Main Your Jungle Guide to Rook Endings, Efstratios Grivas 2020-TLS

Your Jungle Guide to Rook Endings, Efstratios Grivas 2020-TLS

0 / 0
How much do you like this book?
What’s the quality of the file?
Download the book for quality assessment
What’s the quality of the downloaded files?

Please join #TLS_CHESS_NETWORK Links:


Telegram Channel:

TLS CHESS ARCHIVE (Back up Channel)    

All Chess Materials will be shared to improve knowledge...

Telegram Group (Request) Invite Link:

Telegram Group (Practice & Discussion) Invite Link:

TLS CHESS PGN BANK Telegram Channel:

1000s of Rare & Important PGNs have already been shared...


ChessBase Collections-TLS            

Chessable Video Courses-TLS 

Exclusively to share 'Chessable Video Courses'

The Chess World-TLS

The Internet Chess Club-TLS                            

GingerGM-TLS Videos-TLS


Exclusively to share 'Chess24 Video Courses'

Nurtr Chess Gurukul–TLS

Only Nurtr Chess Gurukul by the  Celebrated GMs and Coaches Videos are shared...  

Remote Chess Academy Collections-TLS  


New in Chess-TLS                     

Quality Chess-TLS

Everyman Chess-TLS

Gambit Publications-TLS

Batsford Chess Books-TLS

Chess Magazines Collection-TLS


Contact me in Telegram: @TLSubramanian

Thinkers Publishing
PDF, 14.88 MB
Download (pdf, 14.88 MB)

Most frequently terms


To post a review, please sign in or sign up
You can write a book review and share your experiences. Other readers will always be interested in your opinion of the books you've read. Whether you've loved the book or not, if you give your honest and detailed thoughts then people will find new books that are right for them.

The Best Thing

PDF, 6.17 MB
0 / 0


PDF, 4.91 MB
0 / 0
Computer Evolution

The Lucena Position
The Philidor Position
The Vancura Position
Rook & a+h Pawns vs Rook
Rook vs Three Pawns
Two Rooks vs Rook & Three Pawns
Same Side Pawns 4:3 — Part 1 (1)
Same Side Pawns 4:3 — Part 1 (2)
Same Side Pawns 4:3 — Part 2
Active Rook & 3 Pawns & a-pawn vs Rook & 3 Pawns
Passive Rook — Part 1: Rook & 3 Pawns & a-pawn vs Rook & 3 Pawns (1)
Passive Rook — Part 1: Rook & 3 Pawns & a-pawn vs Rook & 3 Pawns (2)
Passive Rook — Part 1: Rook & 3 Pawns & a-pawn vs Rook & 3 Pawns (3)
Passive Rook — Part 2: Rook & 2 Pawns vs Rook & 1 Pawn


Rook & 3 Pawns & b-pawn vs Rook & 3 Pawns
Rook & 3 Pawns & c-pawn vs Rook & 3 Pawns
Rook & 3 Pawns & d-pawn vs Rook & 3 Pawns
Shattered Pawns — Part 1
Shattered Pawns — Part 2
The Isolani in the Endgame
Drawn Endings
Double Rook Endgame (1)
Double Rook Endgame (2)
The Lasker’s Steps
A Trapped Rook
The Loman Move


Efstratios Grivas
Thinkers Publishing 2020
Managing Editor
Romain Edouard

Assistant Editor
Daniel Vanheirzeele

Bob Holliman

Graphic Artist
Philippe Tonnard

Cover design
Iwan Kerkhof

i-Press ‹›


First edition 2020 by Thinkers Publishing

Your Jungle Guide to Rook Endings
Copyright © 2020 Efstratios Grivas

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 or
ot; herwise, without the prior written permission from the publisher.

ISBN 978-94-9251-074-7

All sales or enquiries should be directed to Thinkers Publishing, 9850 Landegem, Belgium.




a good move


a weak move

!! an excellent move
?? a blunder
!? an interesting move
?! a dubious move
™ only move
= equality
∞ unclear position
© with compensation for the sacrificed material

White stands slightly better


Black stands slightly better

± White has a serious advantage
µ Black has a serious advantage
+– White has a decisive advantage
–+ Black has a decisive advantage
‚ with an attack

with initiative

„ with counterplay
… with the idea of
¹ better is
‹ worse is
N novelty
+ check



Dear Reader,
In my younger chess years, in the early 80’s, I was lucky enough to have as my main coach the
legendary Dr. Nikolay Minev. He had been an Athens resident for approximately two years before
immigrating to the United States. He mainly worked with Greece’s National Men’s Team, but also
with a group of promising youngsters, including myself.
Dr. Nikolay Minev was a prominent coach, quite educated both in life and chess and he acted as my
second father helping me to develop into a Grandmaster. He was quite fond of the great Akiba
Rubinstein and of rook endings. It is no coincidence that he wrote several books on those two
It was natural that I was inspired by his work and I also tended to fall in love with Akiba and rooks!
In my chess career I have played many rook endings, quite brilliantly in most of them. I studied them
with one of the best and I became quite proficient at them!
Having completed my playing career and become a top-level coach it is my turn to author a book on
rook endings dedicated to my wonderful coach. I chose to write a book on advanced rook endings as I
simply did not wish to write another book that would be like the many already available.
I have done my best to present analysis and articles I have written over the past 10–15 years. This
work has been presented in my daily coaching sessions, seminars, workshops, etc. The material has
helped a lot of trainees to develop into quite strong players gaining international titles and
Now, it is your turn to taste and enjoy it!

Sharjah 2019
Efstratios Grivas

Chapter 1 deals with basic knowledge: the Lucena, the Philidor and the Vancura positions. As nearly
all rook endings will result in one of these three positions it is necessary that we go over the basics
before advancing to more complex material.
Chapter 2 deals with some extraordinary endings with rook pawns and rook(s) vs a fair number of
pawns! This is a difficult subject where concrete knowledge is demanded.
Chapter 3 will teach us how to deal with pawns on the same side and if and how we can take
advantage of our extra material.
Chapter 4 is probably the most complicated of endings with 3 vs 3 pawns with an extra a, b, c or d

pawn. In both cases the active or passive rook is examined.
Chapter 5 deals with shattered pawns on one or both sides and how the healthy side can prevail.
Chapter 6 teaches us how to benefit from a useless isolani in the endgame.
Chapter 7 instructs us on how to hold drawn endings. Many such endings have been lost by high
rated players due to a lack of accuracy.
Chapter 8 is full of rooks! We will examine endings with four rooks on the board. This is not an easy
situation as many new elements are presented.
Chapter 9 concludes the book and the subjects are Lasker’s Steps, the Trapped Rook and the Loman


This book is dedicated to my great Coach Dr. Nikolay Minev. Back in 1981 & 1982 he coached me
from being an ordinary kid in the neighborhood to an adult chess player. He made me understand
what chess is and how to deal with it both as a player and as a coach. May he rest in peace.
Nikolay Minev (8 November 1931 – 10 March 2017) was a Bulgarian International Master (IM) and
noted chess author. N.Minev was awarded the IM title by FIDE in 1960 and he was the champion of
Bulgaria in 1953, 1965, and 1966. He played for Bulgaria in the Chess Olympiad six times (1954,
1956, 1958, 1960, 1962, and 1966). N.Minev’s best international results were 3rd at Varna in 1960,
2nd at Warsaw in 1961, tie for 1st at Sombor in 1966, and 2nd at Albena in 1975. He contributed to
early editions of the ‘Encyclopaedia of Chess Openings’ and the ‘Encyclopaedia of Chess Endings’.
N.Minev and his wife immigrated to the United States in 1982 and settled in Seattle, Washington. He
was associated with GM Yasser Seirawan and his magazine ‘Inside Chess’ in the 1980s and 1990s.


It is a testimony to the accuracy of the author’s original analysis that only twenty or so substantive
changes have been made in the 206 relevant games and studies contained herein.
It is interesting to see how the author’s original work benchmarks against the definitive knowledge of
the databases. Of course, some deep wins are ‘over the horizon’ but even shallow tactical shots can be
missed by an over-enthusiastic pruning of the search tree.
Zugzwang situations can confuse the computer if the ‘null-move heuristic’ is not switched off and
fortresses can be rendered magically invisible.

Guy Haworth
London, September 2019
Guy Haworth is an Honorary Fellow in computer science at the University of Reading.
Over the years he has championed the creation and use of chess endgame databases, mainly in the
pages of the ICGA Journal.
Guy’s specific contribution was to extend the author’s sub-7-piecer database checking of the
positions. He has used Ronald de Man’s DTZ50 endgame tables rather than the ‘Lomonosov
DTM(ate)’ ones. Thanks should go to Ronald de Man, Bojun Guo (for the tables) and Niklas Feikas
(for the site, which deserves more publicity.


150 years ago, the quality of played endings was quite low as chess players rarely reached the
As strategy and maneuvering were starting to overcome tactics more and more endgames were
suddenly seen. As a result, a need for endgame material became essential.
‘Basic Chess Endings’ (abbreviated BCE) is a book on endgames which was written by Grandmaster
Reuben Fine and originally published on October 27, 1941.
It is considered the first systematic book in English on the endgame phase of the game of chess. It is
the best-known endgame book in English and is a classic piece of chess endgame literature.
The book was dedicated to World Champion Emanuel Lasker who died in 1941 (the year the book
was published). It was revised in 2003 by Pal Benko.
‘Basic Chess Endings’ was written by Reuben Fine in only four months and was published in 1941 by
McKay (a division of Random House) in hardback. The book used the now obsolete descriptive chess
notation and used the old system of using the abbreviation ‘Kt’ to stand for knight instead of the more
modern ‘N’.
In a 1984 interview Fine said that it took him three months to write the book. He said that organizing
the material gave him no trouble, but it was hard work coming up with exemplar positions therefore
he created many of the examples.
There was a Hardback Limited Edition of 500 signed by Ruben Fine and sold by the USCF in the
1940s.The hardback edition was reprinted at least as late as 1960. The copyright was renewed in 1969
as the book went through many paperback impressions. It went through ten printings in paperback by
The original book contains 573 pages and 607 diagrams of positions. Many other positions were
given by listing the location of the pieces rather than showing a diagram. The 2003 revision contains
587 pages with 1330 positions, most with diagrams.
Many other good books followed BCE. They were written by Yuri Averbakh, Vasily Smyslov and
Paul Keres. But all of them had some analysis flaws as well — it was rather difficult to be accurate.
The computers’ evolution changed everything in the way we learn and analyze endgames today, at
least to some degree!
Nowadays chess players have access to many good and accurate endgame books, mainly because of
the tablebases which changed our endgame vision.
These tablebases contain the game, the theoretical value (win, loss or draw) of each possible move in
each possible position and how many moves it would take to achieve that result with perfect play.
The tablebases act as an oracle always providing the optimal moves. Typically, the tablebase records
each possible position with certain pieces remaining on the board and the best moves with White to

move and with Black to move.
Tablebases are generated by retrograde analysis working backwards from a checkmate position. By
2005, all chess positions with up to six pieces (including the two kings) had been solved — this is
what we call today the Nalimov Tablebases.
The tablebases of all endgames with up to six pieces are available for free download and may also be
queried using web interfaces. They require more than one terabyte of storage space.
By August 2012, tablebases had solved every position with up to seven pieces. The positions with a
lone king versus a king and five pieces were omitted because they were ‘rather obvious’. Today we
call this the Lomonosov Tablebases. The size of all tablebases up to seven pieces is about 140 TB.
The solutions have profoundly advanced the chess community’s understanding of endgame theory.
Some positions which humans had analyzed as draws were proven to be winnable; the tablebase
analysis could find a mate in more than five hundred moves. Clearly this is far beyond the horizon of
humans and even beyond the capability of a computer during play.
For this reason, the 50-move rule has been questioned since many positions are now seen to exist that
are a win for one side but would be drawn because of the 50-move rule.
Tablebases have enhanced competitive play and facilitated composition of endgame studies providing
a powerful analytical tool.
Of course, we can expect the 8-pieces version, but I feel that it will be a bit delayed!
Chess players tend to think that theory only applies to openings. This is a very wrong attitude as logic
and practice have proven that theory counts for every part of a game.
‘To learn and to play endgames well the chess player must love endgames’ — Lev Psakhis.
Different kinds of endgames have specific characteristics and rules. Every serious player must know
many typical positions and main principles of all types of endings.
That knowledge should help us during the game, but it is not yet enough knowledge to make one a
good player as there are just too many different endings, some of them with two or more pieces, some
very complex.
To be comfortable and play those complex endings well requires specific knowledge and specific
ways of thinking. We will call it ‘endgame thinking’.
First, we will separate endgames into two categories:
SOS Tip 1 — Endgame Categories
1. Tactical — Tactical complex endgames must be treated as a middlegame.
2. Strategical — Pure Endgames.
As in the middlegame our thoughts are busy with calculation based on the specific characteristics of
the position.

But more often in our games, as practice proves, we have so-called strategical endgames where even
computers have difficulties finding the best moves. These are the endgames that do not rely on
It is known from the time of Jose Raul Capablanca that in strategical, positional endgames we must
think by plans and schemes. Variations and calculation play a secondary role.
First, we must understand where to put our pieces in order to improve the position. After we have this
understanding, we start to form a plan. This is what we call ‘endgame thinking’.
Thinking correctly, in the right order, will significantly simplify the process of calculation.
SOS Tip 2 — Endgame Thinking
1. Piece placement.
2. Plan forming.
3. Do not rush!
The question is how to approach ‘endgame thinking’, how to switch on the right button in our brain
for the task at hand. The Russian master Sergei Belavenets, a great chess thinker, gives the following
great advice:
‘After tactical complications, when our brains have been busy with calculation of beautiful variations
the exchange of pieces might follow and some kind of prosaic endgame arises.
Every player has to spend a few minutes, if the clock allows, just to relax, and to calm down your
emotions in order to look differently at the position. This investment of time will pay back later in the
I understand this advice was given before World War II when they had very different time controls.
So do not waste your time, just switch on the button in your brain and remember that positional
endgames require ‘endgame thinking’ by schemes, by plans.
And do not forget to respect your opponent’s ideas. The art of endgame play is how to achieve your
plan and subdue the opponent at the same time. And this book is here to help you understand it!


The endgame is the moment of truth. It is the phase of the game where we will try to reap the seeds of
our effort regardless of whether that is the full point of victory or the half point of the draw.
The significance of errors increases in the endgame as the opportunities for correcting them are few.
The following rules are considered Golden Rules of the Endgame. They were recorded by GMs
Reuben Fine and Pal Benko, two of the world’s greatest experts in this field:
1. Start thinking about the endgame in the middlegame.
2. Somebody usually gets the better deal in every exchange.
3. The king is a strong piece: Use it!
4. If you are one or two pawns ahead, exchange pieces but not pawns.
5. If you are one or two pawns behind, exchange pawns but not pieces.
6. If you have an advantage do not leave all the pawns on one side.
7. A distant passed pawn is half the victory.
8. Passed pawns should be advanced as rapidly as possible.
9. Doubled, isolated and blockaded pawns are weak: Avoid them!
10. The easiest endings to win are pure pawn endings.
11. Passed pawns should be blockaded by the king.
12. The only piece that is not harmed by watching a pawn is the knight.
13. Two bishops vs bishop and knight constitute a tangible advantage.
14. Bishops are better than knights in all except blocked pawn positions.
15. Do not place your pawns on the color of your bishop.
16. The easiest endings to draw are those with bishops of opposite colors.
17. Rooks belong behind passed pawns.
18. A rook on the seventh rank is sufficient compensation for a pawn.
19. Not all rook endings are drawn!
20. Perpetual check looms in all queen endings.
21. Every move in the endgame is of the utmost importance because you are closer to the moment of
Of course, there are plenty of other guidelines for dealing with the endgame:
1. Activate you king.
2. If you have more pawns than your opponent then exchange pieces not pawns.
3. If you have fewer pawns than your opponent exchange pawns not pieces.

4. Try to create a passed pawn.
5. Protected passed pawns are very strong.
6. Outside protected passed pawns are usually decisive.
7. Try to promote a passed pawn.
8. If your opponent has a passed pawn, try to blockade that pawn.
9. Bishops are generally stronger than knights.
10. Bishops of opposite color increase the chances of a draw.
11. Be aggressive with your rooks. If your choice is between defense and counterattack, always
12. Rooks belong behind passed pawns.
13. Know the basics.


There are five basic principles that must be followed faithfully in rook endings:
1. Rook behind the pawn: The placement of the rooks in relation to the pawns is very significant.
The rook must be placed behind the pawn whether the pawn is yours or not. With every move the
pawn makes the radius of our rook will increase and that of the opponent’s will decrease.
2. Active rook: In all rook endings the active handling of the rook is almost always the indicated
course of action. The initiative and attacking possibilities must always figure in our plans and moves.
3. Active king: As in all endings the active king has the first say as the endgame is its finest hour.
This is especially true when the king can cooperate harmoniously with the rook as it can dynamically
help us solve the problems posed by the position.
4. Planning: Our moves must be part of one or more plans. Active plans must be directed towards the
sector of the board where we are superior and, correspondingly, defensive plans must be directed
towards the area where we are inferior.
5. Combination of all the above: When we can combine the above- mentioned elements then we
will be able to extract the maximum from our position!
Mark Dvoretsky makes a general quote: Rook activity is the cornerstone in the evaluation and play of
rook endgames.
This activity may take diverse forms: from attacking the enemy pawns, to the support of one’s own
passed pawns, to the interdiction or pursuit of the enemy king.
There are indeed times when the rook must remain passive and implement purely defensive functions.
But even then, one must stubbornly seek out any possibility of activating the rook, not even stopping
at sacrificing pawns, or making your own king’s position worse.’
Study 1
Nikolay Nikolaev Minev
Shakhmatna Misal 1972 □


This looks like an easy draw for White as the extra black f-pawn is going nowhere. But a closer
examination shows that White can get more than a draw.
And White wins. For example 1...Re8+ 2.Kd6!+– or 1...f4 2.Ke6!+– and Black loses the rook. An
excellent example of the importance of piece placement and activity.
Study 2
Shakmaty v SSSR 1939 □


1.Rg8+ Kd7 2.Rg7+ Ke8 3.Rh7! c4 4.Kd6 Rd8+ 5.Ke6


The most important element in modern chess practice is probably correct planning. The plan is
associated with evaluation and execution, three valuable concepts that cannot be separated and which
most of the time determine the fate of our positions.
During my training sessions my trainees are asked to use the famous EPE procedure shown below.
SOS Tip 1 — EPE
1. Evaluation (strategical & tactical)
2. Plan (ours and our opponent’s)
3. Execution (calculation & move)
That’s the right order although many trainers/authors do not really follow it as they prefer to teach
you to first move and then to think. With this method they have some success, but they will never
have a GM or a player near such strength.
Chess is a mind game — it asks you to think to find solutions even if these solutions are obvious.
In every position you must know where you are (evaluation), where you want to go (plan) and how
you will get there (execution). So simple but we tend to forget this procedure in the heat of the battle.
Nowadays chess games between decent players are full of small plans of different types and ideas
based on purely strategic and tactical motives. The one who will evaluate, plan and execute better
than the opponent earns the first option to win the game.
A closer examination of games played between strong players will prove that there is some harmony
among their moves and some central idea that guides the movements of their forces. This is what we
call a plan. A good chess player refuses to act without any plan even if this plan sometimes turns out
to be mistaken in the long run.
First, it is important to identify the most important strategic and tactical elements of each position.
According to the needs of the position we should create our plan and stick to it. Games where only
one plan is used are rarely seen these days. We are usually obliged to create several small plans which
are just parts of our main goal: winning!
Many weak players are not able to construct an acceptable plan. This has nothing to do with
intelligence but just with basic chess education. Today with so many books, electronic help and
trainers available anything can be learned and with constant practice it can be understood and
assimilated. Without basic chess education and knowledge, we can’t go very far. Therefore, the first
step is to understand the basic strategic and tactical elements that govern our game.
Plans are necessary and can be found in every phase of a chess game. From the early opening till the
late endgame chess players create the necessary plans derived from such varied factors as the
occupation of an important square to an aggressive attack.
Most common are the plans we draw right after the end of the opening phase, but this is not the rule.


Nowadays with opening theory having gone very far the choice of a certain opening very much
depends on the plan we would like to use!
Some general advice could be presented to the reader as follows:
SOS Tip 2 — Planning
1. Notice and understand the main merits and disadvantages of each side.
2. Notice the immediate threats of both sides but especially of your opponent!
3. Determine what is the most significant target(s) and how the plan should be executed.
4. Examine what will be the opponent’s reaction to your plan.
5. Examine what your opponent’s possible main plan is (or can be) and how you should react to it.
6. If you can choose among two or more good plans opt for the one that you think can bring the
most benefits.
7. Follow your plan. Do not change it without a good reason.


The ‘Lucena Position’ is one of the most famous and important positions in endgame theory where
one side has a rook and a pawn, and the defender has a rook.
It is fundamental in rook and pawn vs rook endgames that if the side with the pawn can reach this
type of position, he can forcibly win the game. Most rook and pawn versus rook endgames reach
either the ‘Lucena Position’ or the ‘Philidor Position’ if played accurately.
The side with the pawn will try to reach the ‘Lucena Position’ to win; the other side will try to reach
the ‘Philidor Position’ to draw.
The ‘Lucena Position’ is named after the Spaniard Luis Ramírez de Lucena although it is something
of a misnomer because the position does not in fact appear in his book on chess: ‘Repetición de
Amores e Arte de Axedrez’ (1497).
It does appear, however, in Alessandro Salvio’s ‘Il Puttino’ (1634), a romance on the career of the
chess player Leonardo da Cutri, and it is in that form that it is given here.
Salvio attributes it to Scipione Genovino. It is likely that the error arose from the sixth edition of the
‘Handbuch des Schachspiels’, in which editor Constantin Schwede incorrectly attributed the position
to ‘Lucena 96’, possibly as a result of confusion over the references in Antonius van der Linde’s 1874
work ‘Das Schachspiel des XVI. Jahrhunderts’.
The essential characteristics of the ‘Lucena Position’ are:
SOS Tip 1 — Characteristics
1. The pawn is any pawn except a rook pawn.
2. The pawn has advanced to the 7th rank.
3. The stronger side’s king (the one with the pawn) is on the queening square of his pawn.
4. The stronger side’s rook cuts-off the opposing king from the pawn by at least one file.
5. The defending rook is on the file on the other side of the pawn.
6. The winning method: building a bridge.
Like in almost all endings, the rook pawns (a and h) are the least desirable for the attacker in rook
endings as well.
Even in the favorable case that a ‘Lucena Position’ has been reached, the defending king must be cutoff for at least three files.
Thus, if White has an a-pawn and the black rook controls the b-file, the black king must not be nearer
than the f-file for White to win.
Hence, we can conclude:


SOS Tip 2 — Conclusions
1. If the pawn is on the 7th rank, multiple winning methods exist. The most important ones are
building a bridge for protection from checks along files and a rook maneuver for protection from
side checks along ranks.
2. When the king of the weaker side is cut-off from the pawn the only defensive technique
consists of side checks.
3. A rook pursuit of the enemy king can only be successful when the rook and the pawn are
separated by at least three lines.
4. A central or a bishop pawn divides the chessboard into two unequal parts: one is ‘long’, another
is ‘short’. The correct positioning of forces for the weaker side is to keep the king on the short
side and the rook on the long side.
The ‘Lucena Position’ is heaven for the side with the extra pawn: it is the desired outcome of every
rook ending.
In most cases it an easy concept to play but there are certain cases where knowledge and accuracy are
Example 1 ■

Our first example is simple but illuminating. White’s aim is to either promote his pawn or else
compel Black to give up his rook for it — either result will leave White with an overwhelming
material advantage and a straightforward win. White has managed to advance his pawn to the 7th
rank, but it is prevented from queening because his own king is in the way. White would like to move
his king and then promote his pawn but is prevented from moving to the h-file by the black rook and
prevented from moving to the f-file by the black king.


The black rook must stay on the h-file. After 1...Ra1 2.Rh3! and 3.Kh7, White wins.
1...Rh2 2.Rf4!
The first step of the basic winning method called the ‘bridge’.
2...Rh1 3.Re4+! Kd7
Also hopeless is 3...Kf6 4.Kf8!.
4.Kf7 Rf1+ 5.Kg6 Rg1+ 6.Kf6

There are no defensive resource in waiting.
a) After 6...Rg2 7.Re5! and 8.Rg5, White completes the ‘bridge’.
b) Likewise, after 6...Kd6 7.Rd4+! Kc6 (7...Kc7 8.Rd5! and 9.Rg5) 8.Rd8!, White wins.
7.Kg5! Rg1+ 8.Rg4!
The ‘bridge’ is complete, and the white pawn will promote.
The second example shows the building of the ‘Lucena Position’:

Example 2 □

If Black is to move, he would be able to achieve a draw with 1...Rf8!. White has no satisfactory
continuation as both 2.Rxf8 Kxf8 and 2.Rf5 Rxf5! 3.Kxf5 Kf7 lead to a drawn pawn ending. If White
retains the rooks with 2.Ra1 Kf7! Black would succeed in bringing about the ‘Philidor Position’ and a
draw. Now White is ready to advance his g-pawn. Since passive defense does not bring any results
Black has nothing better than to start checking.
1.Kh5! Rh8+ 2.Kg6 Rg8+ 3.Kh6 Rh8+ 4.Kg7 Rh2 5.g6 Rg2
Black is trying to avoid the ‘Lucena Position’ (king on g8, pawn on g7) but all he can achieve is to
slightly delay its appearance.
6.Kh7! Rh2+ 7.Kg8! Rg2 8.g7


White has reached the ‘Lucena Position’ and wins as described in the previous example.
Like in almost all endings the rook pawns (a and h) are the least desirable for the attacker in rook
endings as well.
Even in the favorable case that a ‘Lucena Position’ has been reached the defending king must be cutoff for at least three files.
Thus, if White has an a-pawn and the black rook controls the b-file the black king must be not nearer
than on the f-file in order for White to win.
Example 3 □


Since the black king is already on the e-file the position is drawn regardless of which side is to move.
The outlines of the following play are as follows: In order to win White must free his king, starting
with Rh1–h8–b8. Black cannot prevent this rook maneuver but his king can arrive in time to prevent
the escape of the white king. Assuming that White is to move a possible continuation is:
1.Rh1 Kd7! 2.Rh8 Kc7!
The only move. The defending king must be as close as possible to the pawn.
3.Rb8 Rh2 4.Rb7+ Kc8!


Keeping the white king trapped in the corner.
5.Rb1 Rc2!
By securing a place on the c-file for his king Black also secures the draw. There is no way for White
to improve his position as he cannot free his king. 5...Kc7 6.Rc1+ Kb6 7.Rc8 Rh7, is good as well.
Efstratios Grivas
Gilles Miralles
Bucharest 1984 □

Note that if Black was to move he could continue with 55...Rf8+ 56.Kg6 Rg8+ with a draw.
or 55.Re4 also win.
55.g5! Rf8+ 56.Kg6
56.Kg4, also wins: 56...Rg8 57.Kh5 Rh8+ 58.Kg6 Rg8+ 59.Kh6 and so on.
56...Rg8+ 57.Kf6 Rf8+ 58.Kg7 Rf1 59.g6 Rg1 60.Kf7
The usual way to push the pawn.

60...Rf1+ 61.Kg8 Rg1 62.g7 Rh1
62...Rf1 63.Rh2 Rf3 64.Kh8+–

The ‘Lucena Position’ is reached. Now it is time for the ‘Bridge Method’ — the king goes to g5 and
the rook to g4.
63.Re4! Rh2 64.Kf7 Rf2+ 65.Kg6 Rg2+ 66.Kf6 Kd7
Or 66...Rf2+ 67.Kg5 Rg2+ 68.Rg4.


And as 68.Rg5 is coming, Black resigned.
Example 4 □

It is worth mentioning that White has other winning options: 1.Re1,
or 1.Rg1+ Kh7 2.Rd1! Kg7 3.Kd7 Ra7+ 4.Ke6 Ra6+ 5.Rd6 Ra8 6.Rd8 are also good.
1.Rg1+ Kh7 2.Rg4!
2.Kf7 is premature in view of 2...Rf2+ 3.Ke6 Re2+ 4.Kf6 Rf2+ and the king has only one way to take
refuge from the rook checks: by returning to e8. The rook move prepares an interference at e4. As
already noted, this method is called ‘building a bridge’ or simply ‘bridge’.
2...Ra8+ 3.Kf7.
3.Kf7 Rf2+ 4.Ke6 Re2+ 5.Kf6 Rf2+
If 5...Re1, then 6.Rg5 and 7.Re5.
6.Ke5 Re2+ 7.Re4

Example 5 ■

Now let us see what happens if Black is on move. Let us shift all the pieces except for the black rook
a single file to left. Then the side checks do not help anymore because the rook is not remote enough
from the white pawn: 1...Ra8+ 2.Kc7 Ra7+ 3.Kc8 Ra8+ 4.Kb7+–.
1...Ra8+ 2.Kd7 Ra7+ 3.Kd6 Ra6+ 4.Kc7
4.Kc5 Re6.
Study 1
Kopaev Nikolay 1953 □


The unlucky placement of the king kills Black. With the king on h7 it would have been a draw. In
addition, the black rook is too close to the f-pawn. But it is by no means easy for White to exploit
these disadvantages.
1.Kf6! Rc6+ 2.Ke5 Rc8
If 2...Rc5+ then 3.Kd6 Rc8 4.Re1! Kg7 5.Re8+–. With a rook on b8 the saving check 5...Rb6+ exists.
3.Rg6!! Kh7 4.Rc6! Ra8 5.Kf6
The rook protects the king from side checks. Black is helpless against the maneuver Re6–e8.
Now for some practical examples:
Gyula Sax
Vitaly Tseshkovsky
Zagreb 1975 □


White’s main concern here is to reach the ‘Lucena Position’ as soon as it is possible and maybe the
black king’s position is not helping his plans. But by accurate play White can reach his goal.
A blunder. White should not move the rook away from the d-file where it protects the king from side
a) An easy win was 90.f7! Rc8 (90...Kg7 91.Rg3++–; or 90...Re1+ 91.Kf6 Rf1+ 92.Ke7 Re1+
93.Kf8 Ra1 94.Rh3+ Kg6 95.Kg8+–) 91.Ke7 Rc7+ 92.Rd7+–.
b) Another win was 90.Ke7! Rc7+ (90...Re1+ 91.Kf8+–) 91.Rd7 Rc1 92.f7 Re1+ 93.Kf8+–.
90.Rh3+? Kg6!
90...Kg8? loses to 91.f7+ Kg7 (91...Kf8 92.Rh8++–) 92.Rg3+.


Here something unexpected happened as Black failed to recognize the fact that the position had
become drawn and resigned. Indeed after 91...Kh7 92.f7 Rc8! there’s no win for White. (92...Rc6+?
93.Kd7+–) 93.Ke7 (93.Rd3 Kg7=) 93...Rc7+ 94.Ke8 Rc8+ 95.Kd7 Ra8=. Such an ‘unjustified’
resignation in not common among GMs but strange things can happen in the heat of the battle and
after so many hours of fighting.
Hristos Bousios
Efstratios Grivas
Athens 1987 ■


Cutting-off the white king while preparing a king escape on the e-file.
Note that this isn’t the only win but it is the most practical.
58.Kf2 d3 59.Kf1 Kd1 60.Kf2 d2
Black has reached the ‘Lucena Position’ and White resigned due to 61.Rc6 Rf5+ 62.Kg3 (62.Ke3
Ke1–+) 62...Ke2 63.Re6+ Kd3 64.Rd6+ Ke3 65.Re6+ Kd4 66.Rd6+ Rd5 — the ‘Bridge’.
Efstratios Grivas
Alexander Naumann
Corfu 1999 ■


Black had to find 88...Rc6! 89.Rxf7 Kg5! which draws: 90.Rf8 Ra6 91.f7 Rg6+ 92.Kh7 Rh6+.
88...Rh1? 89.Ra5+ Ke6 90.Ra6+ Kf5
90...Kd7 91.Kxf7 Rf1 92.Kf8, wins as well.
And now White wins as the ‘Lucena Position’ is in the cards.
91...Rh7+ 92.Kg8 Rh6 93.Kg7 Rg6+ 94.Kf7 Kg5


95.Ke7 Kh6 96.Re6
Ulf Andersson
Ralf Akesson
Skelleftea 1999 □


Going for the known position!
79...dxe4 80.Rxe4 Kd7 81.Kg6
Black resigned: 81...Rg1+ 82.Kf7 Rf1 83.f6 Rf2 84.Kg7 Rg2+ 85.Kf8 Rf2 86.f7 (‘Lucena Position’)
86...Rg2 87.Rd4+ Kc7 (87...Ke6 88.Ke8+–) 88.Ke7 Re2+ 89.Kf6 Rf2+ 90.Ke6 Re2+ 91.Kf5 Rf2+
92.Rf4 (Bridge).
The ‘Lucena Position’ is heaven for the side with the extra pawn it is the desired outcome of every
rook ending.
In most cases it an easy to play concept but there are certain cases where knowledge and accuracy are


The ‘Philidor Position’ (or Philidor’s position) usually refers to an important endgame which
illustrates a drawing technique when the defender has a king and rook versus a king, rook and a pawn.
It is also known as the 3rd rank defense because of the importance of the rook on the 3rd rank cuttingoff the opposing king.
It was analyzed by François-André Danican Philidor in 1777. Many rook and pawn versus rook
endgames reach either the ‘Philidor Position’ or the ‘Lucena Position’.
If played accurately the defending side tries to reach the ‘Philidor Position’; the other side tries to
reach the winning ‘Lucena Position’.
From the defender’s point of view the important characteristics of the position are:
SOS Tip 1 — Characteristics
1. The defending king is on the queening square of the pawn (or adjacent to it). The pawn can be
on any file.
2. The opposing pawn has not yet reached the defender’s 3rd rank (its 6th rank).
3. The opposing king is beyond the defender’s 3rd rank.
4. The defender’s rook is on the 3rd rank keeping the opposing king off that rank.
The general ideas are easy to understand and practice. With some good examples, the ‘Philidor
Position’ can be nicely ‘absorbed’. Naturally things are usually more complicated during the real
battle and study is required.
Breaking down the ‘Philidor Position’ is an obligation for the stronger side. But this is easier said
than done as usually happens in this blunderful life!
Example 1 ■


This is the so-called ‘Philidor Position’. The famous French chess player was the first to demonstrate,
as early as the 18th century, the correct method of defense. White would like to get his king to the e6square and threaten mate to force the black king away from the queening square of the pawn, e8. The
black rook on the 3rd rank prevents that. If White checks with the rook from the side Black simply
keeps the king in front of the pawn by alternating between squares e8 and e7. If White offers an
exchange of rooks Black should take it since the resulting king and pawn endgame is drawn therefore
White’s only winning chance is to advance the pawn. The basic idea is for the defender to keep his
rook on his 3rd rank until the pawn advances to that rank and then check the opposing king from
Preventing a penetration of the white king to the 6th rank.
2.e6 Rb1
If the pawn stood at e5 the white king would have had a refuge from vertical checks. However, as
soon as the pawn steps forward the refuge disappears thus Black draws easily.
Example 2 □


Philidor thought the initial position was winning for White until Black found a new defensive idea.
With the following move White poses Black more difficulties.
Black then found the following defensive resource.
Later the second defensive method in the ‘Philidor Position’ was discovered: an attack from the rear
that helps Black to hold as well. If the rook fails to occupy the 6th rank ‘a la Philidor’, it must be
placed behind the white pawn. And his explanation was 1...Rf1+ 2.Ke6 Kf8 3.Ra8+ Kg7 4.Ke7 Rb1
5.e6 Rb7+ 6.Kd6 Rb6+ 7.Kd7 Rb7+ 8.Kc6+–.
2.Ke6 Kf8!
The text move is undoubtedly correct as the king goes to the short side leaving the long side for the
rook but 2...Kd8?! 3.Ra8+ Kc7 does not lose: 4.Re8 (4.Kf6 Kd7=) 4...Rh1! (rather than 4...Re2?
5.Kf7 Rh2 6.Rg8! Rh7+ 7.Rg7 Rh8 8.Ke7 Kc6 9.e6 Kc7 10.Rg1+–) 5.Rg8 Re1! 6.Rg2 Kd8!=.
3.Ra8+ Kg7
Now we can evaluate the position of the black rook. First, it prevents both 4.Ke7 and 4.Kd7.
Secondly, Black can meet 4.Kd6, in an efficient way.


If White tries 4.Re8 preparing 5.Kd7 the black rook occupies the long side: 4...Ra1!=.
And White must retrace his steps.
5.Ra7+ Ke8 6.Ke6 Kf8! ½–½
Example 3 ■


Obviously, such a defense with the king on the long side would have been impossible if the short side
were even shorter with an f or g pawn. But this position is also drawn. Black’s rook comes in time for
a passive defense along the 8th rank:
In this position the attack from the rear does not work anymore: 1...Rg1? 2.Ra6 Kf8 (2...Rf1 3.Ra8+
Rf8 4.Rxf8+ Kxf8 5.Kh7+–) 3.Ra8+ Ke7 4.Rg8! (White prepares 5.Kh7! as the black rook will be
unable to disturb the king from the side.) 4...Rg2 5.Kh7! Kf7 6.g6+ Ke7 7.Ra8 Rh2+ 8.Kg8 Rg2
1...Ra1! 2.Rb6 Ra8 ½–½
Example 4 ■


Black is obviously unable to set-up a ‘Philidor Position’ but he can achieve the draw by activating his
1...Ra1! 2.Ke6 Rd1!
The rook must immediately place itself behind the pawn if the pawn is to be prevented from
advancing. 2...Re1+? 3.Kd6 Kc8 4.Rg8+ Kb7 5.Rg7+! Kc8 6.Rc7+ Kd8 7.Rh7 loses as White will be
able to advance his pawn to the 6th rank keeping his king safely covered in front of it. Thereafter, he
will set-up the ‘Lucena Position’.
3.Kd6 Kc8!
Moving to the short side. Although for this exact case of a central pawn still on its 5th rank the king
can also head for the longer side but it is always simpler and better to move it to the short one abiding
by the rule.
4.Rg8+ Kb7 5.Rd8
Threatening to win with 6.Ke7 and 7.d6.
The rook positions itself for side checks.
6.Re8! Rd1! 7.Re5 Kc8!
The threatened 8.Kd7 must be prevented. Black is now safe (8.Ke7 Kc7!) and draws.


Example 5 ■

If White was to move, he would win by means of 1.Re8! Ra1 (1...Rf2 2.Re6 Rf1 3.Kf8 Kg6 4.f7+!)
2.Kf8 Kg6 3.f7 Kf6 4.Rc8 Ra2 5.Rc6+.
1...Ra1! 2.Re8 Ra7+ 3.Re7 Ra8! 4.Ke6+
Of no help is 4.Rb7 Kh6! 5.Ke7 Kg6!.
4...Kg6 5.Rg7+ Kh6 6.Rg1 Ra6+! ½–½
With the above examples the ‘Philidor Position’ is easy to absorb. But of course, things are usually
more complicated during the real battle and we should examine them.
Eric Lobron
Rainer Knaak
Baden-Baden 1992 ■


To avoid the theoretical draw Black should have played 104...Kd3! as the white king is placed at the
long side and one cannot see how White can survive. For example: 105.Rd5+ Kxe3 106.Kc2 Ke2
[106...Rxg3 107.Re5 Kf4 108.Rd5 (108.Re8 Rd3!–+) 108...Kf3 109.Re5 e3 110.Kd3 Kf2 111.Re8
Rf3 112.Re7 Rf8 113.Re6 Rd8+–+]
107.Rd2+ Kf3 108.Rd7 Rf8 (108...Rxg3 109.Re7 e3 110.Kd3 Kf2–+) 109.Kd1 Kf2! 110.Rd2+ Kf1–
104...Kxe3? 105.Kc2 Rxg3 106.Re8 Rg2+
After 106...Rg2+ 107.Kd1 the position is drawn according to the second defensive method in the
Philidor Position.
Alexey Dreev
Alexander Beliavsky
Odessa 1989 □


White is in a precarious situation as 55...Kd3 is threatened thus he tries his last chance.
55.h4! Rxh4?
Black should have given a rook check and moved his pawn to g4. Later on, he could either trade
kingside pawns under more favorable circumstances than actually happened or move his king to the
g-pawn. The eventual consequences of 55...Rg1+! were:

[56.Kd2 Rg2+ 57.Ke1 g4 58.Kf1 (58.h5 Kc3 59.h6 Rh2 60.Rh8 d3 61.Rc8+ Kd4 62.Rd8+ Ke3
63.Re8+ Kf3–+) 58...Rh2 59.Rxg4 Kc3! 60.Kg1 Rc2! 61.Rg8 (61.h5 d3–+) 61...d3 62.Rc8+ Kb2

63.Rd8 d2–+]
56...d3+ 57.Kd2 (57.Kf2 d2–+) 57...Rg2+ 58.Kd1 g4 59.h5 Rh2 60.Rc8+ (60.Rxg4+ Kc3–+)
60...Kd4 61.Rd8+ Ke3 62.Re8+ Kf3–+.
56.Rxg5 Kc3

The only move to reach the desired draw.
57...Rh1+ 58.Ke2 Rh2+ 59.Kd1 Kd3 60.Kc1! Rh1+ 61.Kb2 Re1 62.Rd8 Re4 63.Kc1 Ke2 64.Kc2
Even great legends have difficulties as the next example shows.
The heat of the battle is always too high.
Bent Larsen
Mihail Tal
Bled 1965 □


The queenside pawns will inevitably be traded and the ‘Philidor Position’ will probably occur
51.Ra7+ Kc8?!
The black king goes the wrong way. He should have tried for the short side. After 51...Ke8! 52.Ke6
Kf8 53.Ra8+ Kg7 54.Kxe5 b3 55.axb3 Rxb3 the draw is obvious.
52.Kxe5 b3?
a) As K. Mueller indicates, after 52...Rh2! the position is still drawn: 53.Ke6 Rh6+ 54.Kd5 Kd8
55.e5 (55.Kc5 Rh2 56.e5 Re2 57.Kd6 Ke8 58.Ke6 Kf8=) 55...Ke8 56.e6 Rh2=.
b) Note that here 52...Kd8, also draws: 53.Ke6 Rh2!, but not by 53...b3? 54.Rd7+! (54.axb3 Rxb3
55.Ra8++–) 54...Kc8 (54...Ke8 55.Rb7+–) 55.axb3 Rxb3 56.e5+–.
53.axb3 Rxb3 54.Kd6! Rd3+


B. Larsen misses his chance to punish his opponent for the grave error and allows him to employ the
second defensive method in the Philidor Position. The winning continuation was 55.Ke7! Rh3 56.Ra4
Rh7+ 57.Ke8 (57.Kf6 Rh6+ 58.Kg7 Rh4 59.Rd4!+–) 57...Rh8+ 58.Kf7+–.
55...Rh3 56.Ra8+
56.Ra4 Kd8!=
56...Kc7 57.Rf8 Re3 58.e5 Re1 59.Re8
59.Kf6 Kd7!=
59...Rh1! 60.Ra8 Re1!


The position is drawn but White played 18 more moves before he agreed to the peaceful outcome of
the game as the result was vitally important for both rivals.
61.Ra7+ Kd8 62.Ra2 Re3 63.Ra8+ Kc7 64.Ra6 Kd8! 65.Ra8+ Kc7 66.Re8 Rh3! 67.Rf8 Re3!
68.Rg8 Re1 69.Rg2 Kd8! 70.Ra2 Re3 71.Ra5 Re1 72.Rb5 Re2 73.Kf7 Rf2+! 74.Ke6 Re2 75.Rb8+
Kc7 76.Re8 Rh2! 77.Kf7 Rh7+ 78.Kg6 Kd7 ½–½
Milan Vukic
Klaus Uwe Mueller
Varna 1975 ■


A losing decision. The black f-pawn is lost anyway but now the black king will forever remain on the
long side where it only obstructs his own rook. Black could hold by means of transferring his king to
the short side: 65...Kf8! 66.Rb6 (66.Kxf6 Ra6+=; or 66.Rb8+ Ke7=) 66...Rf4! (66...Rg4+ 67.Kxf6
Kg8 68.Rb8+ Kh7

is less accurate but still playable. If we shift this position by one file to the left White should have
won: 67.Kxe6 Kf8 68.Ra8+ Kg7 69.Ke7.
Here, however, Black holds since he has two files (‘a’ and ‘b’) for his rook to stay far away from the

white king. The white rook occupies one of them but Black can use the remaining one: 69.Ke6 Re4+
70.Kf7 Ra4=)
67.Kxf6 (67.Rxf6+ Kg8 68.Ra6 Rg4+=) 67...Kg8!

68.Rb8+ Kh7 69.Ke6 (69.Rf8 Ra4=) 69...Kg7=.
65...Ra6? 66.Kg7! Rc6 67.Rb8+ Ke7 68.Rb1 Ra6 69.Re1+ Kd8 70.Kf7 Kd7 71.Rd1+ Kc7 72.Ke7

Black resigned due to 72...Rb6 73.Rd7+ Kc8 74.Rd6.

Michael Rohde
Dan Cramling
Innsbruck 1977 □

70.Kf6? Re1! 71.Ke6 Kf8! 72.Rh8+ Kg7 73.Ra8 Re2 74.Kd6 (74.Re8 Ra2=) 74...Kf7!=
70.Ke6! Kf8
70...Kd8 71.Rh8+ Kc7 72.Ke7+–
An important check to impair Black’s king position. Both 71.Ra7? Re1!= and 71.Rh8+? Kg7 72.Ra8
Re1= are bad.
71...Ke8 72.Ra7 Kf8 (72...Kd8 73.Ra8+ Kc7 74.Ke7+–) 73.Ra8+ Kg7 74.Ke7 Rb1


75.e6 is hopeless as well for Black. The white rook is ideally placed on a8. This is the position White
wants in this ending and he can achieve it — provided that he plays correctly.

a) Wrong would be 72.Ra7? Re1! (72...Kf8? 73.Ra8+ Kg7 74.Ke7+–) 73.Kf6 (73.Kd6 Kf8!=)
73...Rf1+ 74.Ke6 (unfortunately there is no 74.Ke7? Rf7+–+, therefore the white rook should be
placed near the king) 74...Re1=.
b) The move 72.Rc7, is less precise but it does not miss the win after 72...Re1 73.Kf6 Rf1+ 74.Ke7
Ra1 75.Rd7 Ra2 76.Rd4 (or 76.Rd3) 76...Ra7+ 77.Kf6+–.

72...Re1 73.Kf6 Rf1+ 74.Ke7 Ra1

A bad mistake leading to draw.
White could have crowned his efforts with 75.Rd2! (or 75.Rd4) 75...Ra7+ 76.Kf6 Rf7+ 77.Ke6 Rf1.

Unfortunately for Black his rook must leave the long side because 77...Ra7 78.Rd8+ Kg7 79.Rd7+ is
bad. Now White immediately occupies the a-file. 78.Ra2!
[78.Ke7?! Rf7+ 79.Kd8 Ra7 80.e6? (80.Ke8+–) 80...Kf8! 81.Rf2+ Kg7=]

78...Kg7 (78...Re1 79.Kf6! Rf1+ 80.Ke7 Rf7+ 81.Kd6+–; 78...Kf8 79.Ra8+ Kg7 80.Kd6+–)
[79.Ke7? Rf7+ 80.Kd6 Rb7! 81.e6 (81.Ra6 Kf7=) 81...Rb6+ 82.Kd7 Rb7+ 83.Kc6 Rb1 84.Rf2 Ra1=]
79...Kg6 (79...Kf8 80.Ra8+ Kg7 81.Kd6+–) 80.Ra8 Kg7 81.Ke7 Rf7+ (81...Rb1 82.e6 Rb7+
83.Kd6+–) 82.Kd6 Rb7 83.e6 Rb6+ 84.Kd7 Rb7+ 85.Kc6 Re7 86.Kd6+–.
75...Kg7! 76.Rd6 Ra8! 77.Rd1
Black has achieved a perfect ‘Philidor Position’ and there is nothing to be expected.
I have found the exact same position between two 8-year old boys played in the FIDE World
Championship U8.
Well, the two youngsters played more or less fine for their age baring one critical mistake for each
Obviously, they had worked on it and we cannot really blame them for their mistakes, even strong
GMs make the same mistakes!
Haince Patrick de Leon
Agibileg Uurtsaikh
Vung Tau 2008 □

54.Ke6! Kf8 55.Rh8+?


Of course 55.Rf7+!+–.
55...Kg7 56.Rc8 Rb1 57.Kd7 Rd1+ 58.Ke7 Ra1! 59.Rd8 Kg6 60.Rd6+

Too active!
Black should opt for 60...Kg7!=.
61.e6 Ra7+ 62.Rd7 Ra6 63.Rd6 Ra7+ 64.Ke8 Ra8+ 65.Kf7 Ke5 66.Rd7 Ra6 67.e7 Rf6+ 68.Kg7
Re6 69.Kf7
69.Rb7 Kd6 70.Kf7+–
69...Rf6+ 70.Ke8 Ra6 71.Rd8 Rh6 72.Kd7 Rh7 73.Ra8 Kf6 74.Ra6+ Kf5 75.Kd8 Rh8+ 76.e8=Q
Lev Alburt
Maxim Dlugy
Los Angeles 1991 □


Although this is a drawn position, in a practical game Black is suffering. This is due to the shorter
time controls that are used today which demands any player to be familiar with any ‘simple’
On 61.Kd6 Kf5, draws but not 61...Kg6? 62.Kc5++–.
61.Ke6 Kg6?
A very bad mistake.
61...Rb1 (or 61...Rb7) 62.Ra7 Rb6+ 63.Kd5 Rb5+ 64.Kd6 Rb6+ 65.Kc5 Rb1, draws.
62.Ke7+! Kg7
62...Kf5 63.e6 Rb7+ 64.Kf8+–


This is now a won position.
It makes no difference if the white e-pawn is on e5 or e6.
63...Rb8 64.e6 Kg6 65.Ra1 Rb7+ 66.Kd6
66.Kd8 Kf6 67.e7 Rb8+ 68.Kc7 Re8 69.Kd6 wins as well.
66...Rb6+ 67.Kd7 Rb7+ 68.Kc6 Rb8 69.Kc7 Rh8 70.e7 1–0
As noted above, even very strong players can mishandle the position. Next, we will examine two
additional cases:
Levon Aronian
Magnus Carlsen
Moscow 2006 ■


The only drawing move here is 73...Kg6! (73...Ra1? 74.Ke8+–)
73...Kg6! 74.Rd7 (74.Kd7 Kf6= 74.Rd8 Ra7+ 75.Rd7 Ra8 76.Rd6 Kg7!=) 74...Kg7 75.Kd6+ Kf6
76.e7 Kf7=.
73...Ra7+? 74.Ke8
Black resigned because of 74...Ra8+ 75.Rd8 Ra6 76.e7 and with a pawn on the 7th rank White wins.
Ivan Sokolov
Hristos Banikas
Khanty-Mansiysk 2010 ■


Of course, this is a theoretical draw as the pawn is not too advanced, Black’s king is close enough and
his rook has a long side to operate on.
77...Kg7 78.Ke6 Rf6+ 79.Ke7 Ra6 80.Rd8 Ra5 81.Rd5 Ra7+ 82.Rd7 Ra5 83.Ke6+ Kg6 84.e5 Ra6+
85.Rd6 Ra8 86.Ke7+ Kg7 87.e6

This is one of the key positions in this kind of endgame. Black must be careful: he must not move his
rook from the 8th rank and the a-file.
87...Ra1? 88.Ke8! Ra8+ 89.Rd8 Ra6 90.e7 Ra7 91.Rc8 Ra1 92.Rc2, or 87...Rb8 88.Rd8 Rb7+ 89.Kd6
Rb6+ 90.Kd7 Rb7+ 91.Kc6.


Conclusion: as Black’s rook is on its best square (a8) he has only one move left.
87...Kg6! 88.Rd8 Ra7+ 89.Rd7 Ra8 90.Rb7 Kg7 91.Rc7 Kg6 92.Rc1 Ra7+ 93.Kd6 Ra6+ 94.Kd7
Ra7+ 95.Rc7 Ra8 96.Ke7 Kg7 97.Kd7 Kf8
97...Kf6 was also good enough.

White can’t make progress as Black defended accurately.
But here comes the blunder.
98...Kg7 was correct securing the half-point.
And Black resigned as he realized what he had done. His king will be kicked away from the
important promotion square: 99...Ra8 100.e7+ Kf7 101.Rf1+.
This is a method used to shelter the stronger side’s king from the opponent’s rook checks.
It can be applied when the stronger side can sacrifice a pawn.

Petr Velicka
Tomas Polak
Czech Republic 1995 ■

62...f4! 63.gxf4 Rb2+ 64.Kf1 Kf3
White’s own f4-pawn prevents him from saving himself with a check on f5.
65.Ra1 Rh2 66.Kg1 Rg2+ 67.Kh1


Or 67...e2 68.Re1 Rg4–+.
68.Kh2 e2 69.f5 Kf2 70.Kh3 Rg5!
Alexander Zaitsev
Robert Huebner
Buesum 1969 ■

a) The waiting approach with 57...Ra5 was good. After 58.g5 (58.f5 Ra1!=) 58...fxg5+ 59.fxg5 Black
could defend the position either in the Philidor method 59...Ra6 or passively with 59...Ra8.
b) A draw could have been achieved as well by 57...Rb4 58.f5 Rb1! 59.Kh5 (59.g5 Rh1+=) 59...Rg1!.
Now White can implement the umbrella:
57...Rb1? 58.Kh5 Rg1
Otherwise 59.Kg6.
59.g5! fxg5 60.f5! Kf8 61.f6
Black resigned: 61...Rf1 62.Kg6 Ke8 63.Ra7 g4 64.f7+ Kf8 65.Ra8+ Ke7 66.Re8+ Kd6 67.f8=Q+

Rxf8 68.Rxf8 g3 69.Rf3+–.


The ‘Vancura Position’ is a drawing position with a rook and rook’s pawn versus a rook when the
pawn is not beyond its 6th rank and the stronger side’s rook is in front of the pawn.
It was studied by Josef Vančura (1898–1921) and published in 1924. The defending rook keeps
attacking the pawn from the side from some distance away while preventing the stronger side’s king
from finding cover from checks by the rook.
The defending king must be on the opposite side of his rook so as to not block any checks. The
defending rook moves behind the pawn as soon as the pawn moves up to its 7th rank.
Also, the defending king must be near the corner on the opposite side of the board if the pawn
advances to its 7th rank so the stronger side’s rook cannot check the defending king and then support
the advance of the pawn or sacrifice the pawn to skewer the defending king and rook on the 7th rank.
Let’s examine this quite interesting case step by step.
Example 1 □

Protecting the pawn in order to free the rook to move. If 1.a7 Ra6! 2.Kb5 Ra1 3.Kb6 Rb1+ 4.Kc7
Rc1+ 5.Kd7 Ra1 and White cannot win.
Note that if Black’s king was on g6 there would follow 2.Rg8+ Kh7 3.a8=Q,
and if it was on f7 White would win with 2.Rh8! Rxa7 3.Rh7+.
1.Kb5 Rf5+!


The only move! An alternative such as 1...Rf1?, loses to (among others) 2.Rc8.
2.Kc6 Rf6+!
An important square for the rook. Black now checks on the f-file and aims to maintain a sideways
attack on the pawn.
3.Kd5 Rf5+
Alternatively, rook anywhere on the 6th rank is good as well: 3...Rb6, or 3...Rg6, or 3...Rh6, as the
white king is far away from his pawn.
4.Ke6 Rf6+ 5.Ke5 Rb6

Maintaining sideways contact with the pawn.
6.Kd5 Rf6 7.Kd4 Rb6
The usual 6th rank defense.
Note that here 7...Rf4+? loses to 8.Ke5 Rf6 9.Rg8+ Kxg8 10.Kxf6+–.
8.Kc5 Rf6 9.Ra7+
9.a7 Ra6=
9...Kg6 10.Ra8 Kg7
And White cannot win. The white king cannot advance because of the checks and the pawn cannot

advance because the black rook gets behind it.
Study 1
Mueller Karsten 2014 □

The drawn ‘Vancura Position’ is based on the passivity of White’s rook in front of the passed a-pawn.
Black’s rook ties it to the defense of the pawn and Black’s king stays in the drawing zone g7 and h7:
1.Ke5 Rb6
The rook forces White’s rook to defend the pawn.
Or 2.Ra7+ Kg8 3.Kd5 Rf6 4.Kc5 Rf5+ 5.Kd6 Rf6+ 6.Ke5 Rb6 7.Ra8+ Kg7=.
2...Kf7? loses in a typical way: 3.a7 Ra6 4.Rh8 Rxa7 5.Rh7++–. This is the reason why the drawing
zone for the black king is g7 and h7 or g8 and h8 if White’s rook is on the 7th rank.
3.Kc5 Rf5+
It is important that the rook keeps the checking options on the f-file open. 3...Rh6?, loses due to
4.Kb5 Rh5+ 5.Kb6 Rh6+ 6.Kb7 and the checks have run out and White wins as the rook can be
activated after 6...Rf6 7.Rc8+–.


4.Kb6 Rf6+
White’s king finds no shelter on the queenside.
5.Ka7 Rf7+ 6.Kb8 Rf8+ 7.Kb7 Rf7+ 8.Kc6 Rf6+ 9.Kd5

9...Rf5+, draws as well, but after 10.Ke4 Ra5? (10...Rf6= reaches the standard drawing set-up) would
be a mistake due to 11.Kd4 Kf6 12.Kc4 Ra1 (12...Ke6 13.a7+–) 13.Kb5 Rb1+ 14.Kc6 Ra1 15.Kb7
Rb1+ 16.Ka7 Ke7 17.Rb8+–.
Now Black’s rook can move behind the passed pawn as White’s king has no more shelter.
10...Ra6 11.Kc5 Ra1 12.Kb6 Rb1+ 13.Kc7 Rc1+ 14.Kd6 Rd1+ 15.Kc5 Rc1+ 16.Kb4 Rb1+ 17.Kc3
Ra1 18.Kb2 Ra6
A long time ago, in ‘Shakhmaty v SSSR’ in 1950, Peter (Piotr) Romanovsky published the Drawing
Drawing Zone ■


If Black is to move and the white king is on one of the ‘marked’ squares, he draws by reaching the
‘Vancura Position’. Otherwise White wins.
Garry Kasparov
Vassily Ivanchuk
Riga 1995 ■

Black has nothing to fear of here.
49...Rb6 50.Kf5 Rc6 51.Ke5 Rb6 52.Kd5 Rf6 53.Kc4 Rf4+ ½–½


Levon Aronian
Vassily Ivanchuk
Moscow 2009 □

A drawish position.
After 52.h7, Black moves his king to squares a7 & b7, and when the white king is ready to approach
his h-pawn he starts checking from behind.
52.Kf4 Rh2 53.Ke5
A last try...
53...Rxb2! 54.Kd4 Rb6!


Reaching the ‘Vancura Position’.
55.Kxc4 Rg6 56.Kd5 Rf6 57.Ke5 Rc6 58.Rh7+ Kb8
Of course, things are not always so ‘easy’, as we observed in the starting examples.
The ‘Vancura Position’ can be complicated as well. One should know how to play with it but also
how to reach or not reach it if you are on the stronger side.
The chess player is obliged to make quick decisions which are obviously based on sub-conscious
knowledge as there is not enough time in the late stages of the game for long periods of thinking.
Teimour Radjabov
Hikaru Nakamura
Shamkir 2014 ■


The only move! The black king moves to set up the ‘Vancura Position’ drawing set-up.
After 58...Ke7? 59.a7, Black’s king would be caught in the middle of nowhere between the drawing
zones: 59...Kd7 60.Rh8 Rxa7 61.Rh7++–.
58...Kg7! 59.Ke4 Rb5!
Very strong! The rook also prepares to move into position, on the 6th rank. 59...Rc5, with the same
idea, draws as well. All other moves lose, e.g.
a) 59...Kf7? 60.Kd4 Kf6 (60...Ke7 61.a7+–) 61.Kc4 Kf7 62.Kb4 Ra1 63.Kb5 Rb1+ 64.Kc6 Rc1+
65.Kb7 Rb1+ 66.Ka7 Ke7 67.Rb8! (67.Rc8? Kd7 68.Rb8 Rc1=) 67...Rc1 68.Kb7 Rb1+ 69.Ka8 Rc1
70.a7 Kd6


Now a winning procedure follows, which you should know: 71.Kb7 Rb1+ 72.Kc8 Rc1+ 73.Kd8 Rh1
74.Rb6+ Kc5 75.Rc6+! Kxc6 (75...Kd5 76.Ra6+–; 75...Kb5 76.Rc8 Rh8+ 77.Kc7 Rh7+ 78.Kb8 Kb6
79.a8=Q+–) 76.a8=Q++–.
b) 59...Rg5? 60.Ra7+! Kf6 61.Kd4 Ke6 (61...Ra5 62.Ra8 Kg7 63.Kc4+–) 62.Ra8 Kd7 63.a7+–.
After 60.Rc8, the rook simply retreats: 60...Ra5! 61.Rc6 Kf7 62.Kd4 Ke7 63.Kc4 Kd7 64.Rh6 Kc7
65.Kb4 Ra1=.
The g-file is best for the black king, as he can support his rook checking on the f-file.
Or 61.Rb7 Ra5 62.a7 Kf6 63.Kd4 Ke6 64.Kc4 Kd6 65.Kb4 Kc6 66.Rh7 Ra1=.
61...Rb6 62.Ke5 Rc6 63.Ra8 Kg7!


Finally H. Nakamura has reached the textbook ‘Vancura Position’. Black’s rook ties White’s to the
defense of the a-pawn and White’s king finds no hiding place on the queenside. This is the drawing
principle as Black’s king is in the drawing zone of g7 or h7 preventing the rook trick Rh8 followed by
Rh7+. This principle would also work with an additional white pawn on h5.
There is nothing that White can do:
a) 64.Kd5 Rf6 65.Kc5 Rf5+ 66.Kb6 Rf6+ 67.Kb7 Rf7+ 68.Kc6 Rf6+ 69.Kd5 Rb6=
b) 64.a7 Ra6 65.Kd5 Ra1 66.Kc6 Rc1+ 67.Kb6 Rb1+ 68.Ka5 Ra1+ 69.Kb4 Rb1+ 70.Ka3 Ra1+
71.Kb2 Ra6=.
64...Kg6 65.Re7 Rxa6 66.Re6+ Rxe6+ 67.Kxe6
Magnus Carlsen
Levon Aronian
Saint Louis 2014 ■


A mistake left L. Aronian in a difficult rook endgame that required an incredible amount of precision.
M. Carlsen reached a winning position but blundered it away beforehand. With this error L. Aronian
was able to beautifully show the ‘Vancura Position’ defense and obtained a draw out of seemingly
Many spectators without access to tablebases were confused online as the engines keep saying this
position is a win, but it most certainly was not. The ‘Vancura Position’ defense even works against
two additional h-pawns:
The rook moves into position to force White’s rook to stay in front of the a-pawn.
48.Kxc3 Rf5 49.Ra8 Rb5 50.Kc4 Rf5 51.Kb4 Rf4+ 52.Kc5 Rf5+ 53.Kd4 Rb5 54.Ke4 Rc5 55.Ra6
Rb5 56.h4 Rc5 57.Kd4 Rb5 58.Kc4 Rf5 59.Kb4 Rf4+


As the white rook is ready to improve its position Black must start checking.
60.Kc5 Rf5+ 61.Kb6 Rf6+ 62.Kb7 Rf7+ 63.Kc8 Rf8+ 64.Kd7 Rf5 65.Ra8 Rd5+ 66.Ke6
66.Kc6, is not changing anything: 66...Rf5 67.a6 Rf6+ 68.Kb5 Rf5+ 69.Kc4 Rf4+ 70.Kd3 Rf6=.
66...Rb5 67.Kf6 Rc5 68.Ra7+ Kxh6 69.Kf7 Rb5 70.a6

Finally, White moved his pawn to a6 therefore Black must be on the alert.


The only drawing move!
71.Kg8 Rb8+!
This check is necessary. The waiting 71...Rc6? would lose to 72.Rh7+ Kg6 73.Rg7+ Kh6 74.a7+–.
72.Kf7 Rb6!
The 6th rank!
73.Ke7 Kg6 74.Kd7 Rf6 75.Ra8 Kg7 76.Kc7 Rf7+ 77.Kd6 Rf6+ 78.Ke5 Rb6 79.Ra7+ Kg6
79...Kg8= works as well.

After 80...Kxh5? 81.Ra8, Black’s king does not reach the drawing zone in time: 81...Kg6 82.a7 Ra6
81.Kf5 Rc6 82.Re7
82.Ra8 Kg7 83.h6+ Kh7 (83...Rxh6 draws as well) 84.Kg5 Rg6+ 85.Kf5 Rb6=.
82...Rxa6 83.Re6+ Rxe6 84.Kxe6 Kxh5

Things do not always run smoothly as mistakes are lurking around and a minor one can be lethal.
Even 2700+ players can mishandle a well-known mechanism:
David Howell
Hikaru Nakamura
Caleta 2015 □

Preparing for the well-known position, but this is a lethal mistake.
a) White could draw with 63.Rd3+! Ke4 64.Rc3! (64.Rb3? h4! 65.Kb2 Kf4 66.Rc3 Rf1 67.Kc2 Rf3–
+) 64...h4 (64...Rf1 65.Rh3 Rf5 66.Kd2=; 64...Kf4 65.Kd2! h4 66.Ke2!=; 64...Rh2+ 65.Kd1=)
65.Kb2 Rh2+ 66.Kb1, but not 66.Kb3? Rd2!–+.
b) Note that 63.Rc5? also loses: 63...h4 64.Rh5 h3 65.Rh8 Kg2 66.Rg8+ Kh2 67.Kd2 Rg1! 68.Rf8
[68...Rg5 69.Ke2 Kg3–+ also wins; but not 68...Rg4? 69.Ke2! Kg3 70.Rf3+! Kh4 71.Rf8 h2 72.Kf2!
h1=Q (72...h1=N+=) 73.Rh8+=]
69.Rg8+ (69.Ke2 h2! 70.Rg8+ Kh3! 71.Rh8+ Kg3! 72.Rg8+ Kf4!–+) 69...Kh1! 70.Rf8 h2 (70...Rg5–
+) 71.Ke3 Kg2! 72.Rg8+ Kf1! 73.Rf8+ Ke1! 74.Ra8 Rg3+! 75.Kf4 Rf3+! 76.Kg4 Rf1–+.
63.Kb2? h4 64.Rd4


The ‘Vancura Position’ could be achieved with the white rook on c4 where it can safely attack the hpawn and prevent the black king from supporting the pawn by giving side checks. But White is short
a move.
The wrong direction!
Black could have crowed his efforts with 64...Kg3!, protecting the pawn and threatening to activate
the rook to the square f1: 65.Rd3+ Kf2 66.Rd2+
[66.Rd4 h3 67.Rd2+ (67.Rh4 Kg3 68.Rh8 Rh2+ 69.Kc3 Rf2 70.Kd3 h2 71.Ke3 Rf3+ 72.Ke2 Rf4–+;
67.Rd3 h2! 68.Rd2+ Ke3 69.Rc2 Rb1+–+) 67...Ke3! 68.Rc2 Rd1! (68...h2? 69.Rc3+ Kf2 70.Rc2+!
Ke3 71.Rc3+ Kd4 72.Rh3!=) 69.Rh2 Rd2+!–+; 66.Rc3 Re1 67.Rh3 Re4, followed by 68...Kg2 and
Black wins]


66...Ke3 (The point! Black wins a tempo to free his rook because the white rook is unprotected.)
67.Rc2 Rf1 68.Rc4 Rf4–+.
65.Rc4 h3 66.Rc3+!
The ‘Vancura Position’!
66...Kd4 67.Rg3 Rh2+ 68.Ka1 Rh1+
As said beforehand it is important as well to avoid the ‘Vancura Position’ when being on the stronger
Viswanathan Anand
Alexei Shirov
Wijk aan Zee 2004 □


It looks like Black should not have much difficulty reaching a draw.
After 37.Rxa7? Ke6 38.Ra5 Rg4 39.f5+ Kf6,
White cannot improve — see the game but 37.Re5! cutting-off the black king, was the only way to
win: 37...Rg4 38.Re4 Kd5 39.Kh3! Rg7 40.Re2, followed by Kh4 and f5.
37.f5? Ke7! 38.Rxa7+ Kf6 39.Ra5 Rg4 40.a4 Rb4?
Black should keep the white king cut-off in the h-file: 40...Rg8 41.Rb5 Rg4 42.a5 Kg5 43.Rd5 Kf6
44.Kh3 Rg1 45.Kh4 Rg2=.
41.Kg3 Rc4 42.Kf3 Rh4 43.Ke3 Rg4 44.Kd3 Rf4 45.Ra8 Kg7


A precise move as the naive 46.a5?, would allow Black to reach the ‘Vancura Position’ after
46...Rxf5 47.a6 Rf6=.
46...Kf6 47.Ra5?!
Losing time. 47.a5! Rxf5 48.a6 Ra5 49.Ra8! Kg7 50.Kc4 Rf5 51.Rb8, wins, as Black can never reach
the desired set-up.
47...Rh4 48.Kc3
Again 48.Ra8!+– is quicker.
48...Rf4 49.Kd3 Rh4 50.Ra8! Kxf5 51.a5 Kg6
51...Ke6 is losing to 52.a6 Kd7 53.a7 Ra4 54.Rh8+–.
52.a6 Kg7
52...Ra4 53.Kc3 Kg7 54.Kb3 Ra1 55.Kb4+–, as the white king reaches a7.


The only winning move — a position to remember!
53...Kg8 54.Rb7+–
54.Kc3! Re4
54...Rh1 55.Ra8 Ra1 56.Kc4+–
55.Rh7 Ra4 56.a7
And Black resigned: 56...Ke6 57.Kb3 Ra1 58.Kb4 Kd6 59.Kb5+–.
A complicated example completes this Chapter. It strongly resembles the above game M. Carlsen-L.
Aronian: the ‘Vancura Position’ is good even with extra h-pawn(s) for the stronger side:
Nikita Maiorov
Azer Mirzoev
Livigno 2011 ■


Too greedy.
Black must prevent the activation of White’s rook: 73...Rg6+! and now:
a) After 74.Kc5 h6 75.Ra8+ Kh7 76.a7 Ra6 77.Kb5 Ra1= White’s king has no shelter on the
b) 74.Ke5, also fails to win after 74...h5! 75.g5 (75.gxh5 Rb6= is a version of the ‘Vancura Position’
defense: 75.Kf5 Rb6 76.g5 Rc6=) 75...Rb6! (75...Rxg5+? 76.Kf6 Ra5 77.Kg6+–) 76.Kf4 Rg6!
(76...Rb4+? 77.Kf5 Rb5+ 78.Kg6 Rb6+ 79.Kxh5 Rc6 80.g6 Rc1 81.Rb7+–) 77.Ke4 Rc6 78.Kf3
Rc3+ 79.Kf4 Rc4+ 80.Ke5 Rc6 81.Kf5 Rd6 82.g6 (82.Kf4 Rg6=) 82...h4! 83.Kg5 h3 84.Kh6 Rd8=.
c) 74.Kc7 h5 75.Ra8+ (75.gxh5 Rf6!=)


75...Kh7! (but not 75...Kg7? 76.Kb7+– and Black has no check and goes down) 76.gxh5 Rf6!, when
the ‘Vancura Position’ draw is reached: 77.Ra7
[77.a7 Rf7+! (77...Ra6? 78.Kb7 Ra1 79.Rc8+–) 78.Kb6 Rf6+ 79.Kc5 Rf5+ 80.Kd4 Ra5=]
77...Kg7 78.Kb8+ Kg8 79.Kb7 Rf7+ 80.Kc6 Rf6+ 81.Kd5 Rb6 82.Kc5 Rf6 83.Ra8+ Kg7 84.h6+
Kh7! (84...Rxh6? 85.Kb5 Rh5+ 86.Kb6 Rh6+ 87.Kb7 Rh1 88.Rb8+–; 84...Kxh6 85.a7 Kg7
86.Rg8++–) 85.Kb5 Rf5+ 86.Kc6 Rf6+ 87.Kb7 Rf7+= And White cannot make progress as he is
unable to activate his rook.
73...Rxg4? 74.Re7!
Full control of the 7th rank gives White the victory in typical style. 74.Kc7?!, is bad technique but
also wins: 74...Rg7+ 75.Kb8 Rg6 76.Ra8 Kg7 77.Kb7! Rg1 78.Rb8 Rb1+ 79.Ka8 Ra1 80.a7 Kg6
81.Rb5! (the decisive cut-off) 81...h5 82.Kb7 h4 83.Kb6+–.
Alternatives are not of any help:
a) 74...Ra4 75.a7 Ra6+ 76.Ke5 Ra5+ 77.Kf6 Ra6+ 78.Re6 Rxa7?! 79.Re8#;
b) or 74...Rg6+ 75.Re6! Rg7 (75...Rg1 76.a7 Ra1 77.Re8+ Kf7 78.a8=Q+–) 76.Re8+ Kf7 77.Re7+
Kf6 78.Rxg7 Kxg7 79.a7 h5 80.a8=Q+–.
75.Kc5 Ra4 76.a7 h5


77.Re8+ Kg7 78.a8=Q Rxa8 79.Rxa8 Kg6 80.Kd4 Kg5 81.Ke3 Kg4 82.Ke4
The ‘Vancura Position’ is a basic cornerstone of any rook ending as it might arise at any time.
It has been extensively analyzed for centuries and now with so many modern tools available it is quite
easy to learn and practice.


This is a very interesting and uncommon case! Although the win is usually not difficult there are
several exceptional cases that lead to a draw.
Usually, the superior side sacrifices one of the pawns in order to reach a known winning position with
rook & pawn vs rook.
The strongest side should keep in mind that it is desirable to be able to protect both pawns with the
rook to free the king for action.
Although all the above can be applicable in all cases here we will examine the two rook pawns which
is one of the most difficult cases.
We will start with some ‘easy cases’ highlighting the passivity of the defending rook which is a fatal
Evgeny Bareev
Sergei Rublevsky
Khanty-Mansiysk 2005 □

The defending rook is horribly placed. It is passive and unable to do anything so White can easily
prevail by using his h-pawn as a ‘decoy’ which allows his king to approach the queenside.
59.Ra6! Kh5 60.Kf4


Black threw-in the towel.
Yuri Shulman
Leonid Yudasin
St Petersburg 1998 □

Here the stronger side can improve his rook:
50.Rh1 is good as well.
50.Ra1 Kb7 51.a6+ Ka7 52.Ra4!


Black resigned as the white king has a free hand to assist his h-pawn.
Roman Bar
Alexander Finkel
Tel Aviv 2002 ■

Passivity is bad for either rook. This is an easy case as Black cannot improve his rook.

Even the slightest chances are gone after this advance...
55.Kh2 Ke7 56.Kg2 Kd7 57.Kh2 Kc7 58.Kg2 Kb7 59.Ra3 Kb6 60.Ra8 Kc5 61.Rc8+ Kd5 62.Ra8
Ke4 63.Ra4+ Ke3 64.Ra3+ Kf4 65.Ra4+ Kxg5 66.Ra5+ Kxg4
Black won both white pawns but there is no win in sight!

67.Ra4+ Kf5 68.Ra5+ Ke4
So now let’s move to some more ‘natural’ examples:
Example 1 ■


This is the ideal position that White must strive for.
1...Ra4 2.Rf3+ Kg5!
The h-pawn marches on after 2...Ke6 3.h4 Ke5 4.h5.
3.Kf2 Re4
Otherwise the white king will support the advance of the a-pawn by moving to b3.
4.Rg3+ Kf5
4...Kh4 5.Rg4+ Rxg4 6.hxg4 Kxg4 7.a4+–
5.Rg4 Re6 6.a4 Rb6 7.Kg3 Rb3+ 8.Kh4 Ra3


Making use of this tactical device White can achieve the further advance of his pawns.
9...Kf6 10.Rg5 Ra1 11.Kh5 Ra3 12.h4 Ra4 13.a6
Example 2 ■

This is a very instructive example. If White had a pawn on any other file than the g or h files then he

would simply win with 1.a7! followed by the advance of his other pawn. In this case he must seek a
different winning plan.
1...Ra4 2.Kg3 Kh7 3.h5!
Certainly not 3.Kf3? Rxh4! 4.Ke3 Rg4=.
3...Ra5 4.Ra7+! Kh6 5.Kf4 Rxh5
5...Ra4+ 6.Ke5 Ra2 7.Ra8 Kh7 8.Kd6 Kh6 9.Kc7 Kh7 10.Kb7 Rb2+ 11.Ka7 Ra2 12.Rb8 Kh6
6.Ke4 1–0
Let’s have a look at one from own play where I was defending and faced a difficult challenge.
Riho Liiva
Efstratios Grivas
Kallithea 2006 ■

Black rightly tried to survive in this notorious ending. His active king and rook should prove enough
for that purpose.
Activity above all!

The alternative 38.Ra8 Kh3 39.Kg1, fails to impress due to the delicate 39...Kg4!
[39...Rg2+? 40.Kf1 Rb2 41.a4!+–, but not 41.Ke1? Kg4! (41...Kxh2? 42.a4+–) 42.a4 Kf4 43.a5 Ke3
44.Kd1 Kd3 45.Rd8+ (45.Kc1 Rxh2 46.a6 Rc2+ 47.Kb1 Rc7 48.a7 Kc4=) 45...Kc4 46.Rh8 Ra2
47.Rh5 Kd3=]
40.a4 Ra2 41.a5 Kf3 42.h4 Kg3 43.Kf1

And now Black has a unique saving line by threatening mate: 43...Kf3! 44.Ke1 Ke3 45.Kd1 Kd3
46.Kc1 Kc3 47.Kb1 Rb2+! 48.Ka1 Rh2 49.a6 (49.Rh8 Kb3=) 49...Rxh4 50.a7 Rc4!=.
A draw would be in sight with the simple 38...Kh4! (or 38...Kh3) 39.a3 Ra2 40.Ra7 Kh3 41.Kg1
Rg2+ 42.Kf1 Ra2 43.Rh7+ (43.a4 Kxh2 44.a5 Kg3 45.a6 Kf3=) 43...Kg4 44.h3+ Kg3=.
39.Rg3+! Kf4


White could play the winning 40.Rg2! Rb4 41.a3 Ra4 42.Ra2 (42.Rg3+–) 42...Kg4 43.Kg2 Kf4
44.Rf2+ Kg5 45.Rf3 Rc4 46.h3 reaching the position of the first example. A position he should avoid
can be seen after 46...Ra4 47.Kg3 Ra8 48.h4+? (48.Kf2+–) 48...Kg6 49.Rc3 Kf6 50.Rb3

Here Black can draw: 50...Ra4! 51.Rf3+ (51.h5 Kg5) 51...Kg6 52.Kf2 Rxh4 53.Ke2 Ra4 54.Kd2 Ra6
55.Kc2 Rf6! 56.Rd3 Kf7 57.Kb3 Ke7 58.Kb4 Rd6 59.Rh3 Kd7=.
40...Ra2 41.Rb3 Kg4 42.h3+ Kf4


Now the pawn structure is ideal, but the white king is stuck in the first rank.
43.Kg1 Kf5 44.Kf1 Ke5 45.Ke1 Kd4?
Another blunder! Black had to choose between 45...Ke4 or 45...Kf4 or 45...Kf5 keeping the draw.

There was no win with 46.Kd1 Rh2 47.Rg3 Kc4 48.Ke1 Kd4 49.Kf1 Rc2! [49...Ra2? 50.h4!
(50.Kg1? Ke4 51.Rc3 Kf4!=) 50...Rh2 51.Rb3!+–] 50.h4 Ke4 but White could play 46.h4! achieving
a won position. The black king should follow the footsteps of his opponent but with 45...Kd4? he
overdid it.
46...Ke5 47.h4


A bad mistake that should have cost the half point. 47...Kf4 48.h5 Rh2! 49.Rb4+ Kf3, leads to a
White forgot about the already mentioned 48.h5! Kg5 49.Rh3+–.
48...Kg4! 49.Kd1
49.Rb4+ Kf3 50.a4 Ke3=


And now it is an easy draw as the black king will return to the queenside in time to assist his rook in
holding up the white a-pawn.
50.Kc1 Kg4 51.Kb1 Rd2 52.a4 Kf5 53.a5 Ke5 54.Ra3 Rh2 55.Rd3 Rg2 56.Rd7 Ke6 57.Rd3 Rg5
58.Ra3 Kd6 59.a6 Rg8 60.Kb2
That was a good piece of analysis and contains important information to help understand this ending!
When one of the pawns is far advanced and blocked by the opponent’s king the ‘Vancura’ position
can come into play.
Anatoly Karpov
Artur Jussupow
Linares 1991 ■


White’s a-pawn is rather advanced so Black will defend here with the ‘Vancura’ method.
60...Rc5! 61.Rh8 Rg5 62.Ke4 Rc5 63.Kf4 Rc4+!
The only move — Black must start checking!
64.Ke5 Rc5+!
64...Rc6? loses to 65.Rh7+! Ka8 (65...Kxa6 66.Rd7+–) 66.Kd4! Rxa6 67.Rc7+–.
65.Ke6 Rg5
65...Rc6+? loses to 66.Kd7 Rc5 67.h6 Rd5+ 68.Ke6 Rh5 69.Kf6 Rh1 70.Kg7+–. But the alternatives
65...Rb5, or 65...Ra5 were good.
66.Kf7 Rc5! 67.Rh7 Kxa6 68.h6 Rc7+
Draw. The white king has nowhere to hide.
Alexey Suetin
Ferenc Portisch
Belgrade 1977 □


White could move his rook to a more advantageous position but it appears A. Suetin does not suspect
there is any danger of a draw.
Too hasty!
White could prevail with 48.Rb5+ Kc6 49.Rb4 Kc5 50.Rg4+–;
or 48.Rc5 Kb6 (48...Rh3 49.Rc4+–) 49.Rc3 Rf2+ (49...Rf4 50.Rh3+–) 50.Kb3 (50.Rc2 Rf5 51.Rh2+–)
50...Rh2 51.Rc4 and 52.a4 with an easy win.
48.a4? Rf4!
Now it’s a draw!


The threat was 50.Rb5+ followed by 51.h5+–.
50.a5 Re4 51.Kc3 Rf4 52.Kd3 Rg4 53.Ke3 Rc4 54.Kf3 Rc3+
In the ‘Vancura’ position the rook is best placed on the bishop file. 54...Rb4? loses to 55.Rh8 Ka7
56.h5 Rb5 57.h6+–.
55.Ke4 Rc4+ 56.Kd5 Rg4!
And the game was drawn in 73 moves.
John Emms
Liafbern Riemersma
Gausdal 1993 □


A holdable position.
Not losing but obviously a step in the wrong direction. Accurate is 57.Rb5 Kd6 58.Rf5 Ra1 59.Kh2!
a4 60.Rf4! a3 61.Rf3! Kc5 (61...a2 62.Ra3=) 62.Rf5+=.
57.Rh6+?! Kd5 58.Rh5+ Kc4 59.Rxh4+?
Greedy! 59.Rf5! Ra1 60.Kh2 a4 61.Rf4+ is a draw!
59...Kb3 60.Rh5


White’s rook is misplaced, it is important to keep it on the h-file. J. Emms demonstrated that this
could have been achieved by the subtle move 60...Ra3! for example: 61.Kh2 (61.Rg5 Kb4+ 62.Kg2
Rc3 63.Rg8 a4 64.Rb8+ Kc4 65.Kf2 a3 66.Ra8 Kb3 67.Ke2 a2 68.Kd2 Rc4–+) 61...a4! 62.Rh3+
Kb2 63.Rh4 Ra2! 64.Kh1 (64.Rg4 Kb3+ 65.Kh3 Rc2–+) 64...Ra1+ 65.Kg2 a3 66.Rh3 a2–+.
An error in return.
A draw was possible through 61.Rg5! a4 62.Rg3+ Kb4 (62...Kc2 63.Rg2+ Kd3 64.Rg4 a3 65.Rg3+
Kc2 66.Kh2=) 63.Rg4+ Kb5 64.Rg5+ Kc6


65.Rg6+! The king should be driven as far away as possible.
[Premature is 65.Rg4? a3 66.Kh2 (66.Kg2 Kb5! 67.Rg3 Kb4 68.Rg4+ Kb3 69.Rg3+ Kb2–+)
66...Rb1 67.Ra4 Rb3 68.Kg2 Kb5 69.Ra8 Kb4 70.Kf2 Kc3 71.Ke2 Kb2–+]
65...Kd5 66.Rg5+ Ke6 67.Kg2 Rb1 (67...a3 68.Rg3 Ke5 69.Rf3=) 68.Ra5 Rb4 69.Kf2 Kd6 70.Ke2
Kc6 71.Kd2 Kb6 72.Ra8 Kb5 73.Kc2=.
61...a4 62.Kf2
62.Rh3+ Kb2 63.Rh4 a3–+
62...a3 63.Rb5+ Ka2 64.Ke2 Rb1 65.Rd5
65.Ra5 Kb2 66.Rb5+ Ka1 67.Ra5 a2–+
65...Kb2 66.Rd2+ Kb3 67.Rd3+ Ka4 68.Rd4+ Rb4 69.Rd8 a2 70.Kd3 Kb3 0–1
The same position as in the above game appears in the following one!
Laszlo Szabo
Vladimir Tukmakov
Buenos Aires 1970 □


We already know the accurate continuation from the previous example.
66.Kg2?! Kd6 67.Kf2?! Ra2+

After 68.Kg1! Kc6 69.Rf5! the position was still drawn.
68...Ra1+ 69.Ke2
There is no salvation anymore:

a) 69.Kf2 a4 70.Rxh4 a3 71.Rh3 (71.Ra4 Kc5–+) 71...a2 72.Ra3 Rh1;
b) or 69.Kd2 Rh1! 70.Rxa5 h3 71.Rh5 h2 72.Ke2 Ra1–+.
69...a4 70.Rh6+
70.Rxh4 a3 71.Ra4 a2 72.Kd2 Rh1–+
70...Ke5 71.Rh5+ Kf6 72.Kf2 a3 73.Kg2 Rc1 74.Ra5 Rc3 0–1
Konstantin Lerner
Zoltan Gyimesi
Koszalin 1999 □

In order to reach the ‘Vancura’ position the black king should be closer to his ideal square: h7. Now
he is badly placed on h5 and cut-off on the sixth rank, this endgame is lost for Black.
44.Kf2 Kg5
44...Rc2+ doesn’t really help: 45.Ke3 (45.Kg3+–) 45...Rxh2 46.a5 Ra2 47.Kd4+–.


The normal defensive set-up (with the black king on h7 or somewhere close to this square) would be
to keep attacking the a-pawn from the side and, once the white king gets close to it, give checks from
the side as well.
Here, after 45...Kh5 46.Kd3 Rf4 Black cannot draw this position: 47.a5! Rf5 48.Ra8 Kh6 49.a6 Kh7
(49...Rf6 50.a7+–) 50.Rb8+–. This is why the black king belongs on h7 or close to it.
46.a5 Kf5
46...Rxh2 47.Kd4 Kf5 48.Rc6 and the black king is cut-off along the 6th rank.
47.Ra8 Rxh2 48.a6 Rh7


White could go wrong with 49.a7? Rf7 50.Kd4 Kf6 51.Kd5 Kf5 and Black draws!
49...Rf7 50.Kc5 Rf6 51.Kb5 Kf4
Michal Krasenkow
Jan Gustafsson
Berlin 2003 □


Although the white king is stuck on the last rank he can prevail.
43.a4! Kc5 44.Rg4 Kd5 45.a5 Kc6 46.a6
Or 46.Rb4 Kc5 47.Ra4 Kb5 48.a6 Kxa4 49.a7+–.
46...Kb6 47.Ra4 Ka7
Now the black king is passive therefore the white king can start play on the kingside!
48.Kc1 Rg2 49.Kd1 Rh2 50.Ke1 Rg2 51.Kf1 Rh2 52.Kg1 Rb2 53.h5 Rb5 54.Rh4!
The only winning move!


54...Kxa6 55.h6
And Black resigned.
After 55...Rb8 56.h7 Rh8 57.Kg2 the white king reaches g6.
Ivan Saric
Francisco Vallejo Pons
Rijeka 2010 □


Black can win here. His treat is to place his rook behind the passed pawn.
50.Rc6 h5 51.Ra6 Rb4+ 52.Kg5
After 52.Kg3 h4+ 53.Kh3 Kf7 the black king will run to the queenside.
52...h4 53.Kg6 Kf8 54.Kf6 Ke8 55.Ke6 Kd8 56.Kd6 Rd4+
56...Kc8 57.Kc6 Kb8 58.Kc5 Rf4 would win as well.
57.Ke5 Rg4 58.Rd6+ Kc7 59.Rd3 Kc6
Yuri Shulman
Stanislav Savchenko
Minsk 1996 □


There is no doubt that Black will prevail.
After 45.Rc5 the simplest is 45...Re6 46.Ra5 a6–+.
45.Rh7 Kg5 46.Rc7 Kf4 47.Rf7+ Ke4 48.Re7+ Kd3 49.Rd7+ Ke4 50.Re7+ Kf5 51.Rc7 Ra5
52.Kh3 Kg5 53.Kg3 Kh5 54.Kh3
54.Rc3 Rf5 55.Ra3 a5–+
54...Ra3+ 55.Kg2 Ra4 56.Rc5+ Kh4 57.Rc6 h5 58.Kh2 Ra2+ 59.Kg1 Ra3 60.Kh2 Rh3+ 61.Kg2
Rg3+ 62.Kh2 a5 63.Ra6 Rg5 64.Ra7 Rb5 65.Kg2 Rb2+ 66.Kg1 Ra2 67.Rc7 Ra3


If 68.Kh2 then 68...Rh3+ 69.Kg2 Rg3+ 70.Kh2 a4 as ‘usual’!
68...Kh3 69.Rc7 a4 70.Kf2 Rb3 71.Rc4 a3 72.Ra4

White resigned: 73.Ra8 Kh2 (73...Rb2+ 74.Kg1 Rb1+ 75.Kf2 Ra1 76.Ra4 a2 77.Ra8 Rh1 78.Ra3+
Kg4 79.Ra4+ Kf5–+) 74.Ra4 h3 75.Ra8 Rb2+ 76.Kf3 a2 77.Ra7 Kg1 78.Rg7+ Kf1 79.Ra7 h2.



Among nearly all rook vs pawn endings the rook vs three connected pawns on the same side is the
most difficult to handle.
It is also a scary one! I have seen many good players avoiding such cases mainly because of a lack of
knowledge and uncertainty! Add the usual time trouble and you will have the full picture of the
problem. A deeper amount of knowledge is needed to face this challenge.
The basic stuff is:
SOS Tip 1 — Rook vs Three Pawns
1. The pawns can draw if they can primary form a chain.
2. Three pawns on the 5th or beyond win against the rook in the absence of the kings.
3. Three pawns on the 4th, supported by their king and in the absence of the opponent king may
also win.
4. Three pawns that are not all on the 5th, in the absence of their king, lose.
5. If two of the defensive side’s pawns are placed on the 3rd rank or one pawn has reached the 2nd
rank while the other is on the 4th rank a rook cannot stop them. Sometimes, however, the rook
side can be saved by creating mate threats when the opponents king is cornered in an edge of the
6. If the attacking rook gets behind the base of the pawn chain we have a winning position.
7. The defensive side’s king shouldn’t move on the same file as his furthest advanced pawn.
With the attacking king directly in front of the pawns the critical position can be seen in the next
Example 1
Main Drawing Position ■


1...Kf7? loses to 2.Ke5! Kg7 (2...Ke7 3.Re6+ Kf7 4.Rf6+ Ke7 5.Rf4+–) 3.Ra1! Kg6 (3...f2
4.Kd4+–; or 3...e3 4.Kf4+–) 4.Kxd5 Kf5 5.Kd4 Kf4 6.Ra8 f2 7.Rf8+ Kg3 8.Ke3+–.
1...Kd7! 2.Rf6
2.Kxd5?! is risky but still drawn: 2...f2 3.Ra7+ Ke8 4.Ra1 e3 5.Ke6=.
2...Ke7! 3.Rf5 Ke6 4.Re5+ Kf6! 5.Re8 Kf7 6.Ra8
6.Rd8? f2–+
6...Ke7 7.Ra6 Kd7
Example 2
Main Drawing Position □


Another typical drawn position.
1.Rf2 Kg6 2.Kf4 Kf6
2...Kh5 is still OK: 3.Rd2 Kh4? (3...Kg6= or 3...Kh6=) 4.Rd6 Kh5 5.Re6 h2 6.Re8+.
3.Re2 Kf7 4.Re5

4.Kxf5? even loses to 4...g3–+.


Black must avoid squares h5 and h7 and prevent the capture of the f-pawn with check.
5.Rxf5? h2 6.Rg5+ Kh6–+
As previously noted 5...Kh7? loses to 6.Kg5! Kg7 7.Rg6+ Kh7 8.Rh6+ Kg7 9.Rh5+–.

6...Kf7! 7.Rh6 Kg7 8.Rh5
8.Kg5 f4!=
8...Kg6 9.Rg5+ Kh6 10.Rg8 Kh7 11.Rd8 Kg7
Study 1
Kopaev Nikolay 1966 □


If it is Black’s move here we would have a winning position for the rook. With some accurate
maneuvers White loses a tempo passing the move to Black!
1.Ke3! Ke5 2.Re8+ Kf5
Alternatives are:
a) 2...Kd6 3.Kd4 Kd7 4.Re5! Kc6 5.Rf5+–
b) 2...Kf6 3.Kf2 Kf5 4.Kg3

4...Kf6 (4...d4 5.Re7 d3 6.Re8 d2 7.Rd8+–) 5.Kh2! Kf5 6.Rd8 Ke6 7.Kh3 Ke5 8.Kg3! Ke6 (8...Kd4
9.Kf2 Kc4 10.Ke3+–) 9.Kf2 Ke5 10.Ke3 Ke6 11.Kd4

And it is now Black on the move: 11...Kf5 12.Rxd5+ Kf4 13.Rd8+–.
3.Kf2 Kf6
Or 3...Kf4 4.Re6 d4 (4...Kf5 5.Rd6 Ke5 6.Rd7 d4 7.Rd8+–) 5.Rf6+ Ke5 6.Rf8 d3 (6...Ke6 7.Rd8
Ke5 8.Rd7+–) 7.Ke3+–.

4.Ke1 or 4.Kf1 are the same.


4...Kf5 5.Kf1!
Long Distance Opposition!
5...Kf4 6.Kf2 Kf5
6...d4 loses to 7.Rf8+ Ke5 (7...Kg4 8.Rd8 e3+ 9.Ke1+–) 8.Rd8 d3 9.Ke3+–.
7.Kg3 Kf6 8.Kh2! Kf5 9.Rd8 Ke6 10.Kh3 Ke5 11.Kg3

11...Kd4 12.Kf2 Ke5 13.Ke3 Ke6 14.Kd4+–
12.Kf2 Ke5 13.Ke3 Ke6 14.Kd4 Kf6 15.Rd6+
And White wins the pawns.
Colin Crouch
McShane Luke
England 1999 □


White has an optimal drawn position.
A mistake allowing the rook to get behind the pawns. White had to move his king on the 2nd and 3rd
rank for the time being.
Returning the favor! 68...Rh3 or 68...Rh1 would win.
69.Kb3 Kc5 70.Ka3! Kb6 71.Kb3 Kc5 72.Ka3! Rh3+ 73.Kb2 Kb6 74.Ka2!
Again, the only defensive move. 74.Kc2? loses to 74...Ra3.
74...Ka5 75.Kb2 Rg3


Humans crack under pressure and lack of concrete knowledge!
Correct was 76.Ka2! Kb4 77.c7 Ra3+ 78.Kb2 Rb3+ 79.Ka2 when Black has nothing better than
perpetual check.
Winning was the known 76...Kb4! taking advantage of the white king’s unfortunate position on the cfile: 77.Kd2 (77.c7 Rc3+–+; or 77.Kb2 Rg8 78.c7 Kc5–+) 77...Rg8!


78.Kd3 Kxa4 79.Kc4 Ka5 80.Kc5 Rg5+–+.
77.Kb3 Rb4+

78.Kc3! Rb1 79.Kc2 Rf1 80.Kb3?
The final blunder.
80.Kb2! was the only drawing move!


81.Kc3 Ra3+!–+
81...Rxa4+ 82.Kc5 Ra1 83.c7 Rc1+ 84.Kd6 Kb6 0–1
Ilan Manor
Bartlomiej Macieja
Willemstad 2001 □


Here the black king is rather far away allowing White to build the desired fortress.
53.h4 Ka3 54.h5 Kb4 55.g4 Kc5 56.g5 Kd6 57.f4 Ke6 58.h6 Kf5 59.Kh2?
The known 59.Kf2! was good enough!
59...Kg4! 60.Kg2 Rb2+
60...Rg3+ 61.Kf2 (61.Kh2 Rh3+ 62.Kg2 Rh5–+) 61...Rh3–+
61.Kg1 Kf5 62.Kh1 Rb4
62...Kxf4 63.h7 Kg3 64.h8=Q Rb1#
63.Kg2 Rxf4 0–1
Gisela Fischdick
Philipp Schlosser
Passau 1994 ■

White has a won position as her rook has gotten behind the pawn chain.
61...f4 62.Rf8
62.Rd8 Ke6 63.Rf8 Ke5 64.Rf7+– is a bit more accurate.
62...f3 63.Ra8


63.Re8+ Kf6 64.Kg1!+–
This makes it easier. Black had to opt for 63...Kd6 and pray: 64.Rd8+! (64.Ke3? Ke7=) 64...Ke6
65.Ke3 Ke5 66.Re8+ Kf5 67.Kf2 Kf6 68.Kg1+–.
64.Rd8 d3 65.Ke3 Kf5 66.Rf8+ Kg4 67.Rg8+ Kf5
Draw agreed! Of course White wins easily after the simple 68.Re8 f2 (68...Kg4 69.Rxe4+ Kg3
70.Rf4 d2 71.Rxf3+ Kg2 72.Rf2++–) 69.Rf8+ Ke6 70.Rxf2.


This is an even more difficult case than the previous one! An extra rook for both sides creates unusual
situations with fresh problems to solve.
In looking at this ‘second case’ the main question is: can the rooks prevail? Barring some rare mating
traps and nets the usual way is the exchange of a pair of rooks.
This is an easy case as the unhealthy pawn structure doesn’t really help the defender:
Peter Leko
Alexei Shirov
Dortmund 1996 ■

White’s pawn structure is crippled and that’s already a great minus.
55...Rd6 56.Rf2+ Kg6 57.Ra2
57.g4 loses to 57...Rd3+ (57...Rdd4–+) 58.Rf3 Rxf3+ 59.gxf3 Kf6–+.


There is hardly anything good left for White. 58.Ra4 loses to 58...Kh5! 59.Ra5+ Kh6 60.Rf5 Rdd3
(60...Rg6 61.Rf3 Rxf3 62.gxf3 Rf6!–+) 61.Rg5 Rd1 62.Kh2 Rbb1 63.Kh3 Rh1+ 64.Kg4 Rb4+ 65.Kf3
Rf1+ 66.Ke3 Rf6–+ planning ...Rg6.
58...Kh7 59.Ra7+ Kh6 60.Rf7 Rdd3 0–1
Not the easiest case as the rooks have plenty of space to move around. However, defending seems
possible based upon the placement of the pawns.
Sergey Karjakin
Viktor Moskalenko
Bilbao 2004 ■


Black should be fine here but being a rapid game has its minuses...
58...e5 59.R6h5 Ke6 60.Rh8
It is too early to offer a rook exchange with 60.Rh3 as after 60...Rxh3 61.Rxh3 e4 or 61...Kd6 Black
gets an optimal drawn position. Note that here 61...Kd7? loses to 62.Rd3 Ke6 63.Ke2 Kd5 64.Rb3 e4
65.Rb6+– — well, tablebases are unbeatable!
A blunder as the pawns now lose their optimal set-up.
Something like 60...Ra3 would have been fine.


Now this exchange is winning!
61...Rxh3 62.Rxh3 e4 63.Ke2+– But now the black rook stands in a zugzwang position!
62.Ra8 Kd5 63.Ra5+


This loses easier but even after the stubborn 63...Ke6 64.Ra6+ Kd5, the simplest would be 65.Rf3
Kc5 66.Ra1 Kd5 67.Rh1 Kc5

68.Re1! Rxe1 69.Kxe1+–.
And Black’s best pawn falls!
64...d3 65.Rhxe5 d2 66.Rxe4+ 1–0
Alexander Moiseenko
Maxim Matlakov
Berlin 2015 □


White is lost as his pieces are uncoordinated and he can’t provide satisfactory protection to his e3pawn. So he tried his last chance:
47.Rd6+ Kg7 48.Rxh6 Kxh6 49.Kf3

This slows the win by approximately 10 moves.
The black king has to move forward so accurate was 49...Ra7 50.d5 Ra3 51.Ke2 Kh5!–+.



Of course, we must keep in mind that this was a blitz game and humans are not machines so they tend
to err quite often!
The black king goes toward the e5 check advance!
Good was 50...Rd7 51.d5 Kf7 52.Ke3 Ke7 53.Kd4 Ra7–+.
Or 51.e5+ Kf5 52.d5=.
51...Rd7 52.e5+ Kf5 53.d6=
52.d6 Ke6 53.e5 Kf5


And now it is a draw.
54...Ra4 55.Kd3! Ke6 56.Ke3 Rb4 57.Kf3 Kf5 58.Ke3 Ra4 59.Kd3 Ke6 60.Ke3 ½–½
Viorel Iordachescu
Shardul Gagare
Abu Dhabi 2014 ■


To protect the valuable e5-pawn from e6. Black must be careful: 39...Re3? 40.Kd2 c5 (40...Kc5
41.Re1 Rxe1 42.Kxe1+–) 41.R1h3 Rxh3 42.Rxh3 c4 (42...e4 43.Rh6! e3+ 44.Kd3 Ke5 45.Rc6 Kd5
46.Rc8 Kd6 47.Kc4) 43.Rh8 c3+ 44.Kd3 and White wins.
39...Rf6 40.Kd2 c5 41.Rd1
After 41.Re1 Re6 42.Rh8 Rg6 43.Rd8+ Black can go for the rook exchange with 43...Rd6! 44.Rxd6+
Kxd6 45.Rh1 e4 46.Rh6+ as now he can retreat with 46...Kd7! (46...Ke5? 47.Rc6 Kd5 48.Rc7 Kd6
49.Rc8 Kd5 50.Kc2 Kd6 51.Kb3 Kd5 52.Rd8+ Ke5 53.Kc4+–) 47.Rh5 e3+ 48.Kd3 Kd6 49.Kc4
Ke6 50.Rh6+ Kd7!=.
41...Rf2+ 42.Ke1 Rb2 43.Rh8

But here Black resigned in a drawn position! 43...e4 44.Rd8+ Kc4! 45.Rc1+ Kd3 46.Rc8 e3 47.R8xc5
Re2+ 48.Kf1 Rf2+ 49.Kg1 Ke4=.
Contrary to what we have been taught for the value of having central pawns, here things look easier
for the defender as the rooks have a limited ‘active battlefield’!
Problems arise when the pawns are not advanced enough.
Aleksey Goganov

Sergei Zhigalko
Baku 2013 □

A very primitive starting position.
White must push his pawns and get ready to obtain a theoretically drawn position after a rook
exchange. But as the pawns are still not advanced enough Black should win.
53.g4 Ra2 54.Rd1
It is too early for the exchange: 54.Rxa2 Rxa2 55.Kg3 Ra3+ 56.f3 Kf7 57.h4 Rb3–+.
54...Rb2 55.Kg3 Raa2 56.Rf1 Kf7 57.h4 Kf6


58.f3 loses to 58...Rg2+ 59.Kf4 Ra4+ 60.Ke3 Rb2–+.
58...Rc2 59.Kg3 Re2 60.h5
What to do?
If White stays put with 60.Kf3 he loses to 60...Ke5 61.Kg3 Ke4 62.f3+ Ke5 63.Rg1 Kf6 64.Kf4 Rg2
65.Rb1 Ra4+ 66.Ke3 Rh2 67.Rb6+ Ke5 68.Rb5+ Kd6 69.h5 Rh3 70.Rf5 Rxg4–+.
White is in zugzwang, so he plays his last card.


61.f3 Kg5 62.f4+ Kf6–+, or 61.Kf3 Kg5–+.
Wrong direction!
A waiting move like 61...Rd2! would do the job:
a) 62.Kh4 Ra3 63.Rh1 Rg2–+ or 62.Kf3 Ra3+ 63.Ke4 Ra4+ 64.Kf3 Rd3+ 65.Kg2 Raa3–+,
b) or 62.h6 Kg6 63.g5 Kf5–+,
c) lastly 62.Rf3 Rg2+ 63.Kh4 Rg1–+.
62.Kh4? Rh2#.
There is no return for Black!
After 62...Rbb2 63.Rc3! Rg2+ 64.Kh3 Rh2+ 65.Kg3 Rag2+ 66.Kf3,


White seems to hold: 66...Rh3+ (66...Rc2 67.Rxc2 Rxc2 68.g5+ Kf5 69.h6=) 67.Kxg2 Rxc3 68.g5+
Kf5 69.h6=.
Nevertheless, Black had to try to avoid the rook exchange.
63.Rxb3 Rxb3+ 64.Kf2
And it is a draw.
64...Ke6 65.Kg2 Ra3 66.h6 Kf6 67.g5+ Kf5


Of course not 68.Kh2? Kg4!–+.
Black now tried to win for a long time but White defended with accuracy. The game ended on move
Marc Narciso Dublan
Benoit Taddei
Vandoeuvre 2014 □

A drawn position.
50.Rh6 Rc7 51.Kg2 b4 52.Kf3 Kb5 53.Ke4 a5 54.Rh5+ c5
Black has an optimal position and he is in no danger.
55.Kd5 b3 56.Rh8 Rd7+
56...Rb7 57.Rxc5+ Ka4 would be drawn as well!
57.Ke6 Rb7 58.Kd6 b2 59.Rb1 Kc4 60.Ra8 Kc3! 61.Kxc5 Kc2 62.Rxb2+ Kxb2 63.Rxa5 Kb3
64.Rb5+ Rxb5+ 65.Kxb5


Nigel Short
Jonathan Tisdall
Vestmannaeyjum 1985 □

Black cannot be in real trouble here but White can press!
40.b5! axb5 41.a6 bxa6! 42.Rxc6 Rxe4 43.Rxa6
White has won Black’s knight, but Black should be able to draw comfortably thanks to his kingside
pawns eventually reaching f5, g4 and h3 versus a rook provided Black prevents the rook from settling
on f8 or g8. A danger is that if he tries to set-up this pawn chain with White having two rooks his
king could be driven out of the fortress.
43...Rg4+ 44.Kf3 Rf4+ 45.Ke3 h4?!
After the simple 45...b4 46.Rb5 h4 White cannot eliminate the b-pawn without allowing the positional
draw mentioned above.
46.Rxb5 Kh5?!
An ‘optimistic’ but pointless march that does little harm. Black should defend as he did later.
47.Ra2 Kg4 48.Rg2+


Accurate was 48...Kh5 49.Rf2 Rxf2 50.Kxf2 Kg6=.
Although it is still a draw after 49.Rf2 Black would have to find a lot of accurate moves: 49...Kg3!
50.Rxf4 gxf4+ 51.Ke2 Kg2! (51...h3? 52.Rb8 h2 53.Rg8+ Kh3 54.Kf2 h1=N+ 55.Kf3+–) 52.Rb8
f3+ 53.Ke3 f2 54.Rb2 h3 55.Rxf2+ Kg3! (55...Kg1? 56.Kf3 h2 57.Rg2+ Kh1 58.Re2+–) 56.Rxf6
Kg2!= but not 56...h2? 57.Rg6+ Kh3 58.Kf2+– so White should have given it a try!.
49...Ra4 50.Rgf2 Ra6?
A blunder but nevertheless it was hard to find the defense: 50...Ra3+ 51.Ke4 Kg3 52.Rxf6 h3 and
Black should draw due to his active king. It is more than obvious that some positions have been
created only for chess engines.
51.Ke4! Kg4 52.Rg2+ Kh5


Allowing Black to get back to his safe nest! Winning was 53.Kf5! Kh6 (53...h3 54.Rg1 Ra4
55.Kxf6+–) 54.Rb8+–.
53...Kg6 54.Rg8+ Kf7 55.Rd8 Ra4+ 56.Rd4 Ra5?!
Far simpler was 56...f5+! 57.Ke3 Rxd4 58.Kxd4 g4=.
57.Rgd2 Re5+ 58.Kf3 Rf5+ 59.Kg2 Kg7 60.Ra4
After 60.Rf2 a theoretical draw is reached: 60...h3+! 61.Kg3 Rxf2 62.Kxf2 f5=.
60...Kg6 61.Rd3 Kg7


Or 62.Rf3 Rxf3 63.Kxf3 f5!=.
Accurate again! The naive 62...Kg6? loses to 63.Rf3!+– as in this case Black cannot reach the f5, g4,
h3 pawn set-up!
63.Kh3 Kg6 64.Ra8 Kg7 65.Rb3 Kg6 66.Rg8+ Kh6 67.Kg2 Kh7 68.Rbb8 Rf5 69.Kh3 Rf4
70.Rbf8 Kh6 71.Rf7 Rf3+ 72.Kg2 Rf4 73.Rh8+ Kg6 ½–½
A rare case of course but sometimes this is also possible especially when the defender is a stronger
player and the rooks’ side mishandles the position!
Mikhailo Oleksienko
Sergey Nazaryev
Voronezh 2010 ■


White stands great without any risk of losing.
63...Kb7 64.f6 Rh8?!
Not losing but making Black’s life harder! 64...Rb6 65.Kf5 (65.f7 Rf8 66.Rf2 Kc7 67.d5 Rb4+
68.Kd3 Kd8=) 65...Rb1 would easily preserve the balance.
65.e6 looks stronger but Black can defend as well: 65...Rb1! 66.f7 (66.Rf2 Kc6!=) 66...Rf1 67.Ke5
Rf8 68.Kd6 R8xf7 69.exf7 Rxf7= achieving a theoretically drawn ‘Philidor Position’.


A losing blunder. Bad as well was 65...Rb1? 66.Rf2!+–, but Black should again have gone for
65...Rb6! 66.d5 Rf8 67.e6 (67.Rf2 Rb4+ 68.Kd3 Kc7=) 67...Rxf7 68.exf7 Rf6=.
66.d5+ Kc5 67.Rc2+ Kb5 68.e6
The white pawn phalanx is now unstoppable!
68...Rb1 69.Ke5
69.d6 Rf1 70.d7+– was quicker.
69...Rf1 70.Kd6 Rh7 71.Ke7 Re1 72.Ke8 Rf1 73.f8=Q 1–0
In our last example Black has an optimal position while another white minus is that the white king is
far away. Here it is White that should be careful!
Maurizio Bianchi
Mauro Marelli
Monza 2002 ■

40...h4 41.Rdd7 Rf3 42.Rd4 g5 43.Rg4
Although not losing the rooks must not be placed in passive positions.
Fine was 43.Rd5 Kg6 44.Rc8 f6=.


43...Kg6 44.Rc5 f6
44...f5 45.Rc6+ Kh5 46.Rgc4 g4 47.Rc8=
45.Rcc4 h3

Rooks must be active therefore 46.Rc8!= was perfect.
46...f5! 47.Rc6+ Kh5 48.Rg1 g4 49.Rc8
Too late.
Even better was 49...Rb3+ 50.Ka5 g3–+.
50.Rg8+ Kf4 51.Rh8 Rf2!


52.Rh1 Kg3 53.Ra8 Kg2–+
52...h2 53.Rd1 g3 54.Rd4+ Kf3 55.Rd3+ Kg2 56.Rd1 Rf1! 57.Rd2+ Kg1 58.Rd3 f4 0–1


In general, a rook ending with 4:3 pawns on the same side is considered quite drawish. Indeed this is
the truth but there are always ways to go wrong or simply push your opponent to make mistakes
based on your better knowledge.
The following diagram is very useful in our research:

The weaker side should advance his h-pawn on h5. The stronger side, whenever possible, should
prevent this by means of g4. The explanation consists in the fact that the stronger side’s most logical
plan is an advance of his e- and f-pawns in order to create a passed pawn.
To accomplish this plan, he must sooner or later play g4 allowing a pawn exchange on g4. But pawn
exchanges are usually favorable for the weaker side. Without ...h5 the h-pawns would have stayed on
the board.
Let’s study the drawing mechanism with the h-pawn of the 4th (5th) rank:
Efstratios Grivas
Thisefs Liverios
Iraklion 1983 ■


Black is under pressure and he must get ready to sac/lose a pawn and obtain a drawn 4:3 one-side
pawn rook ending.
33...h5! 34.Bxc4 bxc4 35.Rb4 Rd2 36.Rxc4 Rxb2 37.Rxa4

The ending is theoretically drawn but White can torture Black a bit!
38.Kg2 Kg7 39.Kf3 Rc2 40.Ra6 Rb2 41.Ra4 Rc2 42.Ra5 Rb2


43.g4 hxg4+ 44.hxg4 Rc2 45.Kg3 Rb2 46.f4 Rb1 47.Ra7 Rg1+ 48.Kf3 Rf1+ 49.Kg2 Rb1 50.Kf2
Rb2+ 51.Kg3 Rb1 52.Kh4 Re1 53.Re7 Kf6 54.Re8 Rh1+ 55.Kg3 Kg7 56.e4 Ra1 57.e5 Ra3+
58.Kh4 Rf3 59.Kg5 Re3!

And Black is ready for 60.f5 gxf5 61.gxf5 Rxe5! 62.Rxe5 f6+ 63.Kf4 fxe5+ 64.Kxe5 Kf7=. This
drawing mechanism must be learned by heart!

Tigran Petrosian
Paul Keres
Moscow 1951 ■

The usual defense.
29.Rc2 Kg7 30.Kg2 Rb5 31.Kf3 Kf6 32.h4
Nothing is obtained by the immediate 32.h3 Ra5 33.g4 hxg4+ 34.hxg4=.
32...Rf5+ 33.Kg2 Ra5 34.Kh3 Ra4 35.Rd2 Ke5 36.Rb2 Kf6 37.Rb5 Ra2 38.Kg2 Ra4 39.Kf3 Ra3
40.Kf4 Ra2 41.f3 Re2 42.e4 Re1 43.Rb6+ Kg7 44.Ra6 Rb1 45.Rc6 Rg1
Attacking from the rear is a typical defensive device.


46.Rc2 Kf6 47.Ra2 Kg7 48.Re2 Kf6 49.Re3 Kg7 50.e5 Kf8 51.g4
a) After 51.Kg5 Black can simply act with 51...Kg7=.
b) 51...Rxg3+ draws as well: 52.Kf6 Kg8 53.Rd3 Rh3 54.e6 (54.Rd8+ Kh7 55.Kxf7 Rxf3+ 56.Ke7
g5 57.hxg5 h4 58.e6 h3 59.Rd2 Kg6=) 54...fxe6 55.Kxg6 Rg3+ 56.Kxh5 Kg7=.
51...hxg4 52.fxg4 Kg7 53.Kg5 Rf1 54.Re4 Rf3


The naive 55.e6? would allow Black to gain victory with 55...f6!#.
55...gxh5 56.gxh5 f6+ 57.Kg4
Or 57.exf6+ Rxf6 58.Re7+ Rf7 59.h6+ Kg8=.
57...Rf2? 58.Kg3 Rf1 59.h6+! Kxh6 60.e6+–
A little trap before the curtain falls.
And not 58...Kxh6? 59.e6+–.
In the previous game Black didn’t have any serious troubles. It should be mentioned that when the
white pawns had been set into motion P. Keres used a typical strategic policy for this sort of position:
attacking the pawns from the rear.
But not everybody knows what to do! Look at the following youngsters’ game:
Dimitrios Pissadakis
Iordanis Krikoglou
Porto Rio 2013 □


Another drawn position but both opponents didn’t know what to do.
If 35.g4 then 35...Ra5 and ...h5.
35.Rd1 Kf6 36.Rd4 Ra2 37.Rf4+ Ke6 38.Kg4 f6
No need for this the simple answer was 38...Ra5 39.Kf3 h5.
39.Kf3 Rb2 40.Ra4 Rb7?
But there is a good reason for being low rated: your knowledge is poor.
41.Ra6+ Ke5 42.Ra5+ Ke6 43.Ra4?
43.g4! would have given White real winning chances.
An ugly move outside any theoretical knowledge. Good, of course, was 43...h5!.


This loses easily as does 44...Rb2 45.Ra7.
But Black should try to fight with 44...Kf6, although White wins: 45.Ra6+! (45.Kf4 g5+ 46.Kg3 Rb6
47.gxf5 Kxf5 48.h4 h6 is quite uncertain)

45...Ke5 (45...Kg5 46.Kg3+–) 46.Kg3 Rf7 47.f4+ Ke4 48.Re6+ Kd5
[48...Kd3 49.Kf3 Rd7 50.gxf5 gxf5 51.Rf6 Rd5 (51...Re7 52.Kf2!+–) 52.Rf7 h6 53.Rf6 h5 54.Rh6+–]
49.Re5+ Kd6 50.gxf5 gxf5 51.Kh4+–.


45.Ra6+ Ke5 46.Ra5+ Ke6 47.gxf5+

47...gxf5 48.Kf4 loses the f5-pawn.
48.Rxf5 gxf5
Or 48...Kxf5 49.h4 h6 50.Kg3 Ke4 51.h5 g5 52.Kg4+–.
49.Kf4 Kf6 50.f3 h6 51.e4 fxe4 52.fxe4 Ke6 53.e5 Ke7 54.Kf5 Kf7 55.h4 Ke7 56.Kg6 Ke6
57.Kxh6 Kxe5 58.Kg6
Should the stronger side succeed in placing his pawn on h5 then the weaker side faces considerable
difficulties. The main question is what pawn structure the weaker side should choose.
Let’s first examine the pawn structure f2, g2, h2/h3 or f7, g7, h7/h6. In response to the weaker side’s
g3/...g6 the stronger side will react with hxg6/...hxg3.
Then the weaker side can either recapture on g3/g6 with his f-pawn, giving the stronger side a passed
e-pawn, or with his king leaving himself with two weak isolated pawns on the f- and h-files.
The chances for victory are increased when the pawns of the weaker side are isolated. Let’s see how
the job should be finished:
Example 1 □

White has succeeded in implementing his initial plan with h5 and hxg6. As a result, he has created
two weak and isolated black pawns.
The continuation is important and instructive.
1.Rc6! Kf8 2.Rc8+ Kg7 3.Rd8! Rf2
The ‘desirable’ 3...h6 4.gxh6+ Kxh6 5.Rd7 Kg7 6.e6 loses.
The black rook must now forget about its 8th rank and thus allow the white king to come closer!
4...Rf3 5.Ke4! Rf2 6.Ke3 Ra2 7.f5! Rg2 8.Rd7
An alternative win is 8.g6!


8...hxg6 (8...fxg6 9.f6+ Kf7 10.Rd7+ Ke6 11.Re7++–) 9.f6+ Kh7 10.Rd7 Kg8 11.Rd8+ Kh7
8...Kf8 9.f6 Ke8 10.Re7+ Kf8 11.Ra7 Ke8 12.Ra8+ Kd7 13.Rf8+–
9.Kf4 Rg1 10.e6 Rf1+ 11.Ke5 Re1+ 12.Kd6 h5
12...Rd1+ 13.Kc7+–
13.Rxf7+ Kg8 14.f6
Probably the best example from practice is the following one. It has been analyzed in detail by so
many famous experts on endgames that I just had to compile and present all of them here in order to
fully expose the whole picture of the struggle and the various possibilities:
Jose Raul Capablanca
Frederick Dewhurst Yates
Hastings 1930 □


As we have already examined this is an easy position to evaluate (draw) and handle.
The immediate 39.g4 allows 39...Rb5! and ...h5.
39.Ra5 Rc4?!
Simpler was 39...h5!.
Now Black is in trouble.
40...h6 41.Kg3
With the plan f4 and h4–h5.
41...Rc1 42.Kg2 Rc4 43.Rd5 Ra4
43...g5 is possible as well; the position wouldn’t change much.
44.f4 Ra2+ 45.Kg3 Re2 46.Re5 Re1 47.Kf2 Rh1 48.Kg2 Re1 49.h4


The most precise again is 49...f6! 50.Re7+ Kf8 51.Re6 Kf7 52.f5 gxf5 53.gxf5 h5 54.Kf2 Ra1 55.Rb6
Ra5 56.Rb7+ Kg8 57.e4 Ra3= or 49...Ra1 50.h5 gxh5 51.Rxh5 f6! since it is proven that White
cannot win this position. We can easily observe the correct pawn structure set-up for the weaker side.
50.h5 Re2+ 51.Kf3 Re1?!
Black could have also played 51...Rh2!.
52.Ra5 Kg7 53.hxg6 Kxg6!
An obligatory weakening of the pawns. After 53...fxg6? 54.Ra7+ Kg8 (54...Kf6 55.Rh7 Rh1
56.Rxh6!+–) 55.e4! Rf1+ 56.Ke3 Rg1 57.f5! Rxg4 58.f6 Rg1 59.Kf4+– the two connected and far
advanced white pawns are decisive.
54.e4 Rf1+ 55.Kg3 Rg1+ 56.Kh3 Rf1 57.Rf5


The decisive error according to N. Kopaev. Black should have opted for the above-mentioned pawn
structure: 57...f6! 58.Kg2 Re1 59.Kf3 (59.e5 fxe5 60.Rxe5 Rxe5 61.fxe5 h5=) 59...Rf1+ 60.Ke3
Rg1= according to I. Rabinovich; or even 57...Ra1! 58.e5 (58.Kg3 Ra3+ 59.Kh4 Re3 60.Re5 Rf3
61.f5+ Kf6 62.Ra5 Re3=) 58...Ra3+ 59.Kg2 (59.Kh4 Rf3 60.Rf6+ Kg7 61.f5 Re3=) 59...Ra2+
60.Kf3 Ra3+ 61.Ke4 Ra4+ 62.Kd5 Rb4 63.Rf6+ Kg7=.
58.e5! Re3+ 59.Kg2!
White should be on the alert and refrain from something like 59.Kh4? Rf3 60.Rf6+ Kg7 61.g5 hxg5+
59...Ra3 60.Rf6+ Kg7


A typical won position. Black has weakened his pawn structure and is also unable to rid himself of
his weak h-pawn (with the advance ...h5). White’s position is winning — the same evaluation is valid
with the black pawn on h7 and the white pawn on g5). The winning plan is a rook transfer to the 8th
rank followed by f5–f6+. If the black rook aims at the e5-pawn White defends it with the rook from
e8. J. R. Capablanca carried this plan through. However, as renowned rook endgame expert N.
Kopaev (and others) have demonstrated, the opponent made a number of instructive errors on the way
to the final outcome.
Correct is 61.Rd6!. On the d-file the rook will protect the king against side checks.
Black did not take advantage of White’s lack of precision by restoring equality by 61...Ra4!


62.Kf3 (62.Kg3 Ra3+ 63.Kh4 Ra4 64.f5 Ra5 65.e6 fxe6 66.fxe6 Kf6=) 62...Ra3+ 63.Ke4 Ra4+
64.Kf5 White missed the correct way: he has brought his king, not his pawn, to f5. He cannot win
anymore. 64...Rc4 65.Rb7 With the idea e6. 65...Kf8 (65...Kg8!? 66.e6 fxe6+ 67.Ke5 h5!=) 66.Rb3
(66.g5 hxg5 67.Kxg5 f6+! with a draw!) 66...Kg7 67.Re3 Rc6! 68.Ke4 Rc4+ 69.Kf3 Rc6

70.Ra3 (70.f5 Kf8 71.Ra3 Rc1 72.Ra8+ Ke7 73.f6+ Ke6 74.Re8+ Kd5 75.e6 Rc6!= according to N.
Kopaev) 70...f6 71.Ra7+ Kf8 72.Ke4 fxe5 73.fxe5 (73.Kxe5 Rb6=) 73...Rc1 74.Kf5 (74.Rh7 Re1+
75.Kf5 Rf1+ 76.Ke6 Rf4 77.Rxh6 Rxg4 78.Kd7 Re4!=) 74...Rf1+ 75.Kg6 (75.Ke6 Rf4=) 75...Rg1!
(75...Rf4? 76.Kh5 Re4 77.Ra6 Rxe5+ 78.Kxh6 Kg8 79.Ra8+ Kf7 80.g5+–) 76.Ra4 Rh1 77.Kf6 Rf1+


[78...Rg1? 79.Ra8+! (79.Rf4+? Kg7! 80.Kf5 Ra1=) 79...Kg7 80.Kd6!+–; 78...Re1 was also fine.]
79.Rf4+ Kg7!=. It becomes clear that playing the white king to f5 is not the correct winning plan.
White must advance his pawns under favorable circumstances by coordinating his king and rook.
A second consecutive inaccuracy.

White’s plan is to expel Black’s rook from the third rank, transfer his own rook to the seventh or
eighth rank and advance the f-pawn.

a) The appropriate method is the restriction of the black rook with 62.Rb1!, temporarily denying the
black rook the 1st rank then Black is in zugzwang. He must either worsen his king’s position or move
his rook off the e-file where it is best placed. In both cases the invasion of the white rook gains in
effectiveness. For example: 62...Re4 (62...Re2+ 63.Kf3 Rh2 64.f5 h5 65.Rb7 hxg4+ 66.Kg3 Rh5
67.Kxg4 Rh1 68.e6+–) 63.Kf3 Ra4 and now the time has come for the main plan:

64.Rb8! Ra3+ 65.Kg2 Re3 66.Re8! Re2+ 67.Kf3 Re1 68.f5 Rf1+ 69.Ke2 Rf4 70.Ke3 Rxg4 71.f6+
Kh7 72.e6!+–.
b) Note that 62.Rb8? ‘suggested itself’, however after 62...Re4 63.Kf3 Re1 the straightforward
64.Re8? enables salvation through 64...h5! 65.g5 (65.gxh5 Rf1+! 66.Ke4 Re1+ 67.Kf5 Rh1=)
65...Rf1+! 66.Ke3 (66.Kg3 h4+ 67.Kg4 h3=) 66...h4 67.Ra8 h3 68.Ra2 Kg6 69.Ra6+ (69.Rh2 Kf5=)
69...Kf5 70.Rh6 h2 71.Rxh2 Rxf4= according to N. Kopaev.
62...Rc3 63.Kf2?
The great Jose Raul Capablanca displays horrifying ignorance of this particular position. Correct is
63.Rb8, intending f5 (Ν. Kopaev), winning.


Black misses the chance to continue with 63...h5! (I. Levenfish & V. Smyslov), where he either trades
a pair of pawns or after 64.g5 (64.gxh5 Rh3=) 64...h4! obtains enough counter play to save the game:
65.Rb7 (65.Kg2 h3+ 66.Kh2 Rf3!=) 65...Kg6 66.Rb6+ Kg7! 67.Rh6 h3 68.f5 (68.Kg1 Rf3 69.Rh4
Kg6 70.Kh2 Kf5=) 68...Rc5 69.f6+ Kg8 70.e6 (70.g6 fxg6 71.Rxg6+ Kf7! 72.Rg7+ Kf8 73.Re7
Rc2+ 74.Kg1 h2+ 75.Kh1 Rf2=) 70...fxe6 71.Rg6+ Kf8 72.Kg3 Rc3+ 73.Kh2 Rf3 74.Rh6 e5
75.Rxh3 Rf5 76.Rh8+ Kf7 77.Rh7+ Kf8

78.Rg7 Rf3! 79.Kg2 e4=. A similar position has arisen in the game Rukavina-Grivas.


64.Rb8! of course!
Maybe Black should have tried 64...Ra2+!? but he would not survive anyway. It must be noted that
weaker was 64...Kg6 fails due to the decisive 65.f5++–. Black has lost his chances in an earlier stage
as we have already noticed and analyzed.

65.Rb8+! Kg7 66.f5
Threatening 66...-- 67.f6+ Kh7 68.Rf8 Kg6 69.Rg8+ Kh7 70.Rg7++– (R. Fine).
White’s king cannot be cut-off along the third rank e.g. 66...Rc3 67.f6+ Kh7 68.Rf8 Rc7


69.Kf3 (69.Kg3 Rb7 70.Kh4 Ra7 71.Kh5 Rb7 72.g5! hxg5 73.e6! fxe6 74.Re8 Rf7 75.Re7 Kg8
76.Kg6+–) 69...Rb7 70.Kf4 Ra7 71.Kf5 Rb7 72.Rxf7+! Rxf7 73.e6+– (I. Levenfish & V. Smyslov).
67.Kg3!? would win as well: 67...Ra3+ 68.Kh4 Re3 (68...Ra5 69.f6+ Kg6 70.Rg8+ Kh7 71.Rg7+
Kh8 72.Rxf7 Rxe5 73.Re7! Ra5 74.Re8+ Kh7 75.f7+–) 69.Re8 Re1 70.Kg3 Re4 71.f6+ Kh7 72.Kf3
Re1 73.Kf4 Rf1+ 74.Ke4 with the king headed to e7.
67...Ra3+ 68.Ke4 Ra4+


69.Kd5! Ra5+
After 69...Rxg4 70.f6+ Kh7 71.Rf8 Kg6 72.Rg8+ Kh5 73.Rxg4 Kxg4 74.e6 a new queen will appear
(R. Fine).
70.Kd6 Ra6+ 71.Kc7 Kh7
Alternatives such as 71...Ra7+ 72.Kb6 and 73.f6+ or 71...Ra1 72.f6+ Kh7 73.Rf8 Ra7+ 74.Kd8 Kg6
75.Rg8+ Kh7 76.Rg7+ Kh8 77.g5 (77.Ke8 h5 78.g5 h4 79.Rxf7 Ra8+ 80.Ke7 h3 81.g6! Ra7+
82.Ke6+– I. Lisitsin) 77...hxg5 78.Ke8 Ra5 79.Rxg5 Ra7 80.e6! fxe6 81.f7 Ra8+ 82.Ke7 Ra7+
83.Kf6 Ra8 84.Rh5# (R. Fine) were not help