0-period exponential, 20-period simple moving average crossover
entry. See figure 3 . 1 6.

54
Figure 3 . 1 6

?. CSCO 60-1 I!IEI EI

�NH ��1I
1 6 , 200

l w 1
j
1 6 , 000

11 t l II
1 5 , 800
/"
1 5 , 600

P
.--
--"
i + /-/
//-
1 5 . 400

1 5 , 200

1 5 , 000

1 4 , 800

1 1t 12f AprlSm lGt 17w 18t

Chart created by Dynamic Trader (c) 1 996-2001

This trigger was even later, as one woul d expect, with a closing bar price of
$ 1 5 .3 8. But a very interesting thing happened. That price is actually five
cents higher than the l ast trigger using the ' quicker' crossover with the 51 1 5
combination. Now why woul d that be?

One thing I notice quite often is that if you are looking for confirmation
before you initiate your trade i.e. you aren't ' fading' the entry right at the
trade area, you are usually quite a bit away from the trade area. The general
area you see as 'enough confirmation' is also perceived by a lot of other
traders and money managers. This is also an area where profit taking may
set in for those that did fade the entry for a quick trade.

Hence, this confirmation area is frequently an area where a pullback may

occur. Sometimes the price may pull back and j ump above the moving
averages (on a short trade; reverse, of course, for longs). This would be a
better entry, but then you woul d be taking a trade when the price is on the
'wrong' side of the moving average, and moving against you at that
particular moment.

55
Personal ly, I don't like to take trades like that unless I am particularly
looking to enter the trade on a pullback after a crossover. And if I am
looking for that type of a pullback, I wil l usually have some additional
criteria that I am l ooking for, before I would enter. The point I am trying to
make is that many times you will see some type of a move against you after
these triggers i.e. you ' l l have to 'take some heat' .

If the trade is going to be one of the trades in the 'winners group' , this heat
wil l rarely threaten your stop. If you wait for the heat and try to enter at that
point, you may get a better price, but you will miss a lot of trades. This
would be more l ike the ' first pullback after a crossover' technique. It is a
very valid technique, but is in the group of high confirmation techniques. It
can be a very useful and very good technique, but its occurrence wi ll be very
infrequent, as compared to the techniques j ust presented.

Remember that I prefer to use the ' least restrictive' technique that I can. The
pullback after a crossover is only going to suit my needs if I want a great
deal of confirmation. I mention all this because I want to make it clear that
you won't always get a better price j ust because you used a lesser
confirmation technique. Usually this wil l be the case, but since a pullback
may come at any time, in individual cases anything can happen.

Take a look at some heat that woul d have to be taken on this CSCO trade.
With the last trigger, the trade woul d have been initiated after three inside
bars. Another inside bar printed after these three, and this amount of
sideways uncertainty is sure to resolve itself soon. Many times the first move
will be a headfake, and the 'real ' move wil l be j ust the opposite. Look, first,
at the initial move. See figure 3 . 1 7.

56
Figure 3 . 1 7

?. CSCO 60-1 !tEl £!

i�!iH _i,rI
1 6 . 200

jlwl
1 6 . 000

1 5 . 800

J!lj f%B
�./.-'"
II
1 5 . 600

1 5 . 400

1 5 . 200

1 5 . 000

1 4 . 800

1 1t 12f AprlSm 16t 17w 18t l'

Chart created by Dynamic Trader (c) 1 996-2001

CSCO gapped up huge, and above both the moving averages, trading one
cent above the high of the expansion bar down that was used to create the
four inside bars. But it couldn't hold this high and sold off rapidly, creating a
large expansion bar down that closed near its low.

Even if this initial heat would have been very scary to watch, it' s likely you
had your stop somewhere above the grouping in this case. Remember, the
trade timeframe is the daily. If you look on the entry timeframe and try to
micromanage the trade, not only can it scare you out of trades that you
shouldn't get out of, it takes your perspective away and creates impressions
that are out of context for the trade.

Hence, this gap opening l ikely is j ust the type of pullback I was talking
about, and as you remember from the previous chapter, CSCO did drop
considerably. Let's look, one more time, at what CSCO did after it gave the
trader a little heat. See figure 3 . 1 8.

57
Figure 3 . 1 8

?:': eseo 60-1 !le E!

1 6 , 000

1 5 , 500

15 ,000

1 4 , 500

1 4 , 000

1 11 12f AprlSm 161 17w 181 19F Apr22m 231

Chart created by Dynamic Trader (c) 1 996-2001

That gap up was it, and it went basically straight down from there, trading as
low as $ 1 2.28 before reversing back up.

Try not to lose perspective on the trade when you drop down to a lower
timeframe for your entry. Don't expect every entry to be the best possible
point of entry, or to not take any heat. Have your stops pre-decided and let
the trade play out.

I have never had much luck trying to second-guess trades after I have
opened them, especially by micromanaging on the lower timeframe. Just be
aware that pullbacks are common after most of these types of entries. Be
prepared with your trade plan, and thi s type of heat will j ust become part of
the routine.

Let's move on to the next entry technique, one that, in many ways, is similar
to the moving average cross technique, and that is trendline violations.

58
 Chapter 4

Trendline Violations

The next technique I wil l present I call the trendline violation technique, and
for obvious reasons. The trigger into the trade will be when a trendline is
violated by the price action. But violation of what trendline? And when do I
l ike to use the technique?

I am really partial to the trendline violation technique when an issue is

oscillating along an axis as it heads into a potential trade area. This type of
price action is very difficult to use with moving averages, since you may get
multiple false signals. In general, this oscillating price action can be difficult
to trade.

If you still want to trade an issue that is demonstrating this type of price
action, perhaps because the traded timeframe setup is particularly nice, this
can be a good technique to use for entry. Understand, though, what the
expression ' difficult to trade' means. The bottom l ine is, it really means your
stop wil l have to be wider and/or you' l l hit fewer trades . And perhaps make
less when you hit one.

That all translates to making less, on average, per trade. That, then, alters the
reward/risk ratio. Just because you can trade something doesn't mean it fits
into your game plan. Sometimes it does, but what I want to point out is,
sometimes it doesn't. Fully assess trades that are more ' difficult' .

Let's look at AMZN, which, in this example, is headed for a potential trade
area on a daily chart. The potential trade area is a huge Fibonacci support
zone, based on the groupings technique. This is an exampl e from my book
Kane Trading on: Advanced Fibonacci Trading Concepts. Keep in mind, the
entry techniques presented here are not specific to the Fibonacci groupings
techniques only, I j ust use them for examples because they are one of my
main ways of trading. See figure 4. 1 .

59
Figure 4 . 1

?: AtwlZN 0-0 > >

I!!I EI f!4
1 7 , 000

��
1 6 , 000

1 5 , 000

1 4 , 000

�HI
��"\lrJ\ �j
1 3 , 000

1 2 , 000

N� \Jh'�
1 1 . 000

1 0 , 000

�
, 000

,000

, 000

Oct Nov Dec 02 Feb Mar A�

Chart created by Dynamic Trader (c) 1 996-2001

The assumption on this trade is that the uptrend will reassert itself, reversing
off of the area of one of the three F ibonacci groupings. At this point an entry
technique is sought, and that would be on the lower timeframe. In this
example we' l l look at the 60-minute chart.

For the sake of clarity, I wil l not inc lude all the Fibonacci groupings on the
60-minute chart. We can note that the first grouping wi ll be reached as soon
as the price drops below $ 1 2.80. Once that has happened, i .e. the potential
trade area has been 'penetrated', any valid trigger that I have decided on
would be accepted. Let's look at AMZN on a 60-minute as it approaches the
potential trade area. See figure 4.2.

60
Figure 4.2

?: AMZN 60-1 1!!!I[3' £.i

1 7 . 000

���
1 6 . 500

1 6 . 000

1 5 . 500

��
1 5 . 000

1 4 . 500

1 4 . 000

1 3 . 500

1 3 . 000

8 15 22 AprS
Chart created by Dynamic Trader (c) 1 996-2001

AMZN has now traded into the potential trade zone here, with the low on the
last bar being $ 1 2.7 1 . At this point I would now be looking for my trigger,
whatever trigger I have pre-chosen, to initiate me into a trade. For
comparative purposes, let's add in the 5 and 1 5-period simple moving
averages to the chart, and see what observations we might make. See figure
4.3 .

61
Figure 4.3

?. AhtZN 60-1 II!EI EJ

1 7 . 000

1 6 . 500

1 6 . 000

1 5 . 500

1 5 . 000

1 4 . 500

1 4 . 000

1 3 . 500

1 3 . 000

8 15 22 AptS
Chart created by Dynamic Trader (c) 1 996-2001

Notice how AMZN is oscillating up and down in fairly broad swings as it

trends down to the potential trade area. This doesn't preclude the use of a
moving average crossover trigger, but it does make me less inclined to use it.
In my experience, I have seen more false starts when you see action like this.

Admittedly, though, when I look at figure 4.3 I have to say that it is fairly
orderly. Hence, based on what information I have up to this point, this is an
example where I almost feel it is a coin toss between using the moving
average cross trigger and 'paying up' for more confirmation and waiting for
a trendl ine violation. Let's add a trendline in and reassess. See figure 4.4.

62
Figure 4.4

?: AMZN 60-1 1!!!r;J l:I

1 7 . 000

1 6 . 500

16.000

1 5 . 500

8 15 22 AptS
Chart created by Dynamic Trader (c) 1 996-2001

I drew the trendline from the first two points where you can see contact
between the price and the l ine. It' s amazing how the third point of contact
was such a sharp reversal, right at the contact point. I would hazard a guess
that a few people were watching that trendline. The fourth point of contact
slightly exceeded the trendline, with no closes above the l ine. It then sharply
reversed once again and set a new low, this time right at the top of the
groupings. Let's move ahead another four bars and see what's happening.
See figure 4.5.

63
Figure 4.5

_ c' x
1 7 . 000

1 6 . 500

1 6 . 000

1 5 . 500

1 5 . 000

1 4 . 500

1 4 . 000

1 3 . 500

1 3 . 000

8 15 22 Api'S
Chart created by Dynamic Trader (c) 1 996-2001

The price has dropped a little bit further, but it is a very smooth, orderly
decl ine since the reversal off the trendl ine. I particularly like using the
trendline at this point because it has been such a great guidel ine thus far.
Many traders are likely watching it, and a violation will surely initiate action
on the part of some of these observers.

The downside? Everyone is watching a trendline that is guiding trade so

wel l . That may make it difficult to use when the entry time comes. A
moving average crossover trigger would likely come first, and one might opt
to use that as an entry to anticipate the trendl ine violation.

One possible option is to use the moving average crOSsover trigger, and if
the trendline isn't then taken out as expected, close the trade. At that point
the trade would l ikely be a small profit or a scratch. Let's move another five
bars ahead. See figure 4.6.

64
Figure 4.6

_ c x
1 7 . 000

1 6 . 500

1 6 . 000

1 5 . 500

1 5 . 000

1 4 . 500

1 4 . 000

1 3 . 500

1 3 . 000

8 15 22 AprS 12
Chart created by Dynamic Trader (c) 1 996-2001

AMZN has come off the top of the groupings very strongly, and we now
have a moving average crossover trigger. Keep in mind, we are examining
the trendline violation technique in this chapter, but I want to watch the
moving average crossover trigger at the same time, for the additional
educational value this should provide.

As you look at this chart, you can clearly see we have a battle going on
between the very short-term uptrend that has begun from the reversal off of
the top of the groupings (and pointed out to us by the moving average
crossover trigger to the upside), and the downtrend, as clearly defined by the
down trendline. Why would I play a trade like this, then?

Recall that my traded timeframe is the daily chart, and I have formed my
potential trade area based on that daily chart. The downtrend that is guided
by the trendline is on my entry chart, and that is a lower timeframe 60-
minute chart. I expect the daily chart wil l prevail , and the uptrend will
reassert itself.

65
I find that in almost all the trades I look at, there is always a battle between
two (or more) timeframes, and I am forced to play one against the other. As I
have outlined in some of my other works, I call this 'the trader's dilemma ' .
Unless I am speci fically trading against the larger timeframe trend, I just
about always bet on the larger trend. In this case, we would be betting on the
larger timeframe trend by going long at this point. Let' s note that the closing
bar price after the moving average crossover here is $ 1 3 .5 3 . Now let's move
ahead another six bars. See figure 4.7.

Figure 4.7

_ c x
1 7 . 000

1 6 . 500

1 6 . 000

1 5 . 500

1 5 . 000

1 4 . 500

1 4 . 000

1 3 . 500

1 3 . 000

8 15 22 AptS 12
Chart created by Dynamic Trader (c) 1 996-2001

AMZN is now right at the trendline. It' s do or die time, when the buyers and
sellers will be making critical decisions. My expectation, based on my
experience, is that the most common scenario is for the i ssue to pullback,
hence pulling in some shorts who expect the trendline to hold, and then once
it' s dropped a l ittle bit, the real buyers will step up and take all the rest of
what the shorts have to sell, and it will then tum up. Once the trendl ine is
then taken out the shorts will cover and off it will go.

66
I can hear a lot of readers right now saying ' Well, we know what is going to
happen on the next few charts now, since he already knows the outcome, or
he wouldn't have j ust set us all up for that scenario. ' Actually, no. I do know
what the outcome was already, since this is past history we are reviewing,
but I put that comment in j ust as I would have thought about it as this
unfolded in real time. And as you'll see, that scenario isn't how it played
out. Let' s first move j ust one more bar ahead. See figure 4.8.

Figure 4.8

_ c x
1 7 , 000

1 6 , 500

1 6 , 000

1 5 , 500

15 ,000

1 4 , 500

1 4 , 000

1 3 , 500

1 3 , 000

8 15 22 Ap� 12
Chart created by Dynamic Trader (c) 1 996-200 1

AMZN has continued up and violated the trendline with a firm close above
the trendline. Whether you chose a price cross above the trendline or a close
above the trendline (or anything else you may have decided on ahead of
time), the trade would now be initiated at this point.

For our example, let's note the trigger bar's closing price, which is $ 1 4.05 .
This i s a full fifty-two cents above the closing price for the moving average
crossover trigger bar. I f you waited for this trigger, you paid up a lot more. I
would consider this if I felt I got something for that extra money I paid, and

67
in this case what I would feel I got was a greater likelihood this trade is
going to work out as I planned it.

Let's move ahead one bar and see how the market is reacting to this
'breakout' above the trendline. Note that AMZN has been going nearly
straight up for $ 1 .57. I prefer not to hit an area of strong support or
resistance that I hope to break with the issue possibly running out of steam.
I ' d prefer to see a few rest stops before it hits the critical area.

Also note that, as I mentioned before, it' s common for an issue to pullback
near or shortly after a moving average crossover with the type of period
lengths we have been looking at. On the timeframe we are looking at, the
60-minute, we didn't get any notable pullback after the crossover. This may
mean we are out of steam and the slight trendline violation won't hold, or
that this thing is strong and we, as they say, ' are gone' . See figure 4.9.

Figure 4.9

?: AhtZN 60-1 I!![;J £J

1 7 , 000

1 6 , 500

1 6 , 000

1 5 , 500

1 5 , 000

1 4 , 500

1 4 , 000

1 3 , 500

1 3 , 000

8 15 22 Ap� 12
Chart created by Dynamic Trader (c) 1 996-2001

AMZN exceeded the ' breakout' bar by one cent, but otherwise essentially
printed an inside bar. This is a period of indecision, and all we can do is wait

68
and see how it p lays out. Our stops would already be set and we would do
nothing but let the trade play out.

I j ust l ike to assess what I see to help evaluate trade behavior and see i f I can
learn anything useful that I may use in my future game plan. My reason for
such detail here is to help you see what my thought process is and how I
come to make the decisions I do, based on 'experience' . I ' m trying to share
some of my experience, simulating how I get some of that experience. Let' s
go forward and see how this is playing out. See figure 4. 1 0.

Figure 4 . 1 0

::''1. AMZN 60-1 !lEI £I

1 7 , 000

1 6 , 500

1 6 , 000

1 5 , 500

1 5 , 000

1 4 , 500

1 4 , 000

1 3 , 500

1 3 , 000

s 15 22 AptS 12
Chart created by Dynamic Trader (c) 1 996-2001

Well, AMZN went up a bit after the ' inside' bar, and now has pulled back
below entry, right to the trendline. This is all normal behavior and may be
the rest before the launch I frequently see. Regardless of the stop you chose,
it would likely be well below this tiny pullback, and not threatened at all at
this point. Again, we are j ust trying to glean educational information out of
the price behavior at this point, and all we can do is let the trade, itself,
develop, as we stick to our pre-determined game plan. Let' s move forward
and see how the players are going to react to this small dip. See figure 4. 1 1 .

69
Figure 4 . 1 1

1 7 . 000

1 6 . 500

1 6 . 000

1 5 . 500

1 5 . 000

1 4 . 500

1 3 . 500

1 3 . 000

8 15 22 Ap� 12
Chart created by Dynamic Trader (c) 1 996-2001

They decided to take the small dip, and AMZN i s now clearly broken out
above the trendline. The slight pullback was a quick 'test' that was bought,
and the trade i s now working out quite wel l so far. But remember, the traded
timeframe is the daily chart, so I ' m not going to try and m icromanage the
trade on this lower time frame 60-minute chart. F irst, let's see how the trade
played out longer-term, on the 60-minute chart. S ince the chart had to be
compressed quite a bit to show all the data, we' l l follow that up with the
daily chart, to show it in its proper perspective. See figures 4. 1 2 and 4.13 .

70
Figure 4 . 1 2

?. AMZN GO-f !lS I.!

20 ,000

1 9 , 000

1 8 , 000

1 7 , 000

1 6 , 000

1 5 , 000

1 4 , 000

1 3 , 000

8 15 22 AptS 12 19 26 May3 10
Chart created by Dynamic Trader (c) 1 996-2001

Figure 4 . 1 3

j
20 ,000

rJ
1 8 , 000

16,000

� rt
1 4 , 000

f }�;W\l>� j
¥
N�Jkt¥ ,000

,000

Oct Nov Dec 02 Feb Mar Apr May

Chart created by Dynamic Trader (c) 1 996-2001

71
AMZN rested for a while after the trendline break, but hardly pulled back at
all. If your game plan allowed you to be patient i.e. it had no type of fairly
short-term time stop, you would have been able to capitalize on quite a
strong up-move. This up-move was very consistent with the type of moves
already seen on the daily chart, and should have come as no surprise in the
context of your overall plan.

In looking back at this example, it is fairly clear that this was a trade where
you couldn't clearly see what the best choice was for an entry plan. By this, I
don't mean that you couldn't see any good choices, but that one choice
didn't j ump out as clearly superior over the rest, at least in my experience.
This is quite commonly the case.

Since you have no idea what the future holds, you won't ever know until
after the fact what the best choice woul d have been. This is something one
has to get used to as a trader. You play the probabilities, and that includes
the choice of entry techniques. You must decide, using your experience,
what the probabilities tell you about a given setup, and then make your best
decision. After that you can only let it play out according to your plan.

You then incorporate that additional experience into the mix, and try to
assess the probabilities better the next time around. I have never found there
to be one clear-cut ' right' answer, at least not before the fact. I choose my
entry techniques, and all other aspects of my plan, as best I can, by assessing
the likely outcomes of the scenario at hand.

With all that said, I want to present one more example, this one where I
think the trendline violation is clearly superior to the moving average
crossover technique. The problem is, any time this technique is clearly
superior, you may not want to take the trade. Let me explain.

The times when you may want to avoid the moving averages, whether it be
moving average cross or moving average crossover, is when the market is
'choppy'. When the issue is trending, let' s say, but j umping all herky-j erky
above and below a moving average or regression line as it trends. It is still
trending, but not in a smooth, orderly fashion. In general, when I notice this,
fjust stay clear a/the issue.

Sometimes, though, I find a potential trade area that looks just great on the
traded timeframe, but it is choppy on the lower timeframe. The moving

72
average techniques j ust get whipsawed, so I opt to pay up and go for the
more confirmation technique of the trendline violation. Not necessarily
because I want more confirmation, but because this may be the least
restrictive technique that I can use for the issue, the way it' s behaving.

You have to look at the way the issue behaves and try and match up a
technique that will work. If you have more than one that looks good, then, as
I mentioned, I use the least restrictive technique, the one that gets me in the
soonest, paying the least 'premium ' . And when won't the trendline
technique be useful? Well, for example, how about if you can't draw a nice
trendline?

Many times an issue won't grant you the nice points necessary for a
trendline. Or what if you can draw a trendline (or multiple trendlines), but
the issue doesn't seem to respect them to any great degree? Well, then I
wouldn't use this technique. The nice thing about these techniques is that
you can see, before you reach a potential trade area, how the issue is acting
with respect to a broad choice of the different techniques.

To summarize, I find the trendline violation technique as the best choice

when the issue is choppy but a trendline can be drawn, and the trendl ine
looks to guide the trade. This happens here and there, because it may be the
only obvious guide the trade has, and hence many start to watch it. Gaps and
irregular trade are brutal on moving average calculations, but have very little
effect on a trendline.

When you have an issue that is gapping all over the place and chopping as it
trends into the potential trade area, look to the trendline technique as a
possible entry technique. Keep in m ind, though, I think it should be the rare
exception when an issue that behaves l ike this should be traded at all .

Let's look at the example, now, where I think the trendline violation
technique would clearly be the choice. FLEX trades in a very choppy manor,
with large gaps being pretty much the norm. The company is actually
headquartered in Singapore, but it trades on the NASDAQ. It trades
somewhat like an ADR in the way it gaps nearly daily, sometimes with quite
large gaps.

I noticed a pattern develop on the I S-minute chart, and the 60-minute and
daily charts gave me the context I needed to confirm the potential viability

73
of my potential trade area. The trade was actually a reversal trade and not a
trend continuation trade (which is my most common type of trade), and
hence I was looking for a technique with a pretty fair amount of
confirmation. I wanted to be a fair amount away from the potential trade
zone, in exchange for what I would expect to be a greater likelihood of the
trade working out. Let's first look FLEX on the traded timeframe, the 1 5-
minute, as it is approaching the potential trade area. See figure 4. 1 4.

Figure 4 . 1 4

?: FLEX 1 5-1 1!16U3

1 2 . 800

1 2 . 600

1 2 . 400

1 2 . 200

1 2 . 000

1 1 . 800

1 1 .600

1 1 . 400

1 1 . 200

1 1 . 000

1 0 . 800

8t <Jw lOt l lf Jull4m

Chart created by Dynamic Trader (c) 1 996-2001

FLEX approached the potential trade area and backed away, ending the day
in the middle of the day ' s range. Let's move forward to the next bar, which
will be the opening bar for the next day. See figure 4. 1 5 .

74
Figure 4 . 1 5

?. flEX 1 5-1 1!!113 E.t

1 2 , 800

1 2 , 600

1 2 . 400

1 2 , 200

1 2 , 000

1 1 , 800

1 1 .600

1 1 . 400

1 1 . 200

1 1 .000

1 0 , 800

8t 9w lOt 1 1f Jull4m
Chart created by Dynamic Trader (c) 1 996-2001

As is so common with FLEX, it gapped open, and right into the potential
trade area. Once this area is penetrated, I drop down to my lower time frame
and look for my trigger. I wil l show the 5/1 5 simple moving averages on the
charts so we can examine that trigger as we look at the trendline violation
trigger. First, let's look at the 5-min chart the day before the gap up into the
potential trade area. See figure 4. 1 6.

75
Figure 4 . 1 6

?: fLEX 5-1 '

!IS £J
1 2 , 600

1 2 , 500
""-
1 2 . 400

1 2 , 300

1 2 , 200

1 2 , 1 00

1 2 , 000

1 1 .900

1 1 . 800

1 1 . 700

Jull4m
Chart created by Dynamic Trader (c) 1 996-2001

FLEX appears to be behaving fairly orderly with respect to the 5-minute

moving averages. The biggest problem is with all the gaps, which can really
skew the moving average calculations. Let's move forward on the 5-minute
chart, j umping ahead to the end of the next day. See figure 4. 1 7.

76
Figure 4. 1 7

?: FLEX 5-1 !IS £i

1 2 . 700

1 2 . 600

1 2 . 500

1 2 . 400

1 2 . 300

1 2 . 200

1 2 . 100

1 2 . 000

1 1 . 900

1 1 .800

1 1 . 700

Jull4m 1St
Chart created by Dynamic Trader (c) 1 996-2001

Well, now we have some data we can discuss at length. The open on July 15
was the gap up move we saw, into the potential trade area. At was at that
point that I stated I would be willing to take my trigger and initiate a trade. If
you were inclined to use the moving average crossover technique, you
would have a trigger very soon after the open.

But notice how skewed the moving averages are by that gap open. You have
a crossover, but it is clear that at that point the longer period moving
average, the I S-period, still hasn't overcome that gap yet. That indicates that
the crossover is not a ' smooth data' cross, if you will. Take note, though, the
most interesting thing to me is the multiple crossovers and re-crossovers as
the day wears on.

Granted, you could be triggered into this trade by the moving average
crossover and be j ust sitting tight, waiting for your game plan to play out.
But I would not be confident that any of these crossovers mean anything.
Now you can guess that I picked a trade that is going to work out so I can
demonstrate the trendline violation entry, and that is a correct assumption.

77
Given that, it' s apparent that the moving average crossover entry is going to
work out okay.

That may all be true, but this is j ust one example, and one with a pre­
determined outcome. When I evaluate a trade after the fact, one of the things
I like to look at is how the issue behaved in the areas that I considered
technically significant. That' s how I j udge the action in the issue, that is,
how well it appears to react to areas where I expect reactions. Sure, the
moving average crossover may work here, but it appears like it would j ust be
luck, not technical factors. Let' s look at more data on the 5-minute chart, to
see where the trendline is in all this, and what it is I am looking at for my
trigger. See figure 4 . 1 8 .

Figure 4 . 1 8

?. FLEX 5-1 '

, !lEI £'J
1 2 . 700
1 2 . 600
1 2 . 500
1 2 . 400
1 2 . 300
1 2 . 200
1 2 . 1 00
1 2 . 000
1 1 . 900
1 1 . 800
1 1 . 700
1 1 .600
1 1 . 500
1 1 . 400

l lF �1� �
Chart created by Dynamic Trader (c) 1 996-2001

Now that adds the needed perspective. You can clearly see that FLEX has
mostly traded sideways for two full days, and is approaching the uptrendline.
This line is pretty obvious to most traders, and hence I would expect it is
being watched closely. Note, too, that this trendline has just the two points

78
needed for its construction. It hasn't been tested at all. If buyers are going to
come in, this is an area I would expect them to really buy the stock.

If the potential trade area is not to be a reversal point and FLEX is going to
set a new high for the move, this is where I expect the move up to start. This
is what I mean by choosing a trigger technique and area that has some
technical significance. Let's zoom in on this area as FLEX touches the
trendline for the first time. See figure 4. 1 9.

Figure 4. 1 9

?. FLEX 5-1 I!!EI £J

\
\
1 2 , 400

\
1 2 , 350

I\�, 1 2 , 300

1 2 , 250

1 2 , 200

Chart created by Dynamic Trader (c) 1 996-2001

Although this is well after the first moving average crossover following
penetrating the potential trade area, j ust look at those repeating crossovers ! I
would j ust have no faith in using crossover triggers with this issue, on this
timeframe. Let's j ump two bars ahead and assess the situation. See figure
4.20.

79
Figure 4.20

:::1. FLEX 5-1 I!IGI £I

)j
1 2 . 400

1 2 , 350

j
1 2 , 300

j
1 2 , 250

1 2 , 200

Chart created by Dynamic Trader (c) 1 996-2001

Wow, look at the price action. You have a big j ump up off of the trendline, a
quick reversal back down, and a close below the trendl ine. The trade would
now be triggered using the trendline violation technique. But is this real ly a
significant event that j ust happened, or is happening? When I look at the
sideways action on this chart without focusing on the trendline, I would
think that nothing has really happened that is of any significance. Let's move
one bar ahead. See figure 4.2 1 .

80
Figure 4.2 1

?. flEX 5-1 I!I3 EJ

1 2 . 400

1 2 . 350

1 2 . 300

1 2 . 250

1 2 . 200

Chart created by Dynamic Trader (c) 1 996-200 1

Not much has happened yet. We have an inside bar with a close lower than
the open, and below the trendline. A period of indecision, which, as you
know by now, is common as an important technical area is approached. I 've
found that most trades that work out 'play around' the technical breakout
area before launching. Let's keep advancing and see how this one goes. See
figure 4.22.

81
Figure 4.22

\
?, FLEX 5-1 I!II3J E'I
,

\
1 2 . 450

\
\ 1 2 . 400

j
1 2 . 350

\\ 1 2 . 300

1 2 . 250

1 2 . 200

11
Chart created by Dynamic Trader (c) 1 996-2001

Wow, again. This is one choppy trading stock. A huge up bar that traded
way above the trendline. (This bar i s also the opening bar for the next
trading day.) But notice that the bar i s a doj i bar that closed on the low end
of the bar, and couldn't hold its gains.

This wouldn't stop me out since my stop would be up above (likely just
above all the congestion on this chart, or perhaps above the potential trading
area, i f I was very confident in the pattern), but it sure would make my heart
j ump for a minute. Let's keep going and move ahead another bar. See figure
4.23 .

82
Figure 4.23

\
?. flEX 5-1 11!!18 £J

t \ 1 2 , 400

\�
1 2 , 350

1
I
1 2 , 300

I
1 2 , 250

1 2 , 200

Chart created by Dynamic Trader (c) 1 996-2001

Now that's more l ike it. We have a quick reversal back down, trading to the
low of the entire congestion area. Any move below this low should 'kick it
off . Let's move two bars ahead and see what happens. See figure 4.24.

83
Figure 4.24

?. FLEX 5-1 !!IS f!I

1 2 . 400

1 2 . 300

1 2 . 200

1 2 . 1 00

1 2 . 000

1 1 .900

1 1 . 800
lGw
Chart created by Dynamic Trader (c) 1 996-2001

There it is. An inside bar and then the sellers crush it. Notice how the
breakdown in the stock coincides with the breakdown through this trendl ine?
If buyers had decided to come in at the trendline, which is a l ikely place for
them to do j ust that, the moving average trigger would have been a stop out,
but the trendline violation would not have triggered. That's the trade off.
The moving average trigger would have you in earlier, but it wil l give you
more fal se starts.

When I see an issue that chops and gaps all around the moving averages but
I stil l see a trade I l ike, I lean towards an entry l ike this one. On this trade, I
liked the potential trade area because it was not only an Elliot wave pattern,
but also another trading pattern I use, right on top of the same area, with
outstanding Fibonacci convergences. Let' s take one look at how this played
out on the traded timeframe, the I S-minute chart. See figure 4.25 .

84
Figure 4.25

?'. FLEX 1 5-( I!EH!1

t===============:::�===
::t =====================� IH;� �Hff
1-- - 12.737 Exp 0.300

/,rt�f�Jv\r1fl<,�
-------------------- 1 2 . 800

1 2 . 600

1 2 . 400

�\uy
1 2 . 200

1 2 . 000

�W
1 1 . 800

1 1 .600

1 1 . 400

1 1 . 200

1 1 . 000

1 0 . 800

Jull4m 1St lGw 17t

Chart created by Dynamic Trader (c) 1 996-200 1

This trade went on to yield plenty of profit potential after the trigger. It' s
amazing to m e how a stock that trades as choppy as FLEX still frequently
follows technical patterns so well.

The trendline technique can be a very useful technique to have in one' s

arsenal, but experiment with i t and decide if and when i t may help your
trading. It' s a technique that I apply when conditions call for it, but that is
not the majority of the time. I consider it a more ' specialized' technique.

85
 Chapter 5

Regression Channels

My use of regression channels as an entry technique is really a variation, of

sorts, on the trendline violation technique. Although there are many possible
uses for regression channels that relate very little to trendlines per se, my
coverage in this book will be limited to this one application, which, as
stated, is a variation on the trendline violation technique. In practicality, I
see very little difference between the two techniques, and in my trading I
actually use a hybrid technique (which I will present later in this chapter)
quite frequently.

A regression channel is simply a straight-line channel around a linear

regression line. In review, a linear regression line is simply a mathematically
determined 'best fit' line for a series of data points, in this case prices. The
channel lines are a given number of standard deviations above and below
this linear regression line, with the default on most programs being two
standard deviations. All these calculations can be done 'behind the scenes'
with j ust about any modem charting software.

I use the basic, default setting of two standard deviations for j ust about all
my trading, only deviating from that setting if I find a very special set of
circumstances that I feel warrant it. How do I determine if it is warranted?
As I 've said many times, I experiment, and if I find something that works
better, I may use it. Generally, though, this is one where I rarely make a
change. To add to the educational value for the reader, I will choose the
example for this chapter to be one where I might play with the setting.

I find that, like the trendline violation technique, I like to use this technique
best when the issue is oscillating into the potential trade area in a fairly
orderly manner. The advantage, which you'll soon see, is that you can use
the regression channel even if you can't find a nice, clean way to draw a
trendline. It' s almost like creating a ' statistically determined' trendline, in
the event the points to draw a more 'normal' trendline j ust aren't lining up.

Let's go back to the AMZN example from the previous chapter and have
another look at the 60-minute trigger time frame chart, with the trendline
drawn on the chart. See figure 5 . 1 .

87
Figure 5 . 1

?, AMZN 60-1 !l8 £J

1 7 . 000

1 6 . 500

16.000

1 5 . 500

1 5 . 000

1 4 . 500

1 4 . 000

1 3 . 500

1 3 . 000

8 15 22 Apt'S 12
Chart created by Dynamic Trader (c) 1 996-2001

Let' s now look at the same chart, but with a regression channel drawn on the
chart. See figure 5 .2 .

88
Figure 5.2

?. AMZN GO-I 1!! 8 EI

1 7 . 000

1 6 . 500

1 6 . 000

1 5 . 500

1 5 . 000

1 4 . 500

1 4 . 000

1 3 . 500

1 3 . 000

8 15 22 AptS 12
Chart created by Dynamic Trader (c) 1 996-2001

Let me make a few comments on what we see here. By definition, for those
of you who can't recall their statistics right off the top of their heads (that' s a
joke, in case you haven't gotten my sense of humor yet), two standard
deviations will encompass 95% of the data points. As you can see from the
channel, that looks about right. There is one maj or spike outside of the
channel .

That spike is kind o f 'outlying data', where the gap and drop was just a bit
overblown. This will have the effect of creating a wider channel than the rest
of the data may indicate. It is common in statistics to drop the high and low
value from a data set before doing calculations. Although none of the trading
programs I have seen allow any adj ustments besides how many standard
deviation units the channels are, it would be interesting to calculate the
channel with that outlier ' smoothed out' a bit.

Although I didn't intend to go into too much detail on alternate standard

deviation settings, I did choose this example because it is one that may lend
itself to such an adj ustment. Why would I think that? I notice that when

89
AMZN peaks, it reverses j ust shy of the top channel line. When it falls and
reverses, it reverses quite a bit shy of the bottom channel line.

This tells me the channel perhaps isn't representing the price action as well
as it could. If I made a slight decrease to the channel width by decreasing the
amount of standard deviation used for the channel calculation, the channel
may do a better j ob for me of representing the price action. You might ask,
isn't this ' curve fitting' ?

I would say yes, but it is ' curve fitting' o n relatively current data (by current,
I mean with respect to this trade, as I would be watching it unfold in real
time and having what we have on the previous chart at our disposal), with
the obj ective to get the tools at hand to guide me as best they can. To me this
is no different than varying the period of a moving average to find the length
that best represents the current price action in the issue.

Before I do any 'tinkering' with the channel width, let's look at the channel
with the old trendline on the chart at the same time. See figure 5.3 .

Figure 5.3

1 6 . 500

1 6 . 000

1 5 . 500

1 5 . 000

1 4 . 500

1 4 . 000

1 3 . 500

1 3 . 000

8 15 22 Ap� 12
Chart created by Dynamic Trader (c) 1 996-200 1

90
You can clearly see j ust how close these really are. It is a matter, for all
intents and purposes, of splitting hairs at this point. The triggers likely will
be different, but it isn't going to be by much. And if the trigger bar happens
to be an expansion bar, they may trigger on the same bar. Let's skip ahead to
the trigger bar for the channel, which will be the first close above the
channel . See figure 5 .4 .

Figure 5.4

1 7 . 000 ·

1 6 . 500

1 6 .000

1 5 . 500

1 5 . 000

1 4 . 500

1 4 . 000

1 3 . 500

1 3 . 000

8 15 22 Ap� 12
Chart created by Dynamic Trader (c) 1 996-200 1

AMZN didn't trigger until two bars later in this case. The close of the
regression channel trigger bar is $ 1 4.07. Recalling back to the last chapter,
the close on the trigger bar for the trendline violation trigger was $ 1 4.05 . As
I suspected, the difference is nominal. This is a case where either technique
would work equally well.

The trendline was quite clear, so that technique was a good way to go. The
channel represented AMZN quite well, too, but the one outlying data group
perhaps widened the channel more than that which would represent AMZN
under more normal circumstances. Let's spend just a little time exploring a
better setting on the regression channel, to see how that affects the trigger.

91
Keep in mind: this isn't ' after poker' . We would be looking at the data in
figure 5 .3 as we waited for the trade to unfold, j ust the same as if it was in
real time. It is this data, and only this data, that is used for the calculation of
the channel, and for making the decision we have that it does or doesn't
represent the price action like we would want it to.

Let' s drop the standard deviation down to 1 . 8 instead of 2.0. The value
doesn't have to be in whole numbers, it can be any decimal value you
choose. Why choose 1 .8 as a first choice? I am j ust guessing at a ten percent
drop, as that sounds like a good starting place. I ' l l leave the old channel lines
on and add the new ones to the same chart. See figure 5 . 5 .

Figure 5.5

?: AMZN GO-I !IS J!I

1 7 , 000

1 6 , 500

16,000

1 5 , 500

1 5 , 000

1 4 , 500

1 4 , 000

1 3 , 500

1 3 , 000

8 15 22 Ap� 12
Chart created by Dynamic Trader (c) 1 996-2001

Well, that j ust about does it. Nice representation along the top (remember, it
is the top that I am going to trade the break of, so that's where my focus is),
and I bet it' s close to the old trendline. I ' ll add that old trendline in for
comparative purposes, and remove the 2.0 channel lines. See figure 5.6.

92
Figure 5.6

1 6 . 500

16.000

1 5 . 500

1 5 . 000

1 4 . 500

1 4 . 000

1 3 . 500

1 3 . 000

13 15 22 Ap� 12
Chart created by Dynamic Trader (c) 1 996-2001

Well, the difference is what I would call ' statistically insignificant' . As you
can see, the entry is going to be pretty much the same. I find it very
interesting how, more often than not, many of my entry techniques point to
j ust about the same spots. In my experience, when you get your trigger
narrowed down to this small of an area, it' s not worth the effort to try and
haggle over the small difference between possible entry points.

Keep in mind, too, that we have dialed down to a lower timeframe, and
hence a small difference on this timeframe becomes even smaller
(relatively), perhaps imperceptible, on the traded timeframe. Don't try to be
too perfect and go beyond the point of ' diminishing returns' for all the added
effort.

Let's move on to a l ittle 'trick' I frequently do, to blend all of the above into
a hybrid technique that can often be better than any of the individual
approaches. The hub of the regression channel is the regression l ine itself.
As I explained, this is nothing more than a 'best fit' line, mathematically
determined using the data points we have to work with. There will always be
a line that, mathematically, is the best that can be done with the data.

93
The regression channel then adds two parallel l ines a given number of
standard deviations away from the regression l ine, one l ine above and one
l ine below the regression line. The user i s free to set the standard deviation
to any amount he or she wants, including deci mal amounts (assuming the
program allows for this). The usual default setting is two standard
deviations, and by definition, thi s will encompass 95% of the data points.

For most work thi s 2 . 0 standard deviation setting is going to work j ust fine.
But instead of using the default, or even adjusting the default for the
behavior of the i ssue, let's look at a totally different approach. Let' s erase
the channels l ines altogether, and j ust start with the regression l ine itself.

Some programs have this feature as a stand-alone option, and some require
you to delete the channel l ines off once you add the whole channel on to the
chart. Additionally, some programs have a box you can uncheck to not show
the channel l ines, leaving j ust the regression l ine. Once you discover how
your program works, it should be quite easy to get j ust the regression line on
your chart.

94
Let' s look at the previous example, showing j ust the regression line. Note
that this is the exact same regression line as shown in the previous charts,
it' s j ust that the additional l ines have been deleted. See figure 5 .7 .

Figure 5.7

::':'. AMZN 60-1 !!13 Di

1 7 , 000

1 6 , 500

1 6 , 000

1 5 , 500

1 5 , 000

1 4 , 500

1 4 , 000

1 3 , 500

8 15 � �� 12
Chart created by Dynamic Trader (c) 1 996-2001

Notice how the data is oscillating above and below the line in a fairly regular
manner. As an aside, imagine the regression line as running left to right,
parallel with the ground (in other words, rotate the chart in your mind
counterclockwise a little bit), and imagine how the price woul d look then.
Looks a lot like an oscillator indicator (such as stochastics), doesn't it?

Now I am going to clone this regression line. Most programs have a feature
to allow you to clone any l ine. Even better, you can usually move this cloned
line all over the screen, dropping it wherever you like. You can usually pick
it back up and slide it all over, looking for something to j ump out at you. I ' ll
start by j ust dropping this cloned l ine well above the ' action' on the chart, to
show what I 've done. See figure 5 .8.

95
Figure 5.8

. 1 _ c x
1 7 . 000

1 6 . 500

1 6 . 000

1 5 . 500

1 5 . 000

1 4 . 500

1 4 . 000

1 3 . 500

1 3 . 000

8 15 22 AptS 12
Chart created by Dynamic Trader (c) 1 996-2001

I ' m not sure how your eyes wil l be seeing this, but my eyes are creating an
'optical illusion' that these l ines aren't parallel. I see the lines closer together
at the right than at the left. I can assure you, though, that they are paral lel.
What I am going to do now is slide that higher line down, all the while with
the program maintaining the parallel aspects of the line. I will then drop the
line when I see something that looks to be representing the price action.

F or the sake of this example, I will leave the first drop of the line (above the
price action, as shown in figure 5 .8), on the chart, and show that l ine and the
drop that I think represents the price action, on the next chart. Hopefully, it
wi ll be clear, then, that all I 've done is to slide this l ine down until I liked
what I saw, and then I fixed the l ine at that point. See figure 5 .9.

96
Figure 5.9

_ c x
1 7 , 000

1 6 , 500

1 6 , 000

1 5 . 500

1 5 . 000

1 4 . 500

1 4 . 000

1 3 . 500

1 3 . 000

8 15 22 AptS 12
Chart created by Dynamic Trader (c) 1 996-2001

Now that looks absolutely great to me, from a trading perspective. That l ine
is representing the price action with near perfection. And that l ine is
perfectly parallel with the regression l ine that takes into account all the price
action of this part of the downtrend. This would be my preferred method for
setting up the l ine I would use as my trigger. And one more time, j ust to
show how c lose these methods can sometime be, let' s add back in the old
trendline, for comparison. See figure 5 . 1 0.

97
Figure 5. 1 0

_ c x
1 7 . 000

1 6 . 500

1 6 . 000

1 5 . 500

�
1 5 . 000

1 4 . 500

1 4 . 000

1 3 . 500

1 3 . 000

8 15 22 AptS 12
Chart created by Dynamic Trader (c) 1 996-2001

Again you can see j ust how close these different lines are. But that may not
always be the case. Every so often you may see a chart that you want to
trade where the trendline j ust isn't that easy to draw. Perhaps the tops don't
line up real nice, or the trend angle changes during the move. I l ike the
freedom to move a parallel l ine around and make a j udgment call as to
where I want the trigger line to be, if I feel the situation warrants it.

The next chapter wil l change gears significantly, moving from techniques
that use l ines drawn on the chart that represent some aspect of the trend, to a
technique that uses violations of previous prices to signal a trigger.

98
 Chapter 6

Swing-High and Low Violations

One of my favorite techniques for entries is violation of swing-highs and

lows. This is a fairly commonly presented technique, but hopefully the twist
I wil l present, l ike most of my material, wil l be different. In my experience,
whenever I have seen entry via a swing-high or low violation, it has been on
the traded timeframe. I ' m not saying that no one has presented these
violation techniques any other timeframes, only that I haven't seen it in the
material I've covered.

As the reader should now be familiar with, one of my main tenets is to

decide on your 'potential trade area' on the traded timeframe, and then to
drop down to a lower timeframe to decide on, and implement, the entry
technique. I will continue in that vein in this chapter. As I j ust mentioned,
this is the main concept that I have not found when I read the available
works. I see a lot of mention about using these swing-high and low
violations, but always on the traded timeframe.

For me, that j ust doesn' t make sense, for the same reason it doesn't make
sense with other entry techniques. You are not taking advantage of a 'fine­
tuned' entry by going down to a lower timeframe and taking advantage of
the price action in the potential trade area. And on top of that, you will
generally be a lot farther away from a l ogical protective stop area if you
make the entry decisions based on the traded timeframe. Hence, I will
present this technique, again, in the context of a lower timeframe from the
traded timeframe.

Let me first explain what I mean by a swing-high or low, and a violation of

that swing-high or low. When an issue trends it will commonly pullback a
bit and then resume the trend. When it does this, the swing it j ust formed
wil l create two small turning points, termed swing points. If the swing point
is a high point, it is further term a swing-high, and conversely, if the point is
a low point, it is further termed a swing-low.

I will not get further into defining these terms here. Many references I have
read do a fine j ob at coming up with extensive sets of rules to define what is
and isn't a swing point. I've also seen a few sources that refer to a simple set

99
of rules, such as a swing-high simply has the bar to the left and right of it
with lower high points (reverse for swing-lows). From a practical standpoint,
for my own trading, I don't need to define a swing point with a set of rules; I
simply recognize one when I see it.

When I look at a pullback and attempt to decide if it' s a good swing to

consider using in an entry plan, I mostly l ook for proportion. I want the
swing point to be a certain distance from what I expect is the maj or turning
point on the traded timeframe. But I also want the size of the swing to be the
right amount. And I want the time factor i .e. how long before the swing
started and how long the swing has lasted, to be in the right balance.

I expect you are saying ' I ' m fine with all that, j ust tel l me what all those
proportions and guidelines are, and I ' ll be ready to go. ' Unfortunately, you
won't like my answer to that, at least not at first. My answer is, as I've said,
I don't have any rules for that. But the good news is that I can eye a ' good'
one in one second flat. And with practice, you should be able to make a
decision by eye, too.

To me, a good analogy would be deciding on swinging at a baseball pitch.

Sure, some are way out of the box and not even worth discussing. What
about the rest of the pitches, though, the ones that may or may not be good
ones? How do you explain and define what you will swing at? General ly, a
batter will tell you he j ust knows it when he sees it. And I can tel l just by
looking which swing-points I will consider for use as an entry technique.

Let's look at a few swing-points to start, and then I ' ll move on to trying to
show what I do and don't like about certain swing-points. Once we have the
idea of what a 'good' swing-point looks l ike, I ' l l present the entry technique
I use, using the swing-point violation. Take a look at UPS on a daily chart,
with what I consider the important swing-highs and lows labeled on the
chart. See figure 6. 1 .

1 00
Figure 6. 1

?. UPS D-O !IS £J

22 29 Jan 12 19 26 Feb 9 16 23 Mar 9 1& 23 30 Apr 12 20 27

Chart created by Dynamic Trader (c) 1 996-200 1

This labeling shows what I consider the pertinent swing-highs and lows from
a trading standpoint. Depending on the definition used, there could be
others. My goal here is to show not only what a swing-high or low looks
like, but to label only those that seem to fit the timeframe of the chart we are
looking at. To my eye, these points j ump right out.

Some of these swing-points may also have a greater significance on a larger

timeframe, pointing out larger timeframe turning points, but for now we are
only concerned with the swings as they relate to this timeframe. Note how
the swings are all very similar to each other, with none really standing out
from any of the others. UPS is trading with quite a distinct rhythm on this
chart, with all the swings being candidates I would consider for use in
potential trade entries. This is not the norm, though.

Most of the time there will be a lot more variance in the swings, with many
not being of a proportion and character that I would use for a potential trade
entry. The most important thing to remember is that I will apply this
technique just like the others I have presented thus far. I will find a potential

101
trade area on the timeframe I plan to trade, and then drop down to a lower
timeframe to look for an entry technique.

It is on this lower timeframe that I wil l look at and assess the swing-points
that I see on the chart. I wil l be looking for a swing-point right around the
potential trade area that has the proportion and character that I want. If other
swing-points in other areas don't have the desired character, I ' m not going to
be that concerned. As long as the swing-point that I will use for the actual
trigger has what I want, I will util ize it.

Let ' s look at a swing-point on the S&P e-mini 3 -minute chart. The 3 -minute
chart is my lower timeframe chart in this case, as I ' ve spotted a potential
trade area on the larger timeframe I 3-minute chart. I ' m looking at the 975
area as a potential trade area for a short trade. I ' m watching to see if a
tradable swing-point low forms after I penetrate the potential trade area.

Since we haven' t yet gotten to the actual technique, I ' l l present the trade
with the results on the chart, in one shot. I ' ve outlined the swing-low that I
would have used for the trigger. See figure 6.2.

Figure 6.2

::':. ES03U 3-1 I!I3 £I

swi n g
l ow 74.00

72.00

70 .00

68.00

+hl l��f\J
66.00

64.00

f
62.00

60.00

bW
Chart created by Dynamic Trader (c) 1 996-2001

1 02
I l ike this swing-low because it began to form shortly after the potential
trade area was penetrated. It pulled back up shortly after that, and began to
reverse before it reached the previous peak. This would be a so-called ' failed
test' of the immediate high of the move. As you can see from the chart, this
was the l ast swing type move before the mini dropped right off.

Keep this exampl e in mind, as I will refer back to it when I explain the
actual technique itself, and compare the technique to some of the ones we
have covered so far. At this point, let ' s look at another swing-point in the
mini that I consider a tradabl e swing-point. I was looking at a potential trade
area at 978 on the I 3-minute timeframe. I dialed down to the 3 -minute
time frame and was looking for a swing-high point that could possibly be
used for entry. I ' ve labeled the chart with the swing-high that formed right
after the mini traded into the potential trade area. See figure 6.3 .

Figure 6.3

}
:;:.-;. E S 03U 3-( I!EU!I
98 ,00

hJ 96,00

94,00

92,00

90,00

88,00

86,00

84,00

82,00

80,00

78,00

24t
Chart created by Dynamic Trader (c) 1 996-2001

Notice that the proportion and setup of this swing is very similar to the last
example. The main notable difference is in the time balance between the first
impulse of the uptrend that formed the swing-high, and the pullback to the

103
swing-low. In the l ast example it was fairly balanced time-wise, but in this
example the pullback is an approximate 25% time retracement. Generally, I
like the pullback to be between a 50% and 1 00% time retracement, but i f all
the other aspects of the pullback and the formation of the swing-point are
acceptable, I wouldn't l et a 25% time retracement preclude my taking the
trade.

Let's move on to an example of a swing-point that I wouldn't consider.

That's not to say that it couldn't lead to a great trade, as it very wel l might.
But I choose my criteria for a reason. One of the main things I look for when
I ' m choosing an acceptable swing-point is that it is stil l fairly close to the
potential trade area. This allows me to set a protective stop fairly close to my
entry point.

Although a trade using a swing-point that is far away from the potential
trade area could stil l use the area j ust below that swing-point as the
protective stop area, this is technically not as strong, since this stop comes
from the entry timeframe, not the traded timeframe. Hence, I prefer to use a
swing-point entry that is close to the potential trade area and a more
technically strong protective stop, usually j ust below the potential trade area.
Let's look at another example in the mini, again showing the 3 -minute
trigger chart. The potential trade area was around 975 . See figure 6.4.

104
Figure 6.4

?. ES03U 3-1 1!8 EJ

Jul.
Chart created by Dynamic Trader (c) 1996-200 1

Notice how far above the reversal point this pullback starts. And notice how
shallow the pullback is. Generally speaking, I am looking for the pullback to
start sooner and for the pullback (retracement) to be at least 50%. This
pullback is less than a .3 82.

And as I mentioned, you could still trade this swing-high, but logical stop
placement would be more difficult. As you can clearly see, this turned out to
be a smoking trade. But if I were looking to enter using the swing-point
technique, 1 ' d have been left out. Such is trading. If you stick to your trading
plan guidelines, you' l l sometimes miss great trades.

It' s my opinion that if you try to come up with a plan that makes sure you
never miss any great trades, you' l l probably not miss those great trades, but
the trade-off will l ikely be an enormously great deal of losing trades. If your
game plan works, don't fret over missed trades, it' s part of the business.

I can hear some people saying ' Hey, what about that smaller swing that
formed between the reversal point in the potential trade area and the swing­
high you highlighted?' In my opinion, that small swing-high would not meet

1 05
my criteria for a tradabl e swing-high. Not only did it not move far enough
off the reversal point (time-wise or point-wise), it didn't pull back enough
either. An acceptabl e swing-point would have formed at a point between
these two swing-points, and pulled back at least 5 0%. It also would have
shown a good time balance.

Let's move on to the actual technique itself. The trade will be triggered
when the issue trades above the swing-high in an uptrend or the issue trades
below the swing-low in a downtrend. This wil l be a lot easier to see if we
j ust move on to a chart. Let' s go back to the first example in this chapter, as
shown in figure 6.2. I wil l back up the chart a bit in time and show the trade
as it unfolds. See figure 6.5.

Figure 6.5

?: E S 03U 3-1 !l1iI D

j
swin g
low 74 .00

72.00

70.00

68.00

66.00

64.00

62.00

60.00

58.00

Gw
Chart created by Dynamic Trader (c) 1 996-2001

The mini has traded into the potential trade area and has formed an adequate
swing-low for entry purposes. At this point the next step is to draw in a
horizontal line from the swing-low, going to the right. Once this l ine is
breached, the trade will be triggered.

106
As with the other techniques presented so far, you wil l have the choice of
triggering the trade when the price simply trades through the trigger line, or
waiting to trigger until you get a close after crossing the trigger line. Unlike
the previous techniques where I generally wait for a close to trigger me into
a trade, I usually trigger on this technique with a simple price cross of the
trigger line. Let's look at the chart with the trigger line drawn on the chart.
See figure 6.6.

Figure 6.6

::;:. E S 03U 3-1 1!!1 13 E3

sWi n g
l ow 74.00

�j��!�#fMl
L 972.75 GAg14:3G
72.00

70.00

68.00

I��
66.00

l�
64 .00

\�
62.00

60.00

58.00

Gw
Chart created by Dynamic Trader (c) 1 996-2001

The trigger l ine is at 972.75. Any trade below 972.75 would trigger me into
the trade. Since the tick value on the mini is 0.25 point, the trigger price for
initiating a trade for me woul d be 972.50. Let's l ook at the chart, moved
ahead one more bar. See figure 6.7.

107
Figure 6.7

� ES03U 3-1 11!113 £1

sWi ng
low 74 .00
l 972.75 GAg 14:3G
72.00

70.00

68.00

66.00

64 .00

62.00

60.00

58.00

Gw
Chart created by Dynamic Trader (c) 1 996-2001

The mini traded right down to the trigger l ine, and although it i s difficult to
see on the chart because of the trigger l ine, the last bar closed on the low,
right at 972.75. One more tick down is going to trigger the trade. Let's move
ahead one more bar on the chart. See figure 6.8.

108
Figure 6�8

::::. ES 03U 3-1 1I!!6H3

swi ng
low 74.00
L 972.75 6Ag14:36
72.00

70.00

�
�\J
68.00

66.00

64.00

62.00

60.00

58.00

6w
Chart created by Dynamic Trader (c) 1 996-2001

The trade is now triggered at this point, whether you chose the simple price
cross or the price close below the trigger l ine. Let' s look, again, at how the
mini behaved after the trigger. See 6.9.

109
Figure 6.9

?. ES03U 3-1 1!18 F!J

swi n g
l ow 74 ,00
fr'L--'r--- L 972,75 GAg14:3G
72.00

70,00

68,00

66,00

64,00

62,00

60,00

58,00

Gw
Chart created by Dynamic Trader (c) 1 996-2001

I want to make a quick comparison here with a trendline trigger and a

moving average crossover trigger. I ' l l use my favorite for the moving
averages, the 5/1 5 simple moving average combination. I ' ll put both of these
triggers on the chart with the swing-low entry, for comparison. See figure
6. 1 0.

1 10
Figure 6. 1 0

?. ES03U 3-1 !II;] £I

sWi ng
low 74 .00

{-,-�.-"o:'- L 972.75 GAg14:3G

72.00

70.00

68.00

66.00

64.00

62.00

Chart created by Dynamic Trader (c) 1 996-200 1

Now that's interesting. We are seeing j ust about the same thing we've been
seeing all along. I do believe a pattern is emerging. All three entries are,
indeed, very close. Usually, the 51 1 5 simple moving average crossover will
trigger first, and then the trendline. One m ight expect that the swing-point
technique would trigger after both of these techniques because it is a higher
confirmation technique, and one that generally triggers later.

Sometimes this is the case, but sometimes it i s not. This i s due to the fact
that the there is some discretion to the choice of swing-points that are
considered for the trade. I prefer a swing that is close to the potential trade
zone, and hence will trigger fairly early in the move. This means the swing­
point entry may not come up and trigger that often, but can give you a price
that is better, the same, or not as good as the other techniques. It' s not
uncommon for me to have my chart set up l ike figure 6 . 1 0, so I can observe
multiple possible triggers at the same time.

So what are the different trigger prices for the three techniques on chart
6. 1 07 The first trade below the swing-low trigger l ine would be at 972.50,
and the first closing bar price below that trigger l ine is 972.00. The

111
following bar is the trigger bar for the moving average crossover, and that
bar closed at 97 1 .75. The bar that fol lows after that is the trigger bar for the
trendline violation technique, and that bar has a c losing price of 97 1 .25 .

Now, again, I ' m not stating that anyone could or would get filled at those
exact prices, but they allow for comparison. And as you can see, the prices
are al l within 1 .25 points of each other. I think it is getting clearer that the
variation between the techniques is minimal and that any of them can be
adequate for most circumstances. You can do some fine-tuning to choose the
best parameters and the best technique for an individual trade situation, but
don't get carried away with the time invested in making your choices, either.

Let's look at another example, the one from figure 6.3 . I will make this
example short by putting all three of the triggers we looked at in the last
example on the chart in one shot here. See figure 6. 1 1 .

Figure 6. 1 1

?: ES03U 3-1 I!!IGU3

sWin g
high

'r-r-- H 980.75 23JI12 :51

23w
Chart created by Dynamic Trader (c) 1 996-2001

1 12
At this point both the moving average crossover and trendline techniques
have triggered, but the swing-high violation has not. Let ' s move one more
bar ahead and see if the swing-high triggers. See figure 6. 1 2.

Figure 6. 1 2

?. ES03U 3-1 S8 £J

88.00

87.00

86.00

85.00

84.00
swin g
high 83.00

82.00

rr--t- H 980.75 23)1 12 :51 8 1 .00

80.00

79.00

78.00

23w
Chart created by Dynamic Trader (c) 1 996-200 1

One more bar was all it took to trigger the swing-high violation trade. On
this trade the order of triggering was more what one might expect, starting
with the moving average crossover, then the trendline violation, followed by
the swing-high violation.

Again, for comparative purposes, the moving average crossover trigger price
is 979.50, the trendline violation trigger price is 980.75, and the swing-high
violation would be at 98 1 .00. The price on the first bar to close above the
swing-high trigger l ine is 98 1 .25. All in all, these prices are very close, and
the price is greater as the greater confirmation technique is applied, as one
might expect.

1 13
Before moving on to the next chapter and technique, I want to point out one
interesting observation I see on a regular basis with respect to thi s technique.
Let's take another look at UPS, as shown in figure 6 . 1 . See figure 6. 1 3 .

Figure 6. 1 3

2 , 000

1 , 000

0,000

Jtf)tJ!
9 ,000

8 , 000

7 , 000

22 29 Jan 12 19 26 Feb 9 16 23 Mar 9 16 23 30 Apr 12 20 27

Chart created by Dynamic Trader (c) 1 996-2001

Here ' s the same chart as figure 6. 1 , without all the swing-point labels. I ' m
going to add a swing-low trigger l ine to the l ow on February 7 and expand
the chart out a bit. See figure 6 . 1 4.

1 14
Figure 6 . 1 4

1 . 000

L 59,010 7Feb01

12 19 26 Feb 9 16 23 Mar 9 16 23 30
Chart created by Dynamic Trader (c) 1 996-2001

The trade was triggered by violating the swing-low trigger l ine, and UPS
began to fal l off nicely. But then it began to rebound and went right back up
to the trigger l ine, and even traded above it. UPS reversed again in this area
and resumed the downtrend. I have found it to be very common for an issue
to trade back up to, and even slightly through, the trigger l ine. It doesn't
happen all the time, but it sure seems to happen often.

This can scare a trader out of a good trade if he or she stays on the trigger
timeframe or revises the stop loss very quickly. I j ust want to point this out
so you can be on the lookout for it. Let's look at one more exampl e on this
same chart. I ' ll leave the swing-low l ine on the chart, and add a swing-high
trigger that comes up j ust a l ittle bit later on. See figure 6. 1 5 .

1 15
Figure 6 . 1 5

2 . 000

1 . 000

0 . 000

9 . 000

8 . 000

7 . 000

6 . 000

5 . 000

4 . 000

3 . 000

2 . 000
1� 26 Feb � 16 23 Mar � 16 23 30 Apr 12 20 27 May 11 H
Chart created by Dynamic Trader (c) 1 996-2001

Notice how UPS traded through the swing-high trigger l ine and continued up
$3 .55 before pulling back. And when it pulled back, it came back almost all
the way to the trigger l ine. Again, it doesn' t happen every time, but it
happens often, so keep a watch out for it.

The next chapter is on the pullback technique, and that doesn't sound related
at all to the current technique. But, in fact, the way I use it, it is a variation
on the swing-point technique. The pullback technique is a way to anticipate
the trigger on the swing-point entry technique, in order to get a better price.
The downside? You' re getting less confirmation, so you' re more likely to
get a false signal.

1 16
 Chapter 7

Pullbacks

The entry technique I wil l present using pullbacks is based on the

groundwork laid in the l ast chapter on swing-high and low violations.
Although there is a near infinite number of ways that pullbacks can be used
for trading, I will confine this chapter to one basic variation. That is not to
say I don't use other pullback techniques, or that I don't consider other
techniques to be worthwhile.

In fact, quite the contrary is true. It' s j ust that volumes could be written on
pullbacks, and it woul d be outside the scope of this book to do more with the
concept than I plan to. I have chosen one of my most commonly used
pullback entry techniques for presentation. The technique I have chosen
dovetails the best with the material presented thus far, and it is also
incorporated into the 'cool trick' I wil l present in chapter 9.

As I implied at the end of chapter 6, the pullback technique I will present is

j ust a way to anticipate, perhaps 'jump the gun' , on the swing-high or low
violation technique. I am simply going to take a guess, albeit an educated
guess, as to when I think the pullback may end, and take the trade at that
point. This will have the effect of getting me in earlier than the swing-high
or low violation technique, but perhaps, too, getting me into a trade that may
not play out as anticipated.

I generally use this technique when I have the setup for the swing-high or
low violation in motion, and I get the feeling that the trade is really looking
good to play out as I expect. This can be a subtle thing, such as how the
buyers or sellers are acting, what the volume on each bar is looking l ike,
how other indices and issues are behaving, how this issue looks on an even
lower timeframe and a host of other things I watch. I can't lay this out for
you exactly; it' s a subtle, experience-based thing. But when I feel that the
issue is about to explode and I ' m running out of time, I frequently switch to
this technique on the fly.

The setup, as I ' m describing here, is the same as the setup from the last
chapter. The key is to recognize a pullback that has the characteristics that

1 17
you find acceptable. For me, as I mentioned, I want the pullback to start
fairly soon, pullback at l east 50% and have nice time symmetry.

The only real characteristic that I j ust described, as far as I ' m concerned, that
may be hard to decide on is how far the first thrusting move should be before
the pullback starts. I can see on the chart, by eye, where the area is that I
want the first move to go to. The question i s, what characterizes that area, so
that you can see it similar to the way I do? I ' m not sure I can really qualify it
for you, but I can give some hints to help guide you.

Ultimately, though, even if the hints are fol lowed, it may stil l lead to some
areas that I would look at and see right away that they are not what I would
utilize. This is j ust recognition, gained from experience, and I can' t give that
to you. Let' s go back and look at a few charts from the last chapter and
critique them, looking for clues to my thinking process. We' l l start out with
the mini 3-minute chart example from figure 6 . 5 . See figure 7 . 1 .

Figure 7.1

?: E S 03U 3-1 1!J3 £1

j
sWin g
l ow 74.00

72.00

70.00

+r\�lf�j
68 .00

66.00

�j�
64.00

VI j<
62 .00

60 .00

58.00

6w
Chart created by Dynamic Trader (c) 1 996-2001

The first thing I would note here is that the first thrust is a bit on the small
side. I would have preferred the first thrust to go a l ittle further. Why do I

1 18
think this? I ' m l ooking for the first move to go more than what could
amount to a small and simple pullback in the current trend. Pullbacks that
are 3-5 bars are commonly j umped on, in this case from the buy side. A
pullback that goes 5 bars or more starts leaning, in my opinion, towards
being a greater correction, and starts to tip the scales more in favor of the
potential turning point. I ' m also looking for the first move to be proportional
in size to other moves the issue has been making up to this point.

Generally, I ' m not looking for the first move to be the size of the larger
pullbacks to the trendline, but more l ike the first moves off of the trendline.
Keep in mind, I say trendline here j ust as a guideline. Sometimes the issue
doesn't fit a trendline well . In this case I j ust look at the oscillations of the
issue, and see the relative size of the various swings that it lays out. This
gives me an idea in my mind of what I may be looking for. I look most
closely at ' first thrusts' from the larger corrections. This is the approximate
amount I am looking for on a new first thrust in the possible new trend.

In figure 7. 1 I see what looks like a good candidate. See figure 7.2.

Figure 7.2

::''1. E S 03U 3-1 I!! [;l EI

j
sWi ng
low 74.00

72.00

70.00

lr\�l��J
68.00

j�
66.00

/
ji 62.00

Gw
Chart created by Dynamic Trader (c) 1 996-200 1

1 19
The first thrust up from where the arrow is on the chart looks to me to be an
average first thrust move for the i ssue here. The move is 3 .50 points. I also
see two more first thrust moves that I could focus on. Let me add arrows to
the chart for those two moves. See figure 7 . 3 .

Figure 7.3

?-. E S 03U 3-1 1!!113 £I

I
��
sWin g
l ow

Chart created by Dynamic Trader (c) 1996-2001

The first move on the left of the chart was 4.25 points, and the last move, on
the far right of the chart, was 2.50 points. The average of the three moves is
3 .42 points, or to the nearest tick, 3 .50 points. That's the exact value of the
first thrust choice I made from figure 7.2, as looking like an average first
thrust type move.

I picked it because it was in the middle of the values that I saw on the chart,
and thus I expected that it would be a good representation. I didn't do a lot
of calculations to come up with that, you can j ust look at the chart and figure
it out. Now where woul d 3 .50 points put the termination of the first thrust
down off the top? I ' l l add that to the chart, to see where I would be looking
for a first thrust to go. See figure 7.4.

120
Figure 7.4

?. E S03U 3-1 1!!8 £'I

l'

bW
Chart created by Dynamic Trader (c) 1 996-2001

The area for an ' average' first thrust would indeed be l ower. For me, the
area around 97 1 .75 would have been a lot better of an area for the swing-low
to form. Let me add back in the trendline, and we' l l see something quite
interesting. See figure 7 . 5 .

12 1
Figure 7.5

?. E 503U 3-1 1!113 F!I

74.00

72.00

70.00

68.00

66.00

64.00

62.00

60.00

6w
Chart created by Dynamic Trader (c) 1 996-2001

Note the proximity of the trendl ine to the l abeled area of 97 1 .75 . I do not
believe this is a coincidence. I find the first thrusts are frequently in or about
the area of the trendline. Also notice how the area I was looking at around
97 1 .75 is also at j ust about the same price level as the last swing-high before
the final top. I also believe that that is not a coincidence. I frequently
observe the first thrusts going to areas of previous swing-points like this.

Given all this information, why would I have still considered this a good
candidate for a swing-low violation entry? There are several reasons. The
most important thing to consider is that this is not a perfect world. You will
rarely satisfy all the criteria you are looking for. So at some point you make
a judgment call based on experience.

I felt the trade setup, i.e. the potential trade area from the higher time frame,
was a very good setup. I liked the general behavior of the mini on the lower
timeframe chart, as it approached the area. I also noted that I could have a
tighter protective stop loss with the smaller initial thrust, since 1 ' d be closer
to the potential trade area. Based on experience, I felt the swing-point was
acceptable.

122
Let ' s look again at the swing-high from figure 6. 1 1 . I observed that the first
thrusts were approximately the same at this time as in the l ast example,
about 3 .50 points. I added the 3 .50 points to the potential reversal point in
the trade area, and labeled the chart with a l ine at 98 1 .25 . I also included the
trendline that was used for the trendl ine violation technique. See figure 7.6.

Figure 7.6

::'':'. E S 03U 3-1 I![;J £.I

swi ng
high

23w
Chart created b y Dynamic Trader (c) 1 996-2001

It is clear to see, and again interesting to note, that the first thrust is very
close to the expected first thrust and is ' in or about' the area of the trendline.
I think this shoul d provide enough additional information for the reader to be
able to get a handle on the character of acceptabl e swing-points and
pullbacks. To complete this phase I will reassess the example of the swing­
point I didn't accept as a candidate in the last chapter, as shown in figure
6.4. Let's look at that chart again. See figure 7.7.

123
Figure 7.7

?-. E S03U 3-1 !IS F!'I

Jul.
Chart created by Dynamic Trader (c) 1 996-2001

This was the swing that I felt was j ust too far from the potential trade area,
and I accepted not taking the trade using this technique. I did mention that
the area that looked good was perhaps j ust above the very smal l swing and
pullback area (between the reversal point and the swing-high labeled on the
chart), and that that smaller swing-high was j ust too quick off the bottom
and didn't pull back enough for me.

Let's now put the trend l ine on the chart, and, since the first thrusts are still
around 3 .50 points at this time, I ' l l also add the l ine for 3 .50 points off the
bottom. This shows me the area that I would normally be looking at. Keep in
mind, I determined this 3 .50 points as my own approximation as to how I
felt the issue was trading at the times these charts were unfolding. I am in no
way implying that the issue trades l ike this anymore, or that the 3 .50 points
was anything special .

In fact, the 3 .50 points may not have been correct for anyone' s trading
except my own, even at the time these charts actually unfolded. Each trader
must determine for himself or herself what parameters may work for them. I
say all this because it is critically important that all readers understand that I

124
am using these examples to explain my thinking process, not to give details
of how I think any particular issue should be traded. The examples are used
only to show how I reason out my trading process.

Let's move on to the chart. See figure 7 . 8 .

Figure 7.8

_ C x;

94.00

92.00

90.00

88.00

86.00

84 .00

82.00

80.00
---r-:-I---- 978.50
78.00

76.00

25F
Chart created by Dynamic Trader (c) 1 996-2001

This is quite interesting. It turns out that the smaller swing was right at the
expected area for a first thrust (a coinci dence?), and the swing-high labeled
was, in fact, way too far from the potential trade area. But notice the
trendl ine. Even the way too far up swing-high didn't get to the trendline.
What does that say?

When you first look at the trendline, as it is considered as a potential entry

trigger, you must evaluate i f it makes sense to consider its use. In this case
j ust two points detennine the trendline, and the second of those points comes
after a significantly large countertrend move. This is going to have a large
part of the price action wel l away from the trendl ine. This can be seen by the
large air gap between the prices and the trendline between the contact points.

1 25
What this tells me is that I would be using a trendline that hasn't been tested
at all (no third or fourth contact point, etc.), and one that is going to be far
from the potential trade area. Although either of these conditions doesn't
necessarily preclude the use of the trendline as a potential trigger, both
together greatly decrease the utility of the trendline for me. Combine these
two conditions with the fact that the trendline is way above the area where I
would expect the first thrust move to go, and that would pretty much
preclude my using it.

This is all interesting, but this is a discussion right now on the swing-points
and pullbacks, so how does this relate to that topic? Well, I was saying that
generally the first tradable swing-point and pullback frequently happens in
or about the trendline area, and in this case that wasn't even close to correct.
This should be an alert to the trader. As I alluded to before, I will commonly
put three or four techniques on my chart and see how they look as the
potential trade develops.

This not only allows me to ' shift on the fly' with my choice of techniques (if
I want to), but it also allows me to look for ' anomalies ' . I ' m looking for
anything that doesn't fit, and in this case the trendline being so far from the
preferred area for the first pullback to occur would get me ' digging into' the
trade a bit more, looking to understand why that might be. Another thing I
frequently notice, and I consider this one of my best-kept ' secrets' (unti l
now, that is ! ) is that the first pullback will usually occur three to five bars
after the crossover of a 511 5 period simple moving average.

This is not to say that a pullback always occurs, or always occurs in this
area, just that when it does occur my experience is that it most often occurs
in this area three to five bars after the crossover. This also helps me in my
determination of the viability of a swing-point for use in the swing-point
violation technique, as well as the viability of the subsequent pullback for
the pullback entry technique. The lack of this criterion won't preclude the
move from my consideration, but it will be one of the factors I will use in
my consideration.

At this point I feel I have given the reader a detailed look into my thought
process for determining the relative attractiveness of a given pullback for use
with the swing-point and pullback techniques. Using this information, with
some experimentation and practice, each trader should be able to make an

126
informed decision about potential candidates that meet their own personal
criteria. All that is needed, now, is how I trigger a trade that has pulled back.

Like many of the techniques presented so far, there, of course, isn't one
clear-cut answer. I use several different triggers, depending on the individual
situation. Each trader will have to decide if he or she finds benefit from my
particular triggers, or if they want to formulate ones of their own.

I have three basic triggers that I use for most pullback entries. All are fairly
simple and straightforward. To start with, I use a simil ar concept to not
taking a trigger until the potential trade area is penetrated. In this case, the
pullback must be at least 50%, and once it has reached that amount, I will
then accept a trigger. I will first briefly present the three triggers, and then
go into details and examples on them.

The first trigger is simply a fade entry once the minimum 5 0% retracement
has occurred. The fade may be immediate or I may wait for a ' slowdown' or
'pause' in the action, and then enter. The second trigger is taking out the
high of the last bar of the pullback, once it has reached the minimum 5 0%
required retracement (reverse for short trades). The third trigger is the ' cool
trick' entry I will present in chapter 9. If this one sets up, it is my preferred
method.

1 27
Let's look at MSFT on a I S-minute chart. I found a very nice looking
potential trade area on the 60-minute chart, and I have dialed down to the
I S-minute to look for an entry. The potential trade area is around $25 .40. A
pullback is starting and I ' m waiting for the S O% retracement to be
penetrated. See figure 7 .9.

Figure 7.9

?. MSFT 1 5-1 1!!113 £i

26 . 300

�li
�l� )llJj
26 . 200

26 . 100

t�fj! �
tw
26 . 000

25 .900

25 . 800

l�t
- 25.575 R ..t 0 500
25 . 700

25 . 600

25. 500

IIIJII �� 25. 400

25. 300

25. 200

27F Jun30m it
Chart created by Dynamic Trader (c) 1 996-2001

As soon as MSFT trades at $27.S7 I would consider an entry trigger,

assuming, of course, that this happens fairly soon and in an orderly manner.
This goes without saying, but I ' m mentioning it to make it clear that this is a
here and now opportunity. The part I don't mention on each and every
example is that I ' m assuming it continues its pullback and penetrates the
area below the SO% retracement, and doesn't do anything that I don't like
the looks of. If it goes sideways, for example, for the next ten bars and then
goes below the S O%, I will have abandoned the trade long before that point.
Each trader needs to decide individually what is acceptable price action as
potential trigger time nears, but it must always be considered.

Let's advance one more bar and see what happens. See figure 7. 1 0.

1 28
Figure 7 . 1 0

?, M SFT 1 5-1 !IS £!

26 , 300

��
�l�� )l14
26 . 200

26 . 1 00

tj�rl
tW
1�
26.000

25 .900

25. 800

25.700

�
25 .575 Ret o .soo 25.600

25 . 500

��1�•• ��5 25 . 400

25 . 300

25 . 200

25t 27f Jun30m it

Chart created by Dynamic Trader (c) 1 996-200 1

The trade would now be triggered on a fade entry. Although there is no way
to know if an actual fill could have happened at $25 .57, let' s use that price
for the sake of the example. If I were in this trade my most l ikely stop
placement would be just under the potential trade area (which is around
$25 .40). I can now also be looking for trigger entry type number two, above
the top of this last bar. The last bar has a high of $25 .66. Let's add another
bar and see if MSFT is going to trigger the second entry type, or if the first
entry is going to take some ' heat'. See figure 7 . 1 1 .

129
Figure 7. 1 1

?. MSFT 1 5-1 !l13 tJ

26.300

�
�l\� }hl�
26. 200

26 . 1 00

tw
tj�jl � J�
26.000

25.900

25. 800

25. 700

25.575 Ret 0 '::0 0

25 .600

25 . 500

i.lm �� 25 . 400

25 . 300

25. 200

26t 27F Jun30m lt

Chart created by Dynamic Trader (c) 1 996-2001

The trade would now be triggered by the second method, trading over the
high of the previous bar. For the example, let' s use the price of one cent over
the high, so that would be $25 .67. For the additional educational value, let's
note that the swing-high point is at $25 .76. Using one cent over the high as
the example price for the swing-high violation trigger, that trade would
trigger at $25 .77. So you can see that the fade entry is $25 .57, the trade over
the bar high entry is at $25 .67, and the swing-high violation entry is at
$25 .77.

It' s just a coincidence that the increment of increase in each case is ten cents.
You can see how each entry costs you a bit more, but generally will give you
a little bit more confirmation. But, j ust as before, you can also see that, in the
context of the 60-minute traded timeframe, the differences are very smal l.
I ' ll add in quite a bit more data to this chart to show how this trade played
out, to put the differences in the entry into context. See figure 7 . 1 2 .

130
Figure 7. 1 2

?: MSFT 1 5-1 !lEI £I

27, 500

27 ,000

26 , 500

26,000

--11---- 25 .575 Ret 0 .500

25, 500
-----""---- 21,IU ��

26t 27f Jun30m H 2w 3t Jul7m 8t

Chart created by Dynamic Trader (c) 1 996-2001

I will refer back to this exampl e in chapter 9, when I cover the ' cool trick'
entry. I wonder if the ' cool trick' will wind up giving an entry very similar
to the current entries? I think you know the answer to that one.

Let's look at one more exampl e of a pullback entry and then move on to the
next technique. Wel l look at UPS, using the swing-high and pullback from
figure 6. 1 5 . I ' ll start with the daily chart, showing the pullback as it is
forming, with respect to the 50% retracement. See figure 7. 1 3 .

13 1
Figure 7 . 1 3

2 . 000

1 . 000

0 . 000

9.000

8 . 000

7 . 000

6.000

5 . 000

4 . 000

3 . 000

Feb 9 16 23 Mar 9 16 23 30 Apr 12

Chart created by Dynamic Trader (c) 1 996-2001

UPS hasn't quite pulled back my m inimum 50%, so I ' m not ready to accept
a trigger j ust yet. I ' l l move one bar ahead and reassess. See figure 7 . 1 4.

1 32
Figure 7. 1 4

Feb '3 if> 23 Mar '3 if> 23 30 Apr 12

Chart created by Dynamic Trader (c) 1 996-200 1

UPS has now penetrated the 50% retracement minimum and has thrown a
very strong reversal bar. This last bar opened at the low of the bar and closed
at the high of the bar. A fade entry would be triggered at the trade below the
50% retracement line. For this example, let's say that trigger price is $53 .48.
I would be looking, now, for a trade above this last bar, for a trigger of the
'trade above the high of the previous bar' entry. I 'll move one more bar
ahead and see if UPS triggers. See figure 7 . 1 5 .

133
Figure 7. 1 5

2 , 000

1 , 000

0,000

9,000

8 , 000

7 , 000

6,000

5 , 000

4,000

3 , 000

2 , 000
Feb '3 16 23 Mar '3 16 23 30 Apr 12
Chart created by Dynamic Trader (c) 1 996-2001

UPS blasted above that last bar, triggering not only the trade above the high
of the bar entry, but also the swing-high violation, on the same bar. For the
sake of this example, we' ll say the second entry trigger price is $53 .90, and
the swing-high violation entry trigger price is $54.98. The prices are all
fairly close, again, given the price of UPS. You definitely trigger at a better
price, though, by anticipating the swing-point violation with this technique. I
will refer back to this example, also, when I present the 'cool trick' in
chapter 9.

In the next chapter, I will revisit swing-point violations, but this time with a
new twist. By adding this new twist, it will add many potential entry
opportunities for trades. It will also address, at least partially, the problem of
what to do if you simply don't get any swing-points or pul lbacks after the
potential trade area is penetrated.

134
 Chapter 8

Additional Swing-Point Entries

In Chapter 6 the technique of swing-point violation entries was covered. All

of the examples in that chapter had one thing in common, though. All the
swing points occurred after the reversal point in the potential trade area.
What if there was no tradable swing-point after the reversal? Could a swing­
point that occurred before the reversal be used? And what about using a
swing-point that forms between a double bottom or a double top?

I have seen a fair amount of work discussing the relative merits of swing­
points before and after a reversal pivot point. Some authors say that a swing­
point before a larger scale pivot is more important and more useful. Then
others say a swing-point after the pivot is the one that carries the most
weight for trading purposes. What about a swing-point when the reversal
pivot point actually has two points to it, i.e. a double bottom or a double top?

In doing extensive experimentation with respect to my own trading, I

discovered that it didn't seem to matter all that much which of the three
swing-point variations I used for my trading. What did matter was that the
swing-point was formed with a certain proportion and had a certain
character. I covered that proportion and character in extensive detail in
chapters 6 and 7, and it applies j ust the same for swing-points that occur
before the reversal pivot and for swings that form as part of a double bottom
or double top.

Keep in mind, when I use the terms ' double top' and ' double bottom ' here, I
don't necessarily mean in the classical sense. In fact, if the swing-point is to
have the characteristics I have mentioned, it would be precluded from the
classical definition of ' double top' or ' double bottom' , simply because the
two points would be way too close to each other. The term will be used here
j ust for the convenience, and it shouldn't be confused with the much more
spread out classical version of the term.

One thing I have noticed is that I find a lot more trades with the swing point
after the pivot (reversal) point than before the pivot. Although there are
plenty of these swings to be found before the pivot and a fair amount of the
double top and double bottom types, I still see a lot more after the pivot as a

13 5
percentage of the total opportunities. In general, then, if I want to use the
swing-point techniques I focus on swings after the pivot, and keep an eye
out if something nice shapes up before the pivot, or ' around the pivot', in the
case of the ' double' formations. This wil l allow me to util ize these pre-pivot
swings i f they occur, and i f they don't, I simply look for another entry
technique.

Let's start out with an exampl e in AMZN. There is a potential trade area on
the 60-minute chart. I 've dialed down to the I S-minute to look for an entry.
AMZN has penetrated the upper end of the potential trade area and is
starting to pull back. See figure 8. 1 .

Figure 8 . 1

?. AMZN 1 5-1 I!EI E3

23 . 500

23 .000

22. 500

22 .000

2 1 . 500

Nov18m 19t
Chart created by Dynamic Trader (c) 1 996-2001

At this point I would be wil li ng to take a trigger because AMZN has

penetrated the potential trade area. As this chart lays out right now, though, I
have a few options, or at least potential options. If AMZN starts to go up
right from here, and if I l ike the swing-high that has j ust formed (the
pullback so far has penetrated a 5 0% retracement), I can use a trade above
that high as a trigger. I could also use the swing-high that formed j ust before

136
AMZN dropped into the potential trade area, again if I l iked that swing­
point. What I want to do is form possible scenarios, and whether or not I
would consider the triggers, so if they happen I ' l l know how to react. Let's
move ahead on the chart and reassess. See figure 8.2.

Figure 8.2

?, AMZN 1 5-1 I![;J D

23, 500

23 ,000

22 , 500

22,000

21 . 500

Nov18m 19t
Chart created by Dynamic Trader (c) 1 996-2001

AMZN didn't go up from that last point; instead it went down and set a
nominal new low for the move. This sets up a potential ' double bottom' and
defines a swing-high point that I would consider acceptable for a trigger.
This goes on the assumption that AMZN doesn't do anything I don 't like
before the trigger is hit, and that AMZN starts toward the trigger fairly soon.

It' s hard to describe what price action would invalidate the trade for me; I
just know it when I see it. The best thing I can say is that I ' d really prefer for
the price to j ust do what I expect, and that is to tum somewhere in here and
go up in an orderly manner. Anything that deviates too far from that and I ' l l
reconsider taking the trigger. I ' l l label the trigger on the chart with a
horizontal line. See figure 8.3 .

137
Figure 8.3

?: AMZN 1 5-1 1!!113 £I

23 . 500

23 .000

22 . 500

22 .000

H 2 1 .f>10 19Nv 13:45

2 1 . 500

Nov18m 19t
Chart created by Dynamic Trader (c) 1 996-2001

There' s nothing to do now but watch and wait, and see how AMZN reacts to
this area. The trigger is set, and unless some new price action comes in that
invalidates the current trigger, I ' m in watch mode. Let's move ahead a few
bars and see what happens. See figure 8 .4.

13 8
Figure 8.4

?. AtdZN 1 5-1 I![;l t1

23 . 500

23 .000

22. 500

22 . 000

21 . 500

NO\l18m 19t 2
Chart created by Dynamic Trader (c) 1 996-2001

That's it. As can be seen on the chart, the trade has triggered, and it has done
so on the first bar of a new day's trading action. Let's move ahead a bit and
see how AMZN is acting. See figure 8.5.

139
Figure 8.5

?: AMZN 1 5-1 1!!113 EI

23. 500

23 .000

22 . 500

22 .000

H 2 1 .G 10 l'3Nv 1 3 :45
21 . 500

Nov18m 1'3t 20w

Chart created by Dynamic Trader (c) 1 996-2001

There was a small pause in the area of the break (which we now know is not
uncommon), but then AMZN started up strong. Unless you thought that tiny
rest was enough of a pullback to try and use for entry (it certainly doesn't
meet the criteria I have presented thus far), you had no swing-point or
pullback opportunity at all once AMZN started up. Let's move farther ahead
and analyze how AMZN is behaving. See figure 8.6.

140
Figure 8.6

?: AMZN 1 5-1 1!!1 13 E.1

23 . 500

23.000

22 . 500

22 . 000

r.----H---- H 2 1 .6 10 19Nv 1 3 :45

21 . 500

Nov 18m 19t 20.....

Chart created by Dynamic Trader (c) 1996-2001

AMZN is just rocketing up. There was one additional, small pullback on the
way up, but it would not be the type I would use for entering a trade. I f the
highlighted swing-point was not used for entry, that was the last chance for
this technique, as I 've outlined it. This is one of the reasons I utilize this
technique when given an opportunity. I ' d hate to be left on the sidelines in a
potentially great trade, waiting for a post-pivot swing-point that never
comes. Let' s advance a bit more and make another assessment. See figure
8.7.

141
Figure 8.7

?: At.tZN 1 5-1 !Ie £J

23. 500

23 . 000

22. 500

22 .000

...----l-f- H 2 1 .610 19Nv 13:45

21 . 500

Nov1Bm 19t 20w 2

Chart created by Dynamic Trader (c) 1 996-2001

AMZN had a congestionlrest period and then j ust started up again, and on a
big gap (this might be considered a 'continuation' gap). Although there may
be a potential trade entry in this area, it is so far above the potential trade
area that it would be an entirely different trade at that point. Any trade entry
based on the potential trade area has either long since been executed, or long
since abandoned. My point for showing this additional data is to demonstrate
how sometimes, if an opportunity is presented and passed over, no more
opportunities may be presented.

Now some of you might be thinking that there was another potential swing­
point entry that I haven't mentioned, and you ' d be correct. What about the
swing-high point that stood out fairly well, before AMZN penetrated the
potential trade area. I ' l l highlight that potential swing-high trigger with a
horizontal line, and back up the chart on AMZN to the point where it was
j ust bouncing for the first time off the potential trade area. See figure 8.8.

142
Figure 8.8

?. AMZN 1 5-1 !J8 D

23 . 500

23 .000

22 . 500

22 . 000

2 1 . 500

NovlSm 19t
Chart created by Dynamic Trader (c) 1 996-2001

This is a great observation, noting this swing-high as a potential trigger. The

character of the swing, and its placement with respect to the potential trade
area, make it acceptabl e for me. At this point on the chart, an assessment has
to be made. The question is, i f AMZN goes straight up from here and takes
out that trigger, would that meet your own personal criteria for a trigger? For
me, the answer is yes. I definitely woul d have this l ine on my chart, and I
would be ready for that scenario to play out.

I didn't present this at first because I don't want to present multiple

possibilities when I first introduce a concept, it may get too confusing. Once
I feel the reader is l ikely to have the basic idea, I then feel it makes sense to
present multiple choices and opportunities at the same time. One has to walk
before one can run . Let's see how this potential trigger would play out.
We' l l move ahead on the chart and reassess. See figure 8 .9.

1 43
Figure 8.9

?: AMZN 1 5-1 !IS £J

23.500

23 .000

22. 500

22.000

H 2 1 .760 19Nvl0:30

2 1 . 500

Nov18m 19t
Chart created by Dynamic Trader (c) 1 996-2001

As soon as I see AMZN in this position, I would immediately abandon the

highlighted trigger in favor of the lower potential trigger. This puts us in the
position we were in, in F igure 8.2. This i s the point where I added in the
horizontal trigger l ine on this new swing-high point. I ' ll add it back in,
retaining the current horizontal trigger l ine for comparative purposes. See
figure 8 . 1 0.

1 44
Figure 8. 1 0

?. AMZN 1 5-1 !lEI ti

23. 500

23 .000

22.500

22 .000

21 . 500

NovlSm 19t
Chart created by Dynamic Trader (c) 1 996-200 1

Now, using the same assumptions as before, I would focus on the lower
price trigger, having this newer trigger supersede the higher price trigger. As
a potential trade evolves, I make constant assessments and changes in my
plan. When I see a potentially better trigger come up, I frequently ' upgrade'
to that trigger. It' s an ongoing, evolving process.

Let's move on to another example. I want to point out, though, that I chose
the last example with AMZN because it had less than ideal price action. I
will choose the next two examples trying hard for them not to be too 'well­
chosen' . I ' l l select them both to be a bit choppy, with the pullbacks not
being the most perfect, textbook pullbacks. The last exampl e in this chapter,
with CAT, will be slightly ill iquid in the chosen timeframe. I ' ve done this on
purpose, to reflect what a less than 'perfect' behaving trade may look like.
'
I've attempted to do this throughout my works, but I 've particularly focused
on this with the examples in this chapter.

There is a potential trade area on the 60-minute chart on YHOO. I 've

dropped down to the I S-minute chart to look for an entry trigger. I ' ll add a
lot of data onto the I S-minute chart to give you some perspective for the

145
potential trade area as YHOO approaches it, but generally I ' d spread the data
out a lot more and only focus on the price action on the trigger chart as it
comes into the area. See figure 8 . 1 1 .

Figure 8. 1 1

?. YHO O 1 5-r !lS E!

1 8 . 700
1 8 . 600
1 8 . 500
1 8 . 400
1 8 . 300
1 8 . 200
1 8 . 1 00
1 8 . 000
1 7 . 900
1 7 . 800
1 7 . 700
i�i.oi iilo Uii 1 7 . 600
l} :§�3 �@� U8�
1 7 . 500
1 7 . 400
1 7 . 300
7f Febl0m l lt 12w 13t
Chart created by Dynamic Trader (c) 1 996-2001

Let's move ahead one more bar and see what happens. See figure 8. 1 2 .

146
Figure 8 . 1 2

::::: YHO O 1 5-1 11!!1 8 F!'I

1 8 . 700
1 8 . 600
1 8 . 500
1 8 . 400
1 8 . 300
1 8 . 200
1 8 . 100
1 8 . 000
1 7 . 900
1 7 . 800
1 7 . 700
1 7 . 600
1 7 . 500
1 7 . 400
1 7 . 300
7f Febl0m 1 1t 12w 13t
Chart created by Dynamic Trader (c) 1 996-2001

YHOO has penetrated the potential trade area, and I would now be willing to
take an entry trigger. I wil l watch as the potential trade unfolds and make
assessments as time goes on, trying to decide on a trigger. Let' s add two
more bars and see if I can decide on a trigger. See figure 8. 1 3 .

147
Figure 8. 1 3

?. YHOO 1 5-1 1!!!IJ3 F!!

1 8 . 700
1 8 . 600
1 8 . 500
1 8 . 400
1 8 . 300
1 8 . 200
1 8 . 1 00
1 8 . 000
1 7 . 900
1 7 . 800
1 7 . 700
i' :ioi �� 4lii 1 7 . 600
if :§� Ii!�! UEl@
1 7 . 500
1 7 . 400
1 7 . 300
7F Febl0m l lt 12w 13t
Chart created by Dynamic Trader (c) 1 996-2001

YHOO has bounced off the potential trade area, and I ' m thinking that if it
j ust rockets straight up from here, I ' l l miss the trade. That's a downside to
triggers that key off of swing-points. If you don't get any swing points, you
don 't get any triggers. If YHOO just went straight up from here, would other
triggers, like moving average crossovers or trendline violations, work?

Yes, these triggers could be utilized in this case, since they don't require any
swing-points to be triggered. But perhaps you' ve decided you want a trigger
that is based on a swing-point violation for this trade. Then all you can do is
wait for the trigger to set up, and if it doesn't, you have to let the trade go by.
I ' l l add two more bars and see what YHOO is showing us. See figure 8 . 1 4.

148
Figure 8. 1 4

:::' YHO O 1 5-1 1!!1 6H!J

1 8 . 700
1 8 . 600
1 8 . 500
1 8 . 400
1 8 . 300
1 8 . 200
1 8 . 100
1 8 . 000
1 7 . 900
1 7 . 800
1 7 . 700
1 7 . 600
1 7 . 500
1 7 . 400
1 7 . 300
7F Febl0m 1 1t 12w 13t
Chart created by Dynamic Trader (c) 1 996-2001

Now that's a quick change. YHOO decides to set a nominal new low for the
move, and test the lower part of the groupings (my potential trade area). Not
only that, it sets up a possible swing-high point that I might use for a trigger.
I want to see how YHOO behaves as the potential trade unfolds. If YHOO
does what I want, then this swing-high point will make an excellent trigger
for me. And what do I want? It should be clear by now. I want YHOO to
reverse in here and start up, in a nice and orderly manner. This will be a
subj ective evaluation, and I do it based on my experience. Let' s go forward
another bar and see what happens. See figure 8. 1 5 .

149
Figure 8 . 1 5

?: YH OO 1 5-1 1!!!18 l3
1 8 . 700
1 8 . 600
1 8 . 500
1 8 . 400
1 8 . 300
1 8 . 200
1 8 . 1 00
1 8 . 000
1 7 . 900
1 7 . 800
1 7 . 700
!, :ioi �� ,J.t,i 1 7 . 600
l':§� �@f H8�
1 7 . 500
1 7 . 400
1 7 . 300
7f Febl0m 1 1t 12w 13t
Chart created by Dynamic Trader (c) 1 996-2001

YHOO is reversing off the second move into the potential trade area, and I
like the price action so far. I would now consider the swing-high point for a
trigger, assuming YHOO keeps heading up in an acceptable manner. I ' l l add
in my horizontal trigger line at this point. See figure 8 . 1 6.

1 50
Figure 8. 1 6

?. YHOO 1 5-1 !IS £!

1 8 . 700
1 8 . 600
1 8 . 500
1 8 . 400
1 8 . 300
1 8 . 200
1 8 . 1 00
1 8 . 000
1 7 . 900
H 17.8 10 13Fb1 3 :30 1 7 . 800
1 7 . 700
;;;;;;;;
;;;:;; �
;;;l itilan' H� 1 7 . 600
==== l�:§� ��! US�
1 7 . 500
1 7 . 400
1 7 . 300
7F Febl0m 1 1t 12w 13t
Chart created by Dynamic Trader (c) 1 996-2001

There is nothing to do at this point but watch as the potential trade unfolds,
and assess the situation as necessary. I ' ll add one more bar in and re­
evaluate. See figure 8. 1 7.

151
Figure 8. 1 7

?. YH O O 1 5-1 1!8 £J
1 8 . 700

��
1 8 . 600
1 8 . 500
1 8 . 400
1 8 . 300
1 8 . 200
1 8 . 1 00
1 8 . 000
1 7 . 900
H 17.810 13Fb 13 :30 1 7 . 800
1 7 . 700
l'ilci �� .� 1 7 . 600
ir :§� ��� U8�
1 7 . 500
1 7 . 400
1 7 . 300
7f Feb 10m 1 1t 12w 13t
Chart created by Dynamic Trader (c) 1 996-2001

YHOO is acting very well here, heading for the trigger. I would be very
confident at this point that I have done a good job in choosing my trigger. I
would expect that if YHOO is going to respect the potential trade area, it
will overcome this trigger fairly soon. Let's add another bar and see if
YHOO triggers. See figure 8. 1 8.

152
Figure 8 . 1 8

?. YH O O 1 5-1 1!8 F!J

1 8 . 700
1 8 . 600
1 8 . 500
1 8 . 400
1 8 . 300
1 8 . 200
1 8 . 1 00
1 8 . 000
1 7 . 900
H 17.BlO 13Fb13:30 1 7 . 800
1 7 . 700
i'iJOi �� .� 1 7 . 600
il :§� ��� � :3��
1 7 . 500
1 7 . 400
1 7 . 300
7F Feb10m 11t 12w 13t
Chart created by Dynamic Trader (c) 1 996-2001

YHOO paints an inside bar, but still hasn't triggered. If YHOO trades back
down into the potential trade area, I would likely want to reconsider my
trigger. I might be thinking that YHOO is congesting and trading sideways
in the potential trade area, and that is something I don't like to see. My
experience, in most cases, has been that if I have chosen my potential trade
area well, and the area is going to 'play out' and give me a good trade,
something should happen pretty quickly, and usually decisively. I'll add one
more bar and we' ll evaluate the situation. See figure 8 . 1 9.

1 53
Figure 8 . 1 9

?: Y H 0 0 1 5-( 1!If3 1:1

1 8 . 700
1 8 . 600
1 8 . 500
1 8 . 400
1 8 . 300
1 8 . 200
1 8 . 1 00
1 8 . 000
1 7 . 900
H 17.8 10 13Fb 13 :30 1 7 . 800
1 7 . 700
it;1Oi �� 'lei 1 7 . 600
if:m �@! U�ij
1 7 . 500
1 7 . 400
1 7 . 300
7f Feb10m l lt 12w 13t
Chart created by Dynamic Trader (c) 1 996-2001

Well, I didn't have to make any further assessments or make any other
decisions, since YHOO triggered on the next bar. From what I see on the
chart, I think the choice for a trigger was pretty good. Let's see how this
played out. See figure 8.20.

1 54
Figure 8.20

?" YH O O 1 5-1 I!!! EU!I

1 8 . 800

1 8 . 600

1 8 . 400

1 8 . 200

1 8 . 000

,.-d-- H 17.B 10 13Fb 13 :30 1 7 . 800

1 7 . 400

7F Feb 10m 1 1t 12w 13t 14f

Chart created by Dynamic Trader (c) 1 996-200 1

YHOO pretty much j ust blasted off thi s area. There was some 'congestion' ,
or a pause o f sorts that is barely visible on this chart, right after the trigger
point. There aren't any good swing-points for entry from my perspective,
though, once the trigger was hit. Let' s do a close up of the price action after
the trigger. I ' ll do a horizontal l ine trigger on that 'rest area' after the trigger,
so we can do an assessment on that spot as a potential trigger. See figure
8.2 1 .

1 55
Figure 8.2 1

� YHO O 1 5-1 1!!!113 £1

1 8 . 900
1 8 . 800
1 8 . 700
1 8 . 600
1 8 . 500
1 8 . 400
1 8 . 300
1 8 . 200
1 8 . 100
1 8 . 000
1 7 . 900
1 7 . 800
1 7 . 700
1 7 . 600

13t 141'
Chart created by Dynamic Trader (c) 1 996-2001

Although the second trigger l ine up at $ 1 8.02 coul d have been used as a
trigger, the swing-point sure didn't meet my criteria. If you were to look at
the next swing-point up from there, you ' d be way too far up for me to take
the trade as it relates to the potential trade area. This i s another example of
the highlighted swing-point as the last chance for using that technique.
Again, thi s is why I watch for these opportunities as they unfold. 1 ' 1 1 show
one more chart of YHOO, this time the traded timeframe, which was the 60-
minute chart, to show the larger perspective. I ' ll put the trigger l ine on the
chart so it is easy to see where the potential trade started. See figure 8.22.

1 56
Figure 8.22

?. YH OO 60-1

H 17.8 10 13Fb13:30

13t 'c
l ... Feb 19w 20t 2 1 F . Feb 25t 26w 27t
Chart created by Dynamlc Trader (c ) 1 996-200 1

1 57
I ' ll present one final example with CAT. There is a potential trade area on
the I 5-minute chart. This area also corresponds with a trendline, which is
quite prominent on the 60-minute chart. Let's start with the 5-minute, trigger
timeframe, as CAT heads towards the potential trade area. See figure 8 .23.

Figure 8.23

::'-1. CAT 5-1 !Ie EJ

rs3,400

[53,000

[52,400
�2,200
-52,000
rs1,800
8t
Chart created by Dynamic Trader (c) 1996-2001

Let' s move forward in time and assess the situation. See figure 8.24.

1 58
Figure 8.24

?: CAT 5-1 I!! EHl

�Vlr [53.400
[53.200
[53.000
'52.800

J1
'52.600

1 /"0'\
)Jtt lc
'52.400

.[111tJl1ft .1
'52.200
.t� ltil
� J..f -\.J .L '52.000
-51 .800
8t
Chart created by Dynamic Trader (c) 1996-2001

CAT didn't penetrate the potential trade area as expected, but instead
bounced up a bit. It has then started to roll back over and head towards the
area. In doing so, though, it has created a fairly nice swing-high. The swing­
high doesn't have that picture perfect look to it, it has more of a slightly
illiquid trading look to it, but it is potentially more than adequate for trigger
purposes. I ' l l add one more bar to the chart. See figure 8.25 .

1 59
Figure 8.25

?: CAT 5-1 I!!IS EI

f53,400
f53,200
[53,000
[52,800
[52,600
f52,400
f52,200
f52,000
f51,800
III
Chart created by Dynamic Trader (c) 1996-2001

CAT has now bounced off the top of the potential trade area and the
trendline. I would actually prefer that it trades through the top area j ust a bit,
but I would now accept a trigger. I would consider the swing-high I
mentioned as an acceptable trigger. I ' l l label the swing-high with a
horizontal trigger line. See figure 8.26.

160
Figure 8.26

?. CAT 5-1 !lEI £i

-53,400
-53,200
rs3,000
rs2,800
rs2,600
rs2,400
rs2,200
-52,000
-51 ,800
8t
Chart created by Dynamic Tr ader (c) 1996-2001

Now I just wait and see how CAT behaves and if the trade is triggered. I
qualify this, as usual, by mentioning that any price action that I don't like
will have me 'pulling' the potential trigger. Let ' s move ahead several bars
and reassess. See figure 8 .27.

161
Figure 8.27

?. CAT 5-1 1!J3 £'I

3 . 400

3 , 200

3 , 000

2 , 800

2 , 600

2 . 400

2 , 000

1 . 800

8t
Chart created by Dynamic Trader (c) 1 996-2001

CAT has done what I had hoped it would do, and that is to penetrate a little
bit further into the potential trade area. If it trades up from here and triggers,
I would take the trigger. If CAT continues down and trades below the
potential trade area I would cancel the trigger and no longer consider the
trade. Also, if CAT, for example, traded sideways in here for any sustained
period of time, I would also cancel the trigger. Let ' s move ahead a few bars
and see what happens. See figure 8 .28.

162
Figure 8.28

?. CAT 5-1 II!GJ £I

8t 9f
Chart created by Dynamic Trader (c) 1 996-2001

CAT has now triggered on a small gap up shortly after the open of a new
trading day. Let's add two more bars in and see how the trade is progressing.
See figure 8.29.

163
Figure 8.29

?: CAT 5-1 1!!Ir3 EI

-53,400
-53,200
rs3,000
rs2,800
rs2,600
[52,400
[52,200
[52,000
[51 .800
at 9F
Chart created by Dynamic Trader (c) 1996-2001

CAT is really going, as it takes out that swing-high trigger. Let' s add a bit
more data and see how this one played out. See figure 8.30.

1 64
Figure 8.30

?. CAT 5-1 !IE] £I

J
�
)
-,
\/'\\'\l\;�'-�I+---- f_
- - H 52 , 180 8My1 5:20

����==���=§���=���� U�i� iitf.'i

8t 9F
Chart created by Dynamic Trader (c) 1 996-2001

CAT really launched off this potential trade area, and didn't give much in
the way of swing-points/pullbacks on the way up. Although the quality of
the swing-point used to trigger this trade wasn't the greatest, it was as good
as I would expect, given the behavior of the issue up to this point. For my
trading, I felt the I 5-minute traded timeframe was adequately l iquid for
trading, but the trigger time frame was getting close to not having good
enough behavior to trade.

Given that the 5-minute timeframe is not the traded timeframe, I might be a
bit more lenient in assessing the character of the price behavior. Still , I need
a minimum quality to the swing-points and pullbacks in order for me to
consider their use as potential triggers. What all that boils down to is that
although this swing-point trigger wasn't the best I have seen, it did tum out
to be the best, and only, swing-point trigger that met my criteria for this
trade.

I ' ll finish this example, and chapter, up by zooming in on the potential trade
area and trigger area and looking at two other potential entry triggers at the
same time as this swing-point entry trigger, for comparison. First, I ' l l add on

1 65
the 1 0-period exponential and 20-period simple moving averages. I ' l l look at
the moving average crossover trigger. See figure 8.3 1 .

Figure 8.31

2 . 500

J
2 . 400

l�
��
ll
1l J l '
2 . 300

2 . 200
H 52.180 8My15:20

�
j { )j !
2 . 1 00

j 2 . 000

1 .900

1----- 5 1 .e 1 9 Rlt Ule

9F
Chart created by Dynamic Trader (c) 1 996-2001

Notice how the trade triggered one bar after the swing-high trigger, and at a
slightly lower price. All in all, though, the trade triggers at a very similar
spot. Let ' s throw in a trendline, for use in the trendline violation technique,

�
and see where that triggers. See figure 8.32.

1 66
Figure 8.32

_ c x

'"k:-----+-- H 52, 180 8My15:20

1----- 5 1 .e 19 Rlt UIg

9f
Chart created by Dynamic Trader (c) 1 996-2001

The trendline trigger also produces a very similar entry. If you waited until
the first close above the trendline, your trigger price would be lower than the
swing-point entry, at least in theory. It' s hard to estimate what an actual fill
price might be, and, again, discussion of the prices in this regard is only for
comparative purposes. The lesson is in how close these trigger prices all are
to each other.

I hope it was clear in this chapter j ust how similar the application of these
techniques is to the presentation made in chapter 6. For all intents and
purposes, I trade all three variations of the swing-points pretty much the
same. My main concern is the quality and character of the swing-point and
its pullback, and how it fits in the context of the current price behavior. I ' m
not near as concerned with where the swing-point occurs.

In the next chapter, I ' ll introduce one of my favorite variations on an entry

technique that I presented in an earlier chapter.

1 67
 Chapter 9

A ' Cool Trick'

The ' cool trick' is a favorite variation of mine that I apply on pul lback
entries. I am only presenting it here in the context of the pullbacks, as
described in chapter 7 . I apply it to other pullback situations, but for now I
am presenting it only as I apply it to pullback situations l ike those shown in
chapter 7. As always, I encourage experimentation with the techniques, and
their use and appl ication in any situation that works for the individual trader.

One of my main methodologies, one I ' m sure the reader is by now quite
familiar with, is the use of three timeframes. The ' middle' timeframe is the
traded timeframe, the longer timeframe is used for context, and the shorter
timeframe is used to fine-tune an entry into the trade. What I will be
presenting here amounts to dropping down yet one more timeframe, to fine­
tune the entry even more. This is not a 'trick' whose use I take l ightly,
however.

What I mean by that is, I have to be careful not to indiscriminately start

going down to a lower timeframe than my entry trigger timeframe. To do so
may introduce a timeframe that is way too small with respect to the traded
timeframe to have any real meaning. The ' sensitivity' is simply far too great.
What I am doing with this technique is different, though. I am looking at the
entry chart like it is a trading problem in its own right, and looking to solve
that entry problem.

The entry problem on the entry chart, though, is j ust the same as the original
entry problem. Confused? You shouldn't be. If we have a traded timeframe
chart and we drop down to a lower timeframe for entry once we have
penetrated the potential trade area, and we find that the entry chart has a
potential trade area (the 50% retracement), why not apply the very same
technique again? That's the premise here, and for my trading it' s a valid
premIse.

I am only applying this technique to pullbacks that are part of an acceptable

swing-point formation, and only once the pullback has penetrated the
minimum 50% retracement. I generally don't apply it in other
circumstances. I have exceptions, but only because I ' ve experimented

1 69
extensively with the technique and found other ways to apply it. As with any
technique, whether it is one if my techniques or one from any other source,
each trader should experiment and decide if it will work for him or her, and
under what circumstances.

Let's look at that MSFT example from chapter 7. The following chart is
from figure 7.9. Recall that MSFT has penetrated the potential trade area
around $25 .40 and reversed. MSFT then began to form a swing-high point
and pullback that met my criteria. The chart is the lower (trigger) timeframe.
The 50% minimum retracement I look for is labeled. See figure 9. 1 .

Figure 9 . 1

?. M SFT 1 5-1 1!! 13 £1

26 . 300

111
�l�� )hiJj
26 , 200

26 , 100

tw
tj�r! ltr1V
26 ,000

25 , 900

25, 800

l�t
- 25 ,575 Ret 0 .500
25 , 700

25 , 600

25, 500

n,m _fl.5 25. 400

25 , 300

25 , 200

2Gt 27F Jun30m 1t

Chart created by Dynamic Trader (c) 1 996-2001

MSFT hasn't yet hit the 50% minimum retracement, so let's advance
forward one more bar. See figure 9.2.

170
Figure 9.2

::'':'. M SFT 1 5-1 1!!!1 13 £J

26 . 300

�
26 . 200

f\�
26 . 1 00

�
tW
tl�fJ
ill4
26 .000

1�
25 . 900

25 . 800

25 . 700

�
25575 ,,, 0 .500
25 . 600

25. 500

nlm �f(l,5 25 . 400

25 . 300

25 . 200

26t 27F Jun30m 11

Chart created by Dynamic Trader (c) 1996-2001

MSFT has now penetrated the 50% minimum retracement area, and I can
now consider a trigger. Here' s where I see the similarity with the initial trade
process. What if I presented this chart and said I have a potential trade area
at the 50% retracement. Now I want to dial down to a lower timeframe and
look for an entry. Likely, you would see that as a valid plan.

The only difference I can see is that the 50% retracement is j ust one number,
not a grouping of numbers. Normally, I trade areas I think have a
preponderance of reasons for significance, at least as far as my experience
tells me. (They don't necessarily have to be Fibonacci groupings, just solid
potential trading areas.) That ' s why I am confident to consider it as a
potential trade area.

Remember the 'Plan for a Trade' . It includes an area to trade, an area you
have decided you think gives you an edge, an area you think something is
likely to happen. This trade, redone as a standalone, simply doesn't have
that. That's why I say I don't want to j ust apply the technique at random, so
to speak. I have found the technique works for me because it is a refinement
of a valid trade premise I have laid out.

171
As long as I apply it in this context I think it is useful to me. What I am
trying to convey is, don't j ust apply it whenever you want to any situation
you want. It then becomes out of context, and as I have pointed out so many
times, without context, I find very little value in any technique, regardless of
at what stage of the trading plan it is applied.

Let' s now drop down to an even lower timeframe, this time the 5-m inute
chart, as MSPT penetrates the 5 0% retracement area. See figure 9.3.

Figure 9.3

1j 25 .900

25. 800

JJ qrIth j
25. 700

ill! Ill , , <

25 .600

25. 500

25 . 400
1t
Chart created by Dynamic Trader (c) 1996-2001

MSPT is in the area where I would consider a trigger. Down on this 5-

minute timeframe, I have more detail to examine. I want to look at the
pul lback, and decide when I think that pullback may be ending. I have a few
preferred techniques I usually choose from to help me decide. I can see that
this pullback is very orderly and it appears that it would lend itself very wel l
to a trendline violation entry. Sound familiar? I am j ust using techniques and
concepts I already have to address this trading situation. Let ' s draw in a
trendline and see what that shows us. See figure 9.4.

1 72
Figure 9.4

I1MII :...,:11 _1�lx

II 25 .900

25 . 800

JI } I t ! ! p h 25 . 700

�j ftt h tt1
} jt1
f� j
25 .600
25.575 Ret o.soo

25 . 500

25 .400
11
Chart created by Dynamic Trader (c) 1996-200 1

The trendline represents the price action very well. Understand, when you
use a trendline for such a small trend i.e. so few price bars, it frequently will
not have multiple contact points. That doesn't matter to me, since I ' m only
using the trendline to indicate to me that the price direction has changed
right at that instant. I ' m not trying to determine how solid of a trend I feel is
in place, or anything else. Let' s move two bars ahead and assess the
situation. See figure 9 . 5 .

1 73
Figure 9.5

�;��':I"'liI
I _ Clx

11 25 , 900

25, 800

t1 f ) ! hj
tI 1 1
j) l l t Ilf t l
25, 700

n
) I �J
h
J
2S .575 Ret o .soo
25 ,600

25 , 500

25 , 400
1t
Chart created by Dynamic Trader (c) 1 996-2001

The trade would be triggered at this point, with a close above the trendline.
Let's note the closing trigger bar price for comparative purposes, which is
$25.64. If you'll recall, the entry over the high of the 1 5-minute bar was at
$25 .67, as shown in figure 7 . 1 1 . This entry was a slight bit lower, but in the
same general area.

When I see such a nice, orderly pullback, I also look at a moving average
cross entry, as covered in chapter 2 . I use a fairly short period moving
average, usually a 5-period or an 8-period, when I do this 'trick ' . I will try a
few variations and see which one looks like it best represents the price action
that I have available once the 50% retracement area has been penetrated. In
this case I went with an 8-period simple moving average. Let's look at the
chart with this average added on. See figure 9.6.

1 74
Figure 9.6

?. MSFT 5-1 !IS £I

25 . 900

25. 800

tI J I
I I I I)
25 . 700

P
f1 j
25 .600
25.575 Ret 0500

25. 500

25. 400
it
Chart created by Dynamic Trader (c) 1 996-2001

This looks very similar to what we j ust did with the trendline. Let' s move
two bars ahead, again, and do another assessment. See figure 9.7.

1 75
Figure 9.7

?. MSFT 5-1 BEU!!

4 25. 900

25. 800

lI I I 25. 700

It\IJ n 25. 600

25 .575 Ret 0.500

25. 500

25. 400
1t
Chart created by Dynamic Trader (c) 1 996-2001

The trade is again triggered at this same point, and we' l l note the same price
of $25 .64 as the closing bar trigger price for this variation of the technique.
To show just how similar the two variations are, on the next chart I ' ll show
both the trendline and the 8-period simple moving average together. See
figure 9.8.

1 76
Figure 9.8

25 .900

25 . 800

25 . 700

II
25.600

25. 500

25. 400
1t
Chart created by Dynamic Trader (c) 1 996-200 1

Now, to me, that ' s amazing. The two j ust about converge right at the
breakout point. It j ust goes to show how these techniques can frequently
point you to very specific areas if you choose the parameters to represent the
price action well . Let's move ahead one more bar, and look at the trigger bar
with both the triggers on the same chart. See figure 9.9.

1 77
Figure 9.9

_ c x

25. 900

25. 800

25 . 700

25. 600

25. 500

25. 400
it
Chart created by Dynamic Trader (c) 1 996-2001

Either technique variation would have been a good choice within the
guidelines of my trading plan. As we have seen, when the technique is
chosen well and applied correctly, the entry areas are, many times, very
similar to each other.

A third variation I really like is using the bars to trigger me into the trade.
This is real ly a fine-tuned variation on taking out the bar high (low for
shorts), j ust applied to the lower timeframe. I prefer to use some j udgment in
this regard, and not just automatically take the first signal from the first trade
over any bar high once the 500/0 retracement area i s penetrated. I look at the
other two entry variations we j ust covered on the same chart, and I look at
the last few bars. Let's look at the chart for now without anything else on it,
as the decision point is approaching. See figure 9. 1 0.

1 78
Figure 9 . 1 0

?: MSFT 5-1 1!!113 Ell

)1 25 . 900

Ilh j
25 . 600

J} w
25 . 700

25 . 600

25 . 500

25 . 400
1t
Chart created by Dynamic Trader (c) 1 996-2001

How would I assess the bars here? Normally, it' s common to use the high of
an inside bar for a long trigger. It' s even somewhat common to use the high
of yet another successive inside bar as a trigger. If I were in the 'regular'
lower timeframe, I would l ikely use those criteria for my trigger.

In this case, though, since I ' m in an even lower timeframe, which I consider
more ' sensitive' that usual (with respect to my traded timeframe), I want to
be a little less quick to trigger. So, in this case, I would use the high of the
bar that set the low, which is j ust a bit higher than the inside bar that
followed it. Let's see what that looks l ike by drawing in a horizontal trigger
l ine on the chart. See figure 9. 1 1 .

1 79
Figure 9. 1 1

?: M SFT 5-1 !IS E'I

)t 25 .900

25. 800

J ) l h j
}I !j
25 . 700

)1
II I t h j }
n !ff
) Id
I
H 25 600 111 ,,,,, 25 .600
25 .515 . .. 0 500

25. 500

25 . 400
1t
Chart created by Dynamic Trader (c) 1 996-2001

It is interesting to note that the 1 5-minute chart bar trigger is actually the
high of the bar j ust before the trigger bar on this chart. You can see, then,
that this choice of triggers would be at a lower price. That is the reason
behind doing all this extra work, to attempt to get in at a better price with a
smaller protective stop loss, and to not lose all that advantage by greatly
increasing the chance of a ' false start' . Let's move two bars ahead and assess
the situation. See figure 9. 1 2 .

1 80
Figure 9. 1 2

?: M SfT 5·1 !lEI £!

11 25 .900

25 . 800

) l h j
dt
25. 700

II t f J }
}t 1t t ) l t t l
nl
I IIJ
H 25.GOO 1)112 :35 25. 600
tl t 25.575 Ret O .sOO

25 . 500

25. 400
1t

Chart created by Dynamic Trader (c) 1 996-200 1

Well, this is getting repetitive, isn't it? MSFT didn't trigger on the next bar,
as that bar traded to the same high as the trigger bar, but not above it. It did
trigger on the fol lowing bar after that, though. But since the trigger is the
first trade above that bar high at $25 .60, we' l l record the trigger price at
$25 .6 1 . That's not to say that anyone could have gotten filled at that exact
price, of course, but we need some prices for general comparison. This price
is better than the previous two variations at $25 .64.

Let's finish this exampl e by adding one more bar to the chart, and putting all
three of the trigger variations on the chart at the same time. See figure 9. 1 3 .

181
Figure 9. 1 3

_ c x

k 25 .900

25. 800

}
hJh1 It
25. 700

II �
H 25 .&00 lJ1 1 2 :35 25 .600
25.575 Ret 0 500

25 . 500

25 . 400
it
Chart created by Dynamic Trader (c) 1 996-2001

In a case such as this one, I would certainly note that the three techniques all
converge on pretty much the exact same point, right at $25 .60 on the second
from the last bar on this chart. Given this, I would seriously consider a
trigger on any trade above this point. This is a great example of why I
actually like to use multiple techniques at the same time to guide my
decision making process. In a situation like this I would be very unlikely to
wait for the close on that last bar to initiate the trade. This is a subj ective call
based on experience, and all I can do is pass on the reasoning behind my
potential decisions.

I ' d like to present one more example on this 'trick ' . Let's go back to the
UPS example, which was triggered on the daily timeframe. 1 ' d like to dial
down to a lower stil l timeframe of 60-minutes and go through the 'trick' on
this one. Let's go back to the daily trigger timeframe chart of UPS and look
again at the swing-point and pullback, as the 50% minimum retracement is
penetrated. This chart was previously shown as figure 7. 1 4. See figure 9. 1 4.

1 82
Figure 9 . 1 4

Feb '3 1& 23 M.3r '3 1& 23 30 Apr 12

Chart created by Dynamic Trader (c) 1 996-200 1

The 'regular' way to look for an entry would be to utilize this chart and one
of the previously described techniques. Instead, we will attempt to use the
' cool trick' , and go down to an even lower timeframe; in this case I ' l l choose
the 60-minute. Let ' s look at the same pullback on UPS shown in figure 9. 1 4,
this time on the lower timeframe. See figure 9. 1 5 .

1 83
Figure 9. 1 5

?: U PS 60-1 I!!IB F!I

l,)�\j
6 . 500

6 . 000

1 5 . 500

�t
5 . 000

l )�tVu)¥\tln,n
1
4 . 500

4 . 000

L}l J

r� �J
53.495 Ret 0.500 3 . 500

3 . 000

2 . 500

2 . 000
Apr2m 3t 40", St 6f Apr'9m lOt l lw l2t Apr16m l7t

Chart created by Dynamic Trader (c) 1 996-2001

At this point UPS has penetrated the 50% retracement area and I have to
make a decision on what trigger I might consider. I like to put all three
triggers that we have covered for this 'trick' on the chart at the same time
and make a more 'holistic' decision, but it' s up to the reader to decide i f he
or she wants to try them all at the same time, or to pick a particular one and
just use that. First, I ' ll add a simple moving average to the chart. See figure
9. 1 6.

1 84
Figure 9 . 1 6

?. U PS 60-1 !1GB'!!

-------++---:-- 53.435 Ret 0 .500

Apr2m 3t 4w St of Ap�m lOt l lw l2t Aprlom l7t

Chart created by Dynamic Trader (c) 1 996-2001

In this case I opted to use a 1 0-period simple moving average. I tried an 8-

period, but felt the 1 0-period did j ust a slightly better job at representing the
price action. In reality, the difference is extremely minor and l ikely of no
trading significance at all . UPS is not all that smooth or orderly as it trends
into the trigger area, and that wil l usually have me using a slightly longer
period moving average. I suggest you experiment with varying period
lengths and see what works best for you and your particular trading plan. 1 ' 1 1
now add in a trendl ine and see what that shows me. See figure 9. 1 7.

1 85
Figure 9 . 1 7

_ Oi X

6, 500

6,000

\
5 , 500

5 , 000

4 , 500

4 , 000

------++_:_ 53,435 Ret 0500 3, 500

3 , 000

Apr2m 3t 4o.v 5t 6F Apr9m lOt l lw l 2t Apr16m l7t

Chart created by Dynamic Trader (c) 1 996-2001

The first thing I notice now that the trendline is on the chart is that the
trendline is tracking a bit above the moving average. This might give me a
later signal if I use the trendline. If I am feeling a great deal of confidence in
this setup, I might opt for the moving average cross trigger, and if I ' m not
quite as confident I may opt for the trendline violation.

This level of confidence is not something I can quantify for the reader, it is a
matter of my 'gut feel ing', based on having seen a vast quantity of simi lar
setups over a long period of time. All in all, I don't feel the difference is
very great between these triggers, and all I ' m trying to do is increase my
perceived edge by j ust an additional fraction, if I can.

I ' ll now add a bar high trigger to the chart. See if you can guess which bar I
am going to choose as my trigger bar. Look at the bars as UPS heads into the
trigger area and see if anything j umps out as potentially more significant that
anything else you might use. I see the bar I want to use right off. I ' l l add a
horizontal trigger l ine to the high of that bar. See figure 9. 1 8.

1 86
Figure 9 . 1 8

6. 500

6.000

5 . 500

5 . 000

4 . 500

4 . 000

3 . 500

3 . 000

2 . 500

2 . 000
Apr2m 3t 4w 5t 6F Apr.lm lOt l lw 12t Apr16m 17t

Chart created by Dynamic Trader (c) 1 996-2001

When I choose a high of a bar trigger and highlight it with my horizontal

trigger line, I am doing that in the context of the chart, as it exists when I do
the label ing. If the chart changes before the trigger is hit in such a way that I
no longer like my choice, I will change it. The bar trigger is based on the
assumption that the trigger will be hit with no unusual price action, and with
a tum towards the trigger happening fairly soon. I re-evaluate the choice
with every additional bar that's added to the chart. Let ' s move two bars
ahead and assess the situation. See figure 9. 1 9.

1 87
Figure 9. 1 9

H 53 .750 17 Ap 10 :30
-------tl-ccff- 53.435 Ret 0 .500

Apr2m 3t 4<", 5t 6f Apr'3m lOt l lw 12t Apr16m 17t

Chart created by Dynamic Trader (c) 1 996-2001

At this point UPS hasn' t triggered any of the variations. The price has
penetrated the moving average, but has still closed below it. If you were
opting to trigger on any price action above the moving average, you would
have been triggered at this point. Let's add one more bar and see if any
triggers have been set off. See figure 9.20.

1 88
Figure 9.20

6 , 500

6,000

5 , 500

5,000

4 , 500

4 , 000
H 53 ,750 17Ap 10 :30
------ll+---:-I-F- 53 ,435 Ret 0 .500 3 , 500

3 , 000

2 , 500

2 , 000
Apr 3t 4w 5t &F Apt'3m lOt l lw 12t Aprl&m 17t
Chart created by Dynamic Trader (c) 1 996-2001

Wow, that's pretty interesting. One bar has triggered all three variations at
the same time. The close of this trigger bar is $53.87, and I ' l l use that price
for comparison for the moving average cross trigger and for the trendl ine
violation trigger. . The trade above the high of the bar trigger I ' l l record as
one cent over the bar's high, so that one would be $53 . 76. Recall back to the
example using the daily bar high trigger, that the trigger price using that
method was $53 .90.

In a similar way as we have seen in the other examples, we see that the some
potential price improvement may be gained using the 'trick' . It' s up to each
individual trader to decide if this potential improvement is worth the
additional effort, or even if the technique will work at all for his or her
trading plan. Myself, I have found real benefits from this technique.

Let's look at UPS right after this trigger, to see i f it appears as though this
area was important, or i f many traders seem to have been watching this area.
See figure 9.2 1 .

1 89
Figure 9.2 1

?: U PS 60-1 1!! 8 F3

�---tfl'"-:- H 53,750 17Ap 10 :30

-------t+-:-f'F+- 53,435 Ret 0.500

Apr 3t 4w 5t GF Apr'3m lOt l lw 12t Apr16m 17t 18\

Chart created b'»' Dynamic Trader (c) 1996-2001

It does appear that this area was of some significance. Can I guarantee that
this 'explosion' in price action had anything to do with what I have outlined
on the chart? Absolutely not. It may have had nothing to do with the
technical features I 've outlined. All I can say is that in my experience, I've
come to conclude that I see these areas of activity most often in areas that
seem to have technical aspects that lead me to believe something might
happen in that area. It just seems like too great of a coincidence to me.

Perhaps it' s a self-fulfilling prophecy, in that if enough traders and fund

managers all see the same thing and then dog-pile in at roughly the same
time, the move will occur for that very reason. As a trader I ' m not that
concerned why a move happens. I ' m concerned with finding areas where I
think a move might occur, and then trying to find a place to get into a trade
where I feel the reward to risk ratio gives me the potential to profit over the
long run. That's what I ' m trying to accomplish with my methods.

Let' s take one last look at UPS with a l ittle bit more data added after the
trigger. See figure 9.22.

190
Figure 9.22

?: UPS 60-1 1!16U3

8 . 000

7 . 000

�-tfl'�-- H 53.750 17Apl0:30

-----H--:1Ft4--.::::....,.::--- 53.4$5 Ret 0 500

Apr 3t 4w 5t f,f Apr lOt l1w 12t Apr 17t 18w 19t ;

Chart created by Dynamic Trader (c) 1 996-200 1

It turns out that the trigger area was quite significant. UPS did a strong
pullback almost to the trigger area after it' s first thrust up. Perhaps a few
readers are thinking that a pullback like that would shake them out. Keep in
mind, this is not only a lower trigger timeframe, it' s a 'twice-lower'
timeframe, a 'trigger for the trigger' timeframe. Remember I said that as
soon as I ' m triggered I switch back to my traded time frame?

I do this so I don't ' micromanage' the trade too closely on the trigger
timeframe. Well, if you manage your trade on the trigger' s trigger
timeframe, you ' ll l ikely get stopped out in short order almost every time.
Remember the importance of context. If the trade doesn't play out on the
traded timeframe, then it doesn't play out. I don't try to make a winner out
of a loser by ' scalping' a little bit out on a much lower timeframe.

Each trader has to decide, though, if such micromanaging helps his or her
trading. For me, it' s a losing battle plan. I stay with my game plan, and if I
want to make a change, I do. But I always do it when I ' m not in a trade, in
the 'heat of battle' . I make the changes, which would include changes in
trade management, when it' s off market time and I ' m able to think without

191
any bias from an open trade. I suggest that you very strongly consider this
concept. If the reader would like additional information on trade
management with respect to exiting trades, please see Kane Trading on:
Trailing Stops.

192
 Conclusion

I have presented in this book what I consider to be 'the best of the best' of
my entry techniques. They have served me well, and I hope the reader has
gained valuable knowledge from my presentation of them. I have honed
them over many years of experimentation. I suggest that the reader
experiment with them also, to determine if they can be of use to you, and if
they can be improved on.

Never forget, though, the context of the entry technique within the overall
trading plan. It must never be forgotten that it is j ust one small part of the
whole, one part in a larger scale picture. It is in this context that I have
presented the techniques, and it is in this context that I hope the reader
studies, tests and experiments with the techniques. Without this context, I
feel they are next to useless.

Hopefully, though, my presentation was clear enough and my constant

harping about context well tolerated. If so, then I believe the reader has
made a worthwhile investment with his or her time, and hence has benefited
from this book as much as I had hoped. If this is so, I will feel that I have
succeeded at my goal.

1 93