by
2010
Russell Enterprises, Inc.
Milford, CT USA
Elements of Positional Evaluation
How the Pieces Get Their Power
by
Dan Heisman
ISBN: 978-1-888690-58-3
© Copyright 2010
Dan Heisman
All Rights Reserved
No part of this book may be used, reproduced, stored in a retrieval system or transmitted in any manner or form whatsoever or by any means, electronic, electrostatic, magnetic tape, photocopying, recording or otherwise, without the express written permission from the publisher except in the case of brief quotations embodied in critical articles or reviews.
Published by:
Russell Enterprises, Inc.
P.O. Box 5460
Milford, CT 06460 USA
http://www.russell-enterprises.com
info@russell-enterprises.com
Cover design by Janel Lowrance
Printed in the United States of America
Dedication
Preface
Introduction
Chapter 1: The Background of Positional Knowledge
1.1 Definitions
1.2 A Brief History of Positional Theory
1.3 Positional Theory in 1974
1.4 Inadequacies of 1974 Theory
Chapter 2: The Elements
2.1 Introduction
2.2 Mobility
2.3 Flexibility
2.4 Vulnerability
2.5 Center Control
2.6 Piece Coordination
2.7 Time
2.8 Speed
2.9 Summary
Chapter 3: Pseudo-Elements
3.1 Introduction
3.2 Material
3.3 Space
3.4 King Safety
3.5 Development
3.6 Summary
Chapter 4: The Pieces in Relation to the Elements
4.1 Introduction
4.2 Mobility
4.3 The Knight
4.4 The Bishop
4.5 The Rook
4.6 The Queen
4.7 The King
4.8 The Pawn
4.9 Piece Summary
Chapter 5: Static Features and the Elements
5.1 Introduction
5.2 Static Features and the Pawns
5.2.1 Doubled Pawns
5.2.2 Isolated Pawns
5.2.3 Backward Pawns
5.2.4 Passed Pawns
5.2.5 Pawn Summary
5.3 Files and Diagonals
5.3.1 Introduction
5.3.2 Open Files
5.3.3 Semi-Open Files
5.3.4 Ranks
5.3.5 Diagonals
5.4 Weak Squares and Outposts
5.5 Summary
Chapter 6: Miscellaneous Applications of the Theory
6.1 Introduction
6.2 The Elements During the Phases of the Game
6.2.1 The Opening
6.2.2 The Middlegame
6.2.3 The Endgame
6.3 A Brief Guide to Chess Principles
Chapter 7: Epilogue
Appendix: Illustrative Games
Bibliography
Footnotes
Acknowledgements
About the Author
List of Other eBooks
To “Coach” Donald Byrne
“A chessplayer’s chessplayer and a friend’s friend.”
– part of a telegram from then World Chess Champion Robert J. (Bobby) Fischer to International Master Donald Byrne at Byrne’s testimonial dinner in Boalsburg, Pennsylvania, September 7, 1974, shortly before Byrne passed away at an early age as a result of a rare disease that cut short not only his life, but a very promising chess career. We’ll never forget you, Coach!
Elements of Positional Evaluation: How Chess Pieces Get Their Power provides a different perspective on how to evaluate the effectiveness of chess pieces and positions.
Originally written in 1974 on the author’s typewriter, this new, greatly expanded fourth edition of Elements includes material comparing this current edition of Elements with 1974 positional chess theory. It also includes new discussions on how 2009 theory has moved ever closer to the ideas suggested in Elements.
In addition, over 100 new examples and diagrams have been added plus an Appendix of Illustrative Games, making Elements less of a work on theory and more about theory and practice.
I hope my work in greatly enhancing Elements has also enhanced its instructive value and enjoyment for readers.
Dan Heisman
January 2010
The fourth edition of Elements of Positional Evaluation presents a new challenge to the author: how to position a book whose first edition was written in 1974 about the inadequacies of positional theory at that time, and make it relevant to readers in 2009?
The challenge is more daunting because the suggested improvements to 1974 theory have, to a great extent, become reality, so 2009 theory is much more aligned with the ideas in Elements.
One possibility would have been to ignore the changes in theory that have occurred since 1974 and simply leave the book the way it was: compare the proposed theory to “current” theory as if 2009 theory was very similar to that of 1974. This would be the easiest approach, and superficially reasonable; the first three editions of the book sold out – why change a good thing?
However, I agree with my publisher that taking the easy way out is not only incorrect, but overlooks an opportunity to do more for the reader. Therefore, this edition will take a different approach by adding a theme: occasional discussion/examples of how theory has changed between 1974 and 2009 and how that change brings 2009 theory closer to my proposed theory.
The other substantive revision in Elements will be the addition of many new examples, which will make this expanded edition less of a purely theoretical work than the first three editions, and more practical.
After analytical skill, the next most important chess skill is evaluation. For positions, this means answering the questions: Which side stands better? How much better? and Why?
Breaking positional evaluation down into its elemental parts requires an answer to the question How much is that piece worth in this position?
Most chessplayers rely on loosely knit, unstructured methods to evaluate positions and the role of the pieces therein. They learn positional principles (i.e., guidelines/heuristics/rules of thumb) that often lead to inaccurate evaluations.
A similar problem occurs when chess programmers quantify evaluations in the form of a “scoring function.” In the early days of computer chess, programs relying heavily on evaluation played worse than those using primarily “brute force look-ahead.” Hopefully, this book will provide a step towards overcoming some of these difficulties.
This introduction lays the foundation for a new evaluation theory and provides an overview of the book’s organization. We will attempt to show why a new theory is necessary, revolutionary, and novel. Not all of the points made in this book are original; some of the novelty will be highlighted by the systematic way the material is organized and presented.
Hopefully, the new theory will give a better understanding of the nature of the game and should help many readers significantly improve their playing strength. Bertrand Russell wrote:
But when theories change, the alteration usually has only a small effect so far as observable phenomena are concerned. The practical difference between Einstein’s theory of gravitation and Newton’s is very minute, even though the theoretical difference is very great. Moreover, in every new theory there are some parts that seem pretty certain, while others remain very speculative.
Chess knowledge is not the same as chess ability. Moreover, positional knowledge is separate from analytical and tactical ability, which are the predominating factors in a player’s overall chess strength. There is no substitute for analytical and tactical competence. Because Elements is about positional theory, absorbing its material will not directly enhance most players’ analytical ability, but should greatly benefit their positional awareness.
We will attempt to show that the evaluation theory of 1974 needed to be enhanced. We will not only present the new theory, but also prove the deficiencies of the 1974 theory and sprinkle discussion of how 2009 theory more closely follows the new. Many examples will be provided to help illustrate these points.
Before we start, it would be helpful to discuss revolution – scientific revolution. Thomas S. Kuhn’s book The Structure of Scientific Revolutions explained the process of how scientific theories evolve. The following is a short synopsis of Kuhn’s Structure:
In any scientific discipline that deals with unknown, difficult-to-perceive, or inexact phenomena, there exists a certain status quo that is accepted by most authorities. This status quo is a theory or set of theories that attempts to explain observed phenomena and tries to predict future occurrences. Examples include: the structure of the universe, molecular theory, origins of the earth, origins of life, and how physical phenomena occur. In general physics, the well known theories that came to be widely accepted were formed by Aristotle, then Newton, and finally Einstein. There were other rejected theories, and the eventually accepted theories were not accepted immediately by the originator’s peers. Some theories may take centuries before they are generally accepted by specialists in the field.
The general pattern is as follows:
First, there exists an accepted theory that is believed by most of the specialists in a field. This theory has a certain “track record” for explaining and predicting phenomena. As this theory begins to fail, various reasons are given by its adherents to explain these exceptions. This pattern of theory breakdown continues until several things happen. First, the exceptions pile up until they get unwieldy. Then, various new revolutionaries analyze these unsatisfactory results and use them to construct assorted new theories (called paradigms). Each paradigm attempts to better explain existing phenomena and more accurately predict future results. Slowly, the old theory collapses. The revolution continues as the experts in this field devise experiments and tests to help them determine which new theory, if any, is acceptable. Finally, sometimes dramatically, the experts gravitate toward one theory, which becomes the new status quo. Most scientific theory is formed this way.
Therefore, “scientifically,” the tasks for this book are clear: first, present 1974 theory on positional evaluation; second, show how it is inadequate; third, devise a new theory and, finally, prove that the new theory is superior to the old. To accomplish this final task, the author must prove that the new theory explains existing phenomena (observations) with fewer, if any, exceptions.
These tasks parallel this book’s structure. There are seven chapters, in the same order as the steps outlined above. The Introduction and Chapter 1, “The Background of Positional Knowledge,” explain the 1974 theory, attempt to show its inadequacy, and forges onward toward 2009 theory. Chapters 2 and 3, “The Real Elements” and “Pseudo Elements,” show the basis of the new theory. Chapters 4 and 5, “The Pieces and the Elements” and “Static Features and the Elements,” solidify the new theory, show how it relates to 1974 theory, and compare the adequacy of the new and old theories. Chapter 5 also serves with Chapter 6, “Miscellaneous Applications of the Theory,” to take some of the previous stratagems of existing “dogma” and looks at them in a new light. Finally, Chapter 7, “Epilogue,” attempts to solidify the case by summarizing the process, and also presents a final argument in favor of the new theory.
The following are highly credible sources who believed that 1974 theory had too many exceptions.
In his book on Petrosian,1 P. H. Clarke responds to a surprise move with this perceptive annotation:
This magnificent reply must have come as a rude shock to White. After all, it seems to flout a basic law of positional play. But these laws are not so much to be obeyed as interpreted; and it is the interpretations given by the great masters that ensure the continuing evolution of chess.
In A. D. de Groot’s remarkable scientific work, Thought and Choice in Chess, the Dutch psychologist reinforces the premises of this book many times when he shows how chess talent is developed.2
In the chess master’s empirical, specifically inductive way of thinking there are no primary principles from which deductions can be made; nor are there any empirical rules without exceptions. Often a plan or board goal must be given up right after the opponent moves: if shifting to another plan is more ‘advantageous.’ A dogmatist is just as unfit for playing chess as he is for leading a dynamic enterprise. The chess master is of necessity a relativist or even, so to speak, an opportunist in his thinking.
Later de Groot talks about a player’s individual perception of these rules:
What actually happens is best illustrated by looking at playing methods. First, by means of playing experiences and/or textbooks the player gets to know certain important general strategic and tactical rules; next, he learns to recognize and to handle exceptions to these rules – which in their turn grow into new, more refined rules with new exceptions, etc. Finally, the player develops a “feeling” for the cases in which these already highly specialized rules can be applied... A player develops a feeling for those cases where there are winning chances and for the ways in which such chances must be exploited. In this manner the player’s ability to classify and to apply appropriate treatment (in terms of thinking and playing methods) is steadily refined.
De Groot also makes an insightful comment about why chess is more a sport and less a science3:
There was a time when what we now call the scientific conception of chess did not yet exist. Largely as a result of the work of Wilhelm Steinitz (1836-1900) – and not before – chess began to develop so-called scientific traits... He was the first to connect strategic planning with a systematic position investigation in terms of the features of the position... It would be a mistake to suppose that the new school was enthusiastically received by everybody... Nowadays, nearly 70 years later, a battle no longer rages around the theory of Steinitz. In a general sense it has been completely accepted: in a modernized and much more elaborated form it is part of the technical knowledge that every chess master is assumed to possess.
Nonetheless, the game of chess has not become a science... Of course, the quest for the objectively best move is relevant during the thought process – it is known that some grandmasters like Tarrasch, Euwe, and Botvinnik maintain a pronouncedly objective “scientific” attitude towards the choice-of-the-move problem during the game – but the point is that only rarely can the problem be objectively solved. Even when the choice-of-move problem turns out to be objectively solvable in analysis, the time limit and the prescribed fixedness of the pieces on the board make it impossible for the player to attain more than an incomplete proof in a normal match or tournament game. In spite of what laymen may think, Alfred Brinckmann is right when he says “In the chess battle, acting carries more weight than cognition.”
Russian grandmaster Alexei Suetin speaks up against the current set of rules as they apply to the opening:4
... the opening struggle is very complicated and cannot be entirely subjected to general rules. ...It must be emphasized that the dynamism of chess is not an arbitrary, chance process of change, but on the contrary, is subjected to the iron logic of chess. The positional and the material factors are continually transformed during the course of the game, from it first moves onwards.
Another interesting feature of Elements is its application to the art of chess teaching, which had long been a neglected area of chess research in the U.S. before the immigration of many fine instructors from the former Soviet Union. The following is excerpted (and edited) from “The Basis of Positional Chess Teaching,” which I wrote in the early 1970s:5
Have you ever thought about why and when a player’s rating levels off? Was it just lack of talent or did time and ambition enter the picture? I feel that these factors lie within the individual. But there is one aspect that can be controlled externally: the fundamental way that a person learns how to play; that is, the order he learns his chess skills and the priority (and emphasis) he puts on various aspects of these skills. Furthermore, the overwhelming majority of players attempt to improve in a haphazard manner, using a hodgepodge of learning techniques.
Why is this? How can this best be remedied? In short, how does a player get rid of the dogma that can so easily hinder progress? It is not easy to do this if you scoff at aid or considers yourself to be “of sufficient strength.” Most players do not follow suggestions until they are burned by practical experience. That is normal so long as they don’t repeatedly make the same mistakes. Only the very gifted seem to learn by avoiding mistakes instead of making them.
This leads to the question, “What is the correct method?” A player interested in improvement draws, and enlarges upon, his and other’s experiences. I have drawn upon my experiences to form the basis of my theory of teaching.
Elements was originally quite a maverick book, rejected by two publishers back in the 1970s. After the first edition was published in the 1990s, someone on the internet called it a “Cult Classic.” However, I am happy to say that by 1998 not only was Elements enjoying its third edition, but the theory proposed herein was – not so much because of this work, but because the bases are sound – becoming more generally accepted.
In 1998, IM John Watson’s magnificent book on similar (and other) subjects, Secrets of Modern Chess Strategy, was voted the outstanding chess book of that year. At that time I corresponded with John, discussing the similarities of the theory in our books, and have since appeared as a guest on his ICC Chess.fm radio show. While he had no knowledge of Elements when writing his book, I am happy to say that Secrets of Modern Chess Strategy is more than a justification and extension. In my opinion, Secrets and its excellent sequel, Modern Chess Strategy in Action, can also serve as further praxis for many ideas in Elements!
To conclude the introduction, I would like to whet the reader’s appetite with a bit of the proposed new theory.
Static features are factors based upon where the pieces are situated on the board. Dynamic features are based upon analysis; i.e., mentally moving the pieces. This book addresses how static features affect the value of the pieces in a given position. For example, if doubled pawns (a static feature) are often harmful, but are sometimes beneficial, then there must be something more basic than doubled pawns on which one could/should base his positional evaluation – something that will help one determine when doubled pawns are good or bad and by how much. These more basic “somethings,” when found and identified, will then become our elements.
It is also important to differentiate the evaluation of each piece individually in a given position from the evaluation of an entire position. This book is more about the static evaluation of each piece’s value rather than the evaluation of an overall position. However, the new theory can also help evaluate entire positions by providing insight into: Who stands better, By how much; and Why? To this point, one of the proposed elements, coordination, implies the involvement of multiple pieces.
Students who wish to study evaluation usually concentrate on static features such as pawn structures, files, holes, outposts, all of which have been considered the most basic positional ideas. The identification of more basic elements should help the reader achieve a better understanding of evaluation. A discussion of these elements begins in Chapter 2.
1.1 Definitions
Static Features: Aspects of a position that do not require analysis of piece movement. Examples of static features are doubled pawns, open files, and a dark square complex. Proper study of static features nevertheless requires a knowledge of how the pieces move. The most important static features will be defined throughout the book.
This position illustrates many static features:
• Isolated pawns on the a-, c-, and d-files;
• The open b-file;
• The semi-open c- and e-files (for White) and the a- and d-files (for Black);
• Weak squares around White’s king at f3 and h3;
• An outpost white knight on c5 and a strong black bishop on d5;
• Black has a weak bank rank – there are possibilities of checkmate by the white rook if Black is not careful; and
• Possibilities for back rank mates against White, since the bishop covers the white king’s escape square g2
These static features will be discussed in more detail in Sections 1.3, 5.2, and 5.3.
The following is an example of a completely static position, where the position can be evaluated without knowing whose move it is:
This opposite-color bishop endgame is easily drawn no matter who is to play.
Dynamic Features: Aspects of a position that require knowledge of how the pieces move. This includes both potential movement (how pieces could move on an empty board) and movement in legal positions, such as maneuvers and tactics. Dynamic features include both how the pieces move and the time measurement for motion: the tempo.
Tempo measurements are the move – one move by each player – and the ply – one move by one player. So “looking 5 ply ahead” means you are looking ahead three moves for yourself and two for the opponent. The 50-move draw rule requires both players to make 50 consecutive moves without a capture or pawn move for one player to be able to claim a draw.
In the first diagram in this chapter, the a2-pawn is en prise, but it requires dynamics (knowledge of whose move it is) to know if Black could capture it for free. If it were White’s move, 1.a4 or 1.Ra1 could be played, so the safety/vulnerability of this pawn (discussed in Section 2.4) is a dynamic issue.
Here is an even more trivial example, the opposite of the previous static example above:
The dynamic issue of who is to move decides.
This position is statically equal but not dynamically. Both sides have equal material and symmetric positions. But dynamics require that it must be one player’s turn, and whoever is to move can checkmate by capturing the other’s queen.
The following is a more complex example. This position appears fairly quiet.
Therefore, at first it may seem that the positional features should be the key. White’s doubled b-pawns can actually be a strength, since White no longer has an a-pawn and thus not only has more pawns toward the center but also a semi-open a-file. Black’s doubled rooks on the e-file look like a definite plus. It looks as though Black’s pawn on d5 is a fixed target of the white bishop and possibly also the knight if it can get to f4. Black might be able to plant the knight on e4, since for White to remove it by f3 is so weakening.
Black to play
However, all these positional features are somewhat irrelevant since the position is more dynamic than it appears. Black’s move is the key, and if 1...Ng4! is played, threatening the “removal of the guard” sequence 2...N×h2+ 3.Kg1 R×e2, then White has no good defense! For example, 1...Ng4! 2.h3 Nh2+ 3.Kg1 R×e2 4.R×e2 R×e2, and Black will win at least a pawn with a powerful position, as White eventually has to recapture the piece, e.g., 5.K×h2 R×f2. No better is 1...Ng4 2.f3 Ne3+ (clearer than 2...Nh2+) 3.Kf2 Nc2, winning the exchange. Finally, 1...Ng4 2.Kg1 just loses a piece to 2...R×e2.
Strategy: The part of chess that involves evaluation of the position (which side stands better, how much, and why, including assessment of strengths and weaknesses) and the accompanying planning. Strategy may include both long and short term plans, and involves both players’ pieces.
Tactics: A forced maneuver that may include (but not necessarily be limited to) winning material, including pawn promotion, and executing a mating attack. From simplest to most complex, tactics include:
(A) En prise – a piece that is attacked but not guarded
Black’s queen is en prise and White can capture it with 1.R×e8#. If it were Black’s move, then White’s rook would be en prise and Black could play 1...Q×e2.
(B) Counting – in its basic form, this is how a player evaluates material trades by the “I take-you take” thought process. Counting determines whether material can be won by any sequential capturing sequence. Take the basic Scotch Game setup, after 1.e4 e5 2.Nf3 Nc6 3.d4 (D)
Here the black e-pawn is not en prise since it guarded by the knight on c6, but counting on e5 reveals that White is threatening to win the pawn with the simple 4.d×e5. This is an example of basic counting on one square, e5. But, it is Black’s move, and the loss of the pawn can be prevented by capturing on d4, where counting shows the white d-pawn to be safe. The main sequence is usually 3...e×d4 4.N×d4 with equal material. A further 4...N×d4 not only does not win material for Black, but is considered a minor positional mistake by activating the white queen to a strong square after 5.Q×d4. More on this later when “time” is discussed in Section 2.7.
Black to play
One of my students, not a weak player at all, made a very basic multi-square counting mistake in the following position after 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.O-O N×e4 5.Re1 d5 6.Bb5 Bc5 7.d4:
Black to play
Counting indicates that White’s previous move is safe because after 7...e×d4 8.N×d4 B×d4, White can safely capture with the queen, 9.Q×d4, as the knight on c6 is pinned. This unfortunately gave Black the erroneous idea that by removing the pin, pressure could be added to d4, and 7...a6?? was played, ...losing to the counting sequence on the squares c6 and c5: 8.B×c6+ b×c6 9.d×c5 and White has won a piece.
White to play
(C) Single Motifs – Examples are double attacks, double threats, pins, removal of the guard, and back rank mates. We have already seen some back rank mates – let’s consider a couple of other common examples:
White to play
White wins the queen with the deflection 1.B×h7+, winning the queen after 1...K×h7 2.Q×f8. Deflection is a type of removal-of-the-guard motif.
White to play
In this trap in the English Opening, White wins a piece with 1.d4 e×d4 2.e×d4 Bb6 3.d5, forking the knight and bishop.
(D) Combinations of motifs without sacrifice.
White to play
White wins by 1.N×d6, removing the guard on e4 and, regardless how Black recaptures, 2.Qe4 follows, with the double threat of 3.Q×h7# and 3.Q×a8. Since this sequence combines removal-of-the-guard and a double attack, it is a combination.
(E) Combinations of motifs with sacrifice.
White to play
Incorporating a sacrifice for White converts the previous problem into a sacrificial one: White wins by sacrificing the exchange with 1.R×d6, again removing the guard on e4, allowing 2.Qe4 with the double threat of 3.Q×h7# and 3.Q×a8. White wins a piece. This is also an example of a pseudo-sacrifice, since White wins the material back and more with no risk. A real sacrifice, such as a gambit, does not win the material back by force but instead receives positional compensation, such as more mobility or time.
The following is a much more difficult sacrificial combination. It is from an online game I played against a computer to demonstrate to a student how to analyze. (D)
White has won a piece for two pawns, but how to put Black away? I calculated the clever sacrifice 22.Bh6!. If Black now moves the g-pawn, he loses the exchange after the capture on f8, but White is threatening to put the queen on the g-file and win the g-pawn, so 22...g×h6 23.Qg4+ Kf7 24.Qh5+ Kg8 25.Re3!. This is the move I had to have seen when I sacrificed the piece.
White to play after 21...d5 Phillytutor (Heisman)-Molasses Internet Chess Club 2007
White to Play after 25.Re3! Heisman-Molasses Internet Chess Club 2007
Black is helpless. The game concluded 25...Bg7 26.Qg4 Q×e6 (if 26...Kf7 27.Q×g7+ Ke8 28.Q×h8+) 27.Q×e6+ Kf8 28.Q×e7+ Kg8 29.Rg3 f5 30.R×g7#.
Tactics also include defensive tactics that use the same tools to avoid material loss or mate:
White to play and not lose
The only way to stop 1...Qb6+ is 1.Qa5. If White does not play 1.Qa5, then he loses material: 1.Rd2? Qb6+ 2.Kh1 Nf2+ 3.Kg1 (better is 3...R×f2, losing only the exchange) 3...N×e4+, winning the exchange and a pawn. Not as good is 3...Nh3+ 4.Kh1 (4.Kf1? Qg1+ is winning) and the best Black has is to repeat with 4...Nf2+ 5.Kg1 N×e4+, as above, for if 4...Qg1+?? 5.R×g1Nf2+ is not the famous Philidor’s Legacy combination ending in smothered mate because White defends with 6.R×f2. The ability to handle tactics is the prime factor in determining a player’s chess strength; some have called it 99% of the game. While tactics supersede strategy and positional play, superior strategy produces favorable tactics, and positional play is the guidepost of strategy.
Positional Play: This is strategy that emphasizes piece and pawn placement, as opposed to tactics. An example is taking advantage of an opponent’s static weaknesses. Most chess players develop their tactical skills before the more refined positional skills. Advanced positional play is regarded as more subtle and, I believe, more easily misunderstood. This book pertains primarily to positional play and its evaluation.
Positional Player: A player whose planning is dominated by positional and strategic thinking. Positional players de-emphasize tactical play, instead seeking to avoid, create, and parry threats via positional means. The perceived distinction between tactical and positional players is often exaggerated at the highest level of play.6 At that level, everyone is tactically competent. However, whereas a strong tactical player often immediately strives for unbalanced, tactically tense positions, a positional player usually will try to build up a winning position by increasing positional pressure (via accumulating positional advantages) until the opponent “breaks down,” allowing a simple tactical finish.
One other relevant definition, from mathematics:
Positive Correlation: If factor A has a positive correlation to factor B, then there will be a tendency that if A occurs, that B will occur, and vice versa. This does not imply that one causes the other, or that one must occur if the other does. Let’s give examples from outside and inside the chess world.
In investing, there should a positive correlation between risk and reward: if you take more risk, you want more reward; otherwise it is just gambling. For example, companies with risky balance sheets usually are required to to issue bonds at higher interest rates than more stable companies. In school, students with higher scores on IQ tests have a positive correlation with good grades. In the latter case the correlation is weaker, as companies perceived with risk almost always pay higher interest rates; good grades and high IQ, unfortunately, don’t always go together.
In chess, there is a positive correlation between doubled pawns and semi-open and open files. When you double a pawn, you will almost always create an open or semi-open file.
White to play
This is the most common situation. White’s capture of 1.b×c3 creates a semi-opens b-file for White.
The next position is the exception: the b-file is already semi-open for White and the e-file for Black, but capturing 1.c×d4 does not create any more open or semi-open files, since White is just transferring the doubled pawns from one file to another. Note that it is impossible to have doubled pawns without at least one semi-open or open file because there are only eight pawns and eight files! (D)
But the converse is not necessarily true: when there are open and semi-open files, there are not necessarily any doubled pawns. (D)
So although there is a positive correlation between doubled pawns and open/semi-open files, it is not very high.
White to play
White to play
1.2 A Brief History of Positional Theory
The first major name connected with positional theory is the 18th century Frenchman, André Philidor. Philidor, an accomplished musician, is recognized as the first player to realize and popularize the importance of pawns, although his references to pawns were not always in a positional context. “Pawns are the soul of chess” is Philidor’s famous maxim. However, there is no specific lasting school of positional play emanating from Philidor. Another century would pass before there was any extensive, written positional theory.
The next person to make a contribution to positional theory, though not written, was the American “world champion” Paul Morphy. Morphy’s tour of Europe in the late 1850s established him as head and shoulders above any contemporary. It was not until years after Morphy retired that the “secrets” to his success were fully appreciated and understood. Morphy was the first player to fully implement principles that would not gain recognition until later – “develop all your pieces before you attack,” “develop each piece and put it at once on the correct square,” “a premature attack is doomed to failure,” and its converse, “if you have the advantage, you must attack (“... or convert it to another advantage,” Larry Evans) or you will lose the advantage.” Consider Morphy’s rapid mobilization of forces in his famous game with the Duke of Brunswick and Count Isouard:
1.e4 e5 2.Nf3 d6 3.d4 Bg4? 4.d×e5 B×f3 If 4...d×e5, 5.Q×d8+ gets out of the pin and wins a pawn after 5...K×d8 6.N×e5. 5.Q×f3 d×e5 6.Bc4 Nf6 7.Qb3 Qe7 So that if 8.Q×b7, Black gets some relief by trading queens with 8...Qb4+. 8.Nc3 c6 9.Bg5 b5? 10.N×b5 A sacrifice made possible by White’s large lead in activity and the coordination of his army. 10...c×b5 11.B×b5+ Nbd7 12.0-0-0 Rd8 13.R×d7! The most efficient way to increase the pressure on the d7-pin. 15...R×d7 14.Rd1 Qe6 Allowing the knight on f6 to guard d7 and hoping to trade queens. 15.B×d7+ N×d7 (D)
16.Qb8+ N×b8 17.Rd8#
Morphy’s command of tactics and these principles made him superior to his contemporaries and one of the best players of all time.
White to play Morphy-Duke of Brunswick and Count Isouard
Even though Philidor and Morphy had contributed to positional theory, there still was not a general recognition or acceptance of a “positional school,” and neither player had established a following to carry on “a positional school of thought.”
It was not until the late 19th century that the Austrian-American World Champion, Wilhelm Steinitz, became the first to systematically advocate an all-encompassing positional chess theory. It is interesting to note that the American World Champion, Robert Fischer, whom many believe to be the strongest player of all time, did a comprehensive study of Steinitz’s games and theories. For example, Fischer credited Steinitz for the ideas in the Two Knights Defense, which Bobby used to defeat Arthur Bisguier at the New York State Open in 1963.7 Then, beginning in April 1964, Fischer began analyzing Steinitz’s games in his Chess Life column. These articles showed that Fischer had seriously scrutinized even the lesser known Steinitz games.
Steinitz conceived the theory of pawn formations, systematized ideas such as pawn chains and islands, isolated pawns, doubled pawns, and other fragmented formations. He was the first to publicize the importance of the set of static features that accompany such ideas: weak squares and weak square complexes, holes, and outposts.
Today, Steinitz’s contributions seem very basic and logical, yet it required considerable genius to be the first to recognize and advocate such an encompassing theory.
In the early 1900s, Dr. Siegbert Tarrasch of Germany continued Steinitz’s work, advocating (his critics would say “dogmatizing”) the “classical” theory that included the classical pawn center (pawns on e4 and d4 for White), bishops on their original longer diagonal (e.g., White’s f1-bishop to c4), knights on c3/c6 and f3/f6, early kingside castling and similar principles that dealt with occupation of the center.
Example of classical “Tarrasch” development
Tarrasch’s theories seemed logical successors to Steinitz’s, and the common perception from 1900 until around 1920 that the Tarrasch school was the “correct” one; “classical play” was the rule of the day, with 1.d4 and the Queen’s Gambit Declined replacing the “weak” 1.e4 as the most popular opening workhorse. This school was exemplified by the rise and play of the Cuban World Champion José Raúl Capablanca. Capablanca was a natural talent who easily assimilated the classical information into a very simple and straightforward positional style. Most of Capablanca’s games are marked by classical openings: direct center occupation as opposed to indirect control using a fianchetto.
But in the 1920s, a new and vocal school, the hypermoderns, rose to challenge the Tarrasch dogma. Hypermodern theory was championed by three grandmasters in particular: the Czechoslovakian Richard Réti, the Hungarian Julius Breyer, and the Latvian-Dane Aron Nimzovich. They argued that the Tarrasch school was too closed-minded and was suffocating expansion of positional theory.
The hypermoderns claimed that occupation of the center did not imply control, and that “premature” occupation of the center would only set up targets for attack. Réti further suggested that a non-centralized, fianchettoed bishop was a strong piece, and that a delay in fixing the center pawns could be strong and flexible, often enabling a hypermodern player to establish superiority in the center via counterattack.
As a result of hypermodern theory, many openings became acceptable: the King’s Indian Defense, the Nimzo-Indian Defense, the Queen’s Indian Defense, the Benoni, the Grünfeld, the Reti, the Alekhine Defense, Pirc Defense and later the Modern Defense. More traditional openings and defenses, such as the popular Sicilian Defense in response to 1.e4, were also influenced by the popularity and gradual acceptance of the hypermodern openings.
Example of hypermodern development
Aron Nimzovich is usually remembered for his fine legacy—the classic book on positional play, My System. Even today, many grandmasters point to that book as the turning point in their personal development, and the development of positional chess theory.
In that classic work, Nimzovich attacked Tarrasch’s classical school, advocating several fianchetto openings and other indirect methods of attacking the classical center by means of flank attacks and timely center counter-thrusts. In doing so, Nimzovich developed basic and subtle ideas that are vital to understanding modern chess: attacking a pawn chain at its base, placing rooks on the seventh rank, prophylactic play (that is, as we shall see, limiting the opponent’s mobility and flexibility), and overprotection.
Overprotection deserves a simple explanation: For example, if a pawn is attacked four times and defended four times, all four of the defending pieces are usually tied down to the defense. The defender might consider these four defending pieces as useless, even though the pawn is adequately protected. However, if the pawn were to be protected by a fifth piece (overprotection), then all five of the protecting pieces would be freer for other duties.
White overprotects e5 by playing Re1 as a third defender.
This freedom is a tremendous asset, the result of a simple and brilliant piece of strategy. Nimzovich made many other interesting suggestions, and his theories are considered to be at least as noteworthy as world championship contender Tarrasch’s. Moreover, Nimzovich’s theories were more flexible than the German doctor’s earlier contributions. For more on overprotection and other types of guarding, see Section 2.4.
Hypermodern theory eventually garnered the full support of the chess establishment, culminating in the victory of the Russian expatriate Alexander Alekhine over Capablanca in their 1927 world championship match. Although Alekhine chose mostly classical openings in the match with Capablanca, he was the first world champion to incorporate both classical and hypermodern theories not only in his overall opening repertoire, but also in the overall approach to the game.
Alekhine was more than just a world champion who incorporated hypermodernism into his overall approach; he was the messiah of the dynamic school, a forerunner of contemporary grandmaster play8
The post-Alekhine schools, for instance the Soviet School of Chess, were more difficult to put into historical perspective in 1974 because their effect was not yet in clear historical perspective. What they primarily accomplished was to amalgamate previous theories, refine them, and pick out the best of the ideas that passed the test of time. After 1974, the Soviets, especially Kasparov beginning in the 1980s, laid the basis for modern dynamic play (see below). This sharpening of knowledge and overall play has caused a greater and greater emphasis on openings, so that whereas Capablanca or the American champion Samuel Reshevsky could rise to or near the top of the chess world without an intensive study of opening theory, today such a feat becomes more and more unlikely. Even an original player like Bent Larsen wrote:
At this time I was about to give up the Najdorf Variation, which I had used a great deal for several years.
So much theory had been published on it that nearly all masters knew a good line for White and could play many moves without thinking very much. I had also found some improvements for White myself and was afraid that others would find them and play them against me.9
The contributions by Soviet world champions Mikhail Botvinnik, Vassily Smyslov, Tigran Petrosian, and Anatoly Karpov have solidified the current positional chess theory. Fischer possessed quite a universal style, often positional, but rarely avoiding dynamic possibilities, especially in the opening. In that sense, Fischer is more like Capablanca in clear positions and Alekhine and Botvinnik in dynamic ones.
These players learned to avoid slavish adherence to static theory. Former world champion, Russian Garry Kasparov, may be thought of as a super-Alekhine, the ultimate dynamic school player. Kasparov often seemed to provide flair with his “anti-positional” attacking moves; anti-positional for 1974, but quite in keeping with 2009 dynamism that he, more than any other single player, helped promote. The Bulgarian star Veselin Topalov follows in Kasparov’s dynamic footsteps, while another champion, Vladimir Kramnik, usually plays in a more classic positional style. Finally, the champion in 2009, Viswanathan Anand, possesses a universal style reminiscent of Fischer.
Whether he was the chief protagonist or not, Kasparov’s influence on dynamic play led the way for modern theory to embrace the more practical, “If it works, do it” style over the former “it doesn’t look right and can’t be justified by static evaluation” mode. What boosted the Kasparov stance even more was the ascendency of computers to grandmaster level play in the 1990s and “even-better-than-world-champion” level in the next decade, and the “style” of their super-strong play.
The advent of super-strong computer chess programs such as Deep Blue, Fritz, Junior, Shredder, Hydra, Hiarcs, Zappa, and Rybka have shown that not only does it take a lot to lose a chess game. There are formerly “incorrect” lines that have be shown to be quite playable, and “best moves” occasionally might not “look right” to someone steeped in the “Petrosian-Karpov” statically sound school. Computers are the ultimate pragmatists, and while their excellent tactical play is unquestioned, tactics flow from superior positions and their positional play sometimes is based on the kind of flexibility and coordination that will be discussed in Elements.
I believe that Elements, originally written years before Kasparov burst onto the scene (the first edition was written 1974-75 but was not released until 1990), was one of the first books to advocate a “more-basic-than-static theory” approach founded on a completely different system. As a by-product chess instructors have additional fuel to aid in teaching beginners. This fuel should help instructors impart understanding, rather than just the ability to recognize various static features as assets and liabilities.
There were many fine dynamic players in the 1960s (Tal, Geller, Kortschnoi, Suttles), just as there are many fine positional players today (Kramnik and Leko immediately spring to mind). The key is that in the Petrosian – and later Karpov – school, dynamic play was viewed as an aggressive by-product; whereas today it is much rather the norm, and players who tend to shy away from dynamically imbalanced positions are considered rather conservative.
Entire books have been devoted to the subject of chess theory development, the foremost being Euwe’s The Development of Chess Style and Saidy’s The Battle of Chess IdeasModern Ideas in ChessMasters of the ChessboardDynamic ChessThe Soviet School of Chess