Endgame Books

As the decade of the twenty-teens comes to an end, I thought it appropriate to consider the value of endgame books. I posted a photo on Facebook of the shelf that holds my collection of endgame books, offering a witty remark about ends and beginnings. I have long been in agreement with Jose Capablanca, Chess Fundamentals (1934) that aspiring players need to begin with the endgame.

I thought it would be a simple matter to list the books on this shelf and one or two other places, and then write brief annotations about all or most. However, the limits of this simple project became clear. Before basic pawn endings, Capablanca presents checkmates. Should checkmate pattern books be on the list? Pandolfini's Endgame Course is listed below and begins with simple checkmates. Why exclude others? 

Then, there is the matter of ebooks. Listed below are a few that are available from Amazon. For beginning chess players, no other author offers clearer analysis and better diagrams that the self-published work of Rodolfo Pardi. And yet, I exclude from the list the many classic texts in my Google Books library that are the result of scanning rare books from libraries. For instance, Johann Berger, Theorie und Praxis der Endspiele (1890) is perhaps the originator of the genre. Why not list it?

So, the list below starts with one shelf of books, listing a few other that might be on that shelf if they fit. As print books, they are the core of the reference materials I turn to frequently while studying or teaching the endgame. Studying the endgame positions in GM-RAM: Essential Grandmaster Knowledge (2000) by Rashid Ziyatdinov, for example, sends me looking through other works because Ziyatdinov offers no solutions. Thomas Engqvist, 300 Most Important Chess Positions (2018) contains many endgames with explanations, but the books on the shelf offer greater detail. I study because I enjoy the process as much as for the benefits to my game.

Alburt, Lev, and Nikolay Krogius. Just the Facts!: Winning Endgame Knowledge in One Volume. New York: Chess Information and Research Center, 2000.

This book is part of a set of seven that claimed to bring secrets of Soviet chess training to an American audience. I have not read it, but the Chess Training Pocket Book that is also part of the series is one of very few chess books that I have read cover-to-cover. And, I've done so twice. 

Averbakh, Yuri. Chess Endings: Essential Knowledge. London: Everyman Chess, 1993.

When someone asks where they should start in their study of endgames, this book is the first that comes to mind. It is short, focused, and written by one of the all-time great chess teachers. Averbakh produced a longer set of books on the endgame that belong in any complete library. I have PDF copies that someone gave me.

Ban, Jeno. The Tactics of Endgames. Mineola, NY: Dover, 1997 [1963].

When I bought this book about the year it came out, my chess library still numbered in the dozens of volumes. I spent some time working through part of it and my play improved.

Barden, Leonard. How to Play the Endgame in Chess. Indianapolis, NY: Bobbs-Merrill Company, 1975.

This book looks good, although I have not spent any time with it.

Chernev, Irving. Practical Chess Endings. New York: Dover, 1961.

I bought this book as a teenager, spent minimal time on chess in my twenties, and returned to serious play in the 1990s after graduate school. For two decades, it was my only endgame book. I studied it a bit in the 1970s, and still pull it off the shelf from time to time. Many of the exercises are composed.

_______. Capablanca's Best Chess Endings. New York: Dover, 1968.

In 2012, after beating me in the match for title of Spokane City Champion, John Julian credited his study of this book for guiding him in an endgame we had played. I bought it later that year, if I recall correctly. In November and December 2020, I worked through ten of the sixty games with my students. Although, in truth, I pay little attention to Chernev's analysis. Capablanca often simply outplayed his opponent from positions that could have been drawn with best play, but errors were made. Good examples of practical play.

De la Villa, Jesus. 100 Endgames You Must Know: Vital Lessons for Every Chess Player, 4th ed. Alkmaar, The Netherlands: New in Chess, 2015.

If any book supersedes Averbakh's Chess Endings: Essential Knowledge, it is this terrific book by Jesus de la Villa. Chessable also has a course based on this book that I understand has been very useful for promoting its innovative approach. Before I had a copy, I watched a video of Magnus Carlsen being taken through the Chessable course. One hundred endgames seems like a reasonable beginning and the explanations are well-written.

Donaldson, John. Essential Chess Endings for Advanced Players. Dallas: Chess Digest, 1995.

I bought this book from the author in February at the only OTB tournament I played this year. I spent my morning coffee time with it much of the next few weeks. It offers good challenging instructive positions. 2020 became a much busier year when my teaching went online, and chess suffered until mid-summer.

Dvoretsky, Mark. Dvoretsky's Endgame Manual. Milford, CT: Russell Enterprises, 2003.

This book is not for beginners, and some have suggested that it is only for players substantially stronger than me (i.e. masters). Of course, I've beaten masters quite a bit in online blitz, and some in online correspondence. I've also held my own in two rook endgames with a FIDE Master in the 2008 Spokane City Championship. Credit time spent with this book. It is a standard reference that I use frequently. I have a newer edition in Kindle format on my iPad. The most recent edition has been revised by Karsten Muller because Dvoretsky is no longer with us. See "Pawn Endings Flash Cards".

Emms, John. Starting Out: Minor Piece Endgames. London: Everyman Chess, 2004.

I keep pulling this book off the shelf with some intent to study, but then doing something else.

Erwich, Frank. Endgame Tactics: Magnus Carlsen. Alkmaar, The Netherlands: New in Chess, 2018. ebook.

Frank Erwich has produced a series of short ebooks for New in Chess. The general pattern is to offer one hundred exercises from the games of the featured player. A diagram is presented with the solution on the next page. I work through a problem or two every now and then. I'm 45% through this one.

Fine, Reuben. Basic Chess Endings. New York: David McKay, 1969 [1941].

The formatting of this book makes this book hard to read. There, I have said it. For its day, it was a mammoth achievement. I mostly look at it when someone else has referred to it. Most recently, Harold van der Heiden referenced it in his Endgame Study Database in an endgame I was studying (see "Textbook Ending"). Dvoretsky proved more useful.

Fishbein, Alex. King and Pawn Endings. Macon, GA: American Chess Promotions, 1993.

Someone was raving about this book, and it seemed to offer insights helpful with one or two of the exercises in GM-RAM. I found a copy online and ordered it. I found it useful. I might spend more time with this book in the future. Unfortunately, it is out of print.

Flear, Glenn. Improve Your Endgame Play. London: Everyman Chess, 2000.

I have probably opened this book a few times, as well as the other two in the set. 

_______. Mastering the Endgame. London: Everyman Chess, 2001.

_______. Test Your Endgame Thinking. London: Everyman Chess, 2002.

Kasparian, Genrikh Moiseyevich. 888 Miniature Studies. Belgrade: BeoSing, 2010.

Genrikh Kasparian has another book on the endgame that I should buy, Domination in 2545 Endgame Studies. It is not cheap in a quality edition. These studies are challenging miniatures (problems with few pieces). When I first acquired a copy, I spent a bit of time most days for several weeks working through the problems. Terrific book, but not necessarily ideal for improving practical play.

Levenfish, Grigory, and Vasily Smyslov. Rook Endings. Dallas: Chess Digest, 1971.

Vasily Smyslov was a terrific endgame player. Positions that he played correctly appear in standard reference works, and you can bet the authors of these books studied this one. I dip into it from time to time.

Matanovic, Aleksandar, et al. Encyclopedia of Chess Endings (Queens). Beograd: Sahovski Informator, 1989.

Part of a five volume set. I picked this book as my prize after a blitz tournament at the Spokane Chess Club a few years ago. It is a terrific reference work.

Mednis, Edmar. Practical Rook Endings. Coraopolis, PA: Chess Enterprises, 1982.

This book is another that I bought from John Donaldson in February. I have not made the time to study it, but Donaldson told me the time would be rewarded.

Minev, Nikolay. A Practical Guide to Rook Endgames. Milford, CT: Russell Enterprises, 2004.

This little book belongs in a small collection of books like Averbakh's Chess Endings: Essential Knowledge. It goes to the heart of the subject in a few pages. terrific beginning point before getting lost in the rook endings in Dvoretsky.

Muller, Karsten, and Frank Lamprecht. Fundamental Chess Endings: A New Endgame Encyclopedia for the 21st Century. London: Gambit Publications, 2001.

This book has served me well as a companion to Dvoretsky's Endgame Manual. It is a terrific endgame reference work that sometimes explains matters more clearly than Dvoretsky.

Nunn, John. Nunn's Chess Endings, vol. 1. London: Gambit Publications, 2010.

When Borders Books opened in my city, they started by stocking nearly everything. Then, the local market affected the selection, and there was a diminishing supply of vital texts like chess books and the works of William Faulkner. In the last few months of its existence, these two were the only ones in stock that were worth having and I did not already own. Owning copies is one thing; reading them is another. Nunn's books are challenging.

_______. Nunn's Chess Endings, vol. 2. London: Gambit Publications, 2010.

Pandolfini, Bruce. Pandolfini's Endgame Course. New York: Fireside, 1988.

I worked through much of this book in a short period of time about twenty years ago, and many of the fundamental positions became central to my teaching. I gave away my copy as a prize at a youth chess tournament. A few years later, I acquired another copy.

Pardi, Rodolfo. Opposition and Critical Squares. Self-Published, 2014. ebook.

This book concerns one elementary chess endgame position and a few that are derived from it. The diagrams are the best I have seen in an ebook. This explanations are clear. Perfect choice for young players starting out.

Polgar, Laszlo. Chess Endgames. Koln: Konemann, 1999.

This is part of a set of three massive reference works that one might assume formed part of the curriculum for training the Polgar sisters. However, Susan Polgar has claimed online that at least one of them was her work when she was a teenager. Chess Training in 5334 Positions is the best known, and the most widely available. This endgame book has more than 4000 endgame positions organized by 171 themes. It has been a useful resource.

Seirawan, Yasser. Winning Chess Endings. London: Everman Chess, 2003.

During the broadcast of a grand master chess tournament a few years ago, Wesley So came into the broadcast booth with Yasser Seirawan to go over the game he had just won. Then they talked more generally. So credited the Winning Chess series that Seirawan wrote with having helped him when he was starting out. I find the story credible and think that Seirawan's books are all very good. They are not comprehensive, but well-written and the examples are quite instructive.

Shereshevsky, Mikhail. Endgame Strategy. London: Cadogan Chess, 1994 [1985].

I bought this book a few years ago from John Donaldson who always brings a batch of new and used books to a tournament in my city that he has attended twenty times or so. Most often, he wins the event. He sells these books at low prices. Shereshevsky's text has a good reputation, but my time with it has been too limited to offer much insight.

Silman, Jeremy. Silman's Complete Endgame Course: From Beginner to Master. Los Angeles, Siles Press, 2007.

Jeremy Silman structures this book as the endgames a player should know at different rating levels. The overall structure tends toward superficiality, which makes it the perfect resourse for those who do not want to spend much time studying. I find the positions that Silman lists for master and above where my current study takes me. I worked through the whole book up to my rating level in the space of a few hours the day I bought it. Now, it mostly gathers dust. Nonetheless, it has influenced me substantially in the material I choose to teach. Silman convinced me to cease teaching checkmate with bishop and knight as a regular practice, and he offers no instruction concerning that elementary checkmate in this book. He also convinced me of the importance of the Philidor and Lucena rook endings.

Smyslov, Vasily. Vasily Smyslov: Endgame Virtuoso. London: Everyman Press, 2003 [1997].

I used this book as supplementary material while studying some of Smyslov's own rook endings via Dvoretsky.

Stripes, James. Essential Tactics: Building a Foundation for Chess Skill. Self-Published, 2017. ebook.

This book was my first self-published book through Amazon. Previously, I had used the copiers and spiral binding services at Kinkos (now FedEx Office). It is not an endgame book, bit rather a primer on tactics aimed at beginning players that was inspired by the success that I found taking some young students through Bruce Pandolfini, Beginning Chess. Every position has ten or fewer pieces and a simple tactic, the idea I got from Pandolfini. I composed slightly more than 130 of the 150 exercises included here. However, after solving the tactic, one often finds an endgame where there may be a series of only moves. I offer reasonable explanations of these endings in the text.

Van Perlo, C. G. Van Perlo's Endgame Tactics, new ed. Alkmaar, The Netherlands: New in Chess, 2014.

The exercises in this book are unmatched and the quality of the analysis is first rate. However, this is more of a workbook than textbook. It offers not systematic study of endgame themes, but rather a generous collection (1300+) of problems to challenge and entertain.

Reti's Checkmate

The checkmate pattern know as Reti's Mate originates in a miniature played against none other than Saviely Tartakower. The game was played sometime in 1910 in Vienna. Both men were in their early 20s. Tartakower studied law in Geneva and Vienna; Reti came to Vienna to study mathematics. They had met in at least two tournament games prior to this game, which was a casual game played for a stake.

The mating combination itself has processors, including Morphy's Opera Game. Edward Winter's article, "Reti vs. Tartakower, Vienna 1910" is worth a look for some of these, as well as what little historical background exists. If Tartakower would have moved his king back to its starting square, the checkmate with rook and bishop would have been the same as Morphy's.

Reti,Richard -- Tartakower,Saviely [B15]
Vienna, 1910

1.e4 c6 2.d4 d5 3.Nc3 dxe4 4.Nxe4 Nf6 5.Qd3

This unusual move has been played a mere eight times in more than 3000 games reaching this position. This game is the earliest in MegaDatabase 2020 to feature this move. The same database contains fifteen games up to and including this one with 4...Nf6. In fact, the Caro-Kann Defense was relatively new at the time. Horatio Caro and Marcus Kann had published their analysis in 1886. Tartakower was born the following year; Reti two years after Tartakower.

Databases are incomplete. For example, we do not have all the games from the Second Trebitsch Memorial, played in Vienna in 1909-1910, which Reti won ahead of Tartakower. I have spent enough time reading old chess magazines from the last few decades of the nineteenth century to be confident that even published games played between those recognized as masters are more likely absent from databases than found there. 

Using the database as our point of reference, 5.Qd3 would be a novelty. But, there is a very good chance that Reti had seen it before.

5.Nxf6+ is played almost every time. 5.Ng3 is a distant second.

Black to move

5...e5?

Tartakower finds a clever idea that fails.

5...Nxe4 6.Qxd4 Qd5 appears equal.

5...Nbd7 6.Bd2 Nxe4 7.Qxe4 Nf6 8.Qd3 Bg4 9.f3 Be6 10.0-0-0 Qd6 11.Kb1 0-0-0 is given in an article on Chess.com by AksanAkhmad.

6.dxe5 Qa5+

6...Qxd3 seems reasonable, but is not the reason Tartakower played 5...e5. Perhaps he should have reassessed his plan. 7.Bxd3 Nxe4 (7...Ng4) 8.Bxe4 and White seems slightly better.

7.Bd2 Qxe5

It would seem that this was Tartakower's idea.

White to move

Reti's pinned knight is attacked twice.

8.0-0-0 Nxe4??

Tartakower has won a whole piece, but Reti saw further.

8...Be7 9.Nxf6+ Qxf6 (9...Bxf6 10.Re1) 10.Nf3 and White seems slightly better.

White to move

9.Qd8+! Kxd8 10.Bg5+

Tartakower resigned here, according to some accounts. Later, in his A Breviary of Chess (1937), he stated, "nothing could better illustrate the power of a double check" (as quoted by Winter, referenced above).

10...Kc7 11.Bd8# 1-0

If 10...Ke7, 11.Rd8#

The game is worthy of memorization, especially by chess teachers. It pairs nicely with the Opera Game.

Checkmate in One or Two

I am in the process of creating a checkmate exercise book aimed at beginning players. It consists of 152 checkmates in one organized by the piece that delivers the final blow. Then, another 152 exercises are checkmate in two from the same games. All 304 exercises are repeated in a series of 17 sets, but no longer organized in the sequence found the first time through. This book should be available from Amazon in early January both as a print text and in Kindle format.

Below is half of one of the sets. White to move. Checkmate in one or two.

1.

2.

3.

4.

5.

6.

7.

8.

Rubinstein -- Heilmann 1905

In October 2012, my wife and I loaded into the SUV our three month old puppies, and two dog crates, and headed to Eden Valley Guest Ranch. I also carried along a book that had recently arrived in the mail, Akiba Rubinstein: Uncrowned King (1994) by John Donaldson and Nikolay Minev. I wrote about this trip and the game annotated here after returning home (see "Practicing Visualization"). The Hauptturnier was a tournament for aspiring masters. Rubinstein scored 12 points in 15 games, finishing in a tie for first. He was then recognized as a master.

Rubinstein,Akiba -- Heilmann,Ernst [D40]
Hauptturnier-A Barmen (2), 1905

1.d4 d5 2.Nf3 e6 3.c4 c5 4.e3 Nf6 5.Nc3 Nc6

White to move

There are more than 9000 games with this position in Mega Database 2020, reached via several move orders.

6.a3

The main move has been known since Saint Amant played it against Staunton (see "Staunton Annotates").

6...b6

White scores over 70% after this move. Donaldson and Minev call it an error in Akiba Rubinstein: Uncrowned King (1994). Staunton criticizes his own play of this move, calling it "premature" (Chess Player's Companion, p. 342). Daniel Fiske notes this in reference to his own game in the First American Chess Congress, where it was played against him by Napoleon Marache. However, Fiske switches players, asserting that St. Amant played it against Staunton.

6...Ne4;
6...a6;
6...cxd4

7.cxd5! exd5

7...Nxd5 does not fare much better; White wins at least a pawn, according to Donaldson and Minev, citing Minev -- Morcken 1956 (Olympiad). Sherwin -- Chertkof 1957 New Jersey State Championship also offers an instructive reference game.

8.Bb5

First played by Saint Amant in the eleventh match game against Staunton

Black to move

8...Qd6

This move was a novelty at the time that worsens Black's position.

8...Bb7 was played in the two nineteenth century games referenced above. 9.Ne5 Rc8 10.Qa4 Qc7 11.Qxa7 Ra8

(Both players may have studied 11...Be7 12.Bxc6+ Bxc6 13.Qxc7 Rxc7 14.Nxc6 Rxc6 1-0 (50) De Saint Amant,P -- Staunton,H, Paris 1843)

12.Bxc6+ Qxc6 13.Qxa8+ Bxa8 14.Nxc6 Bxc6 1-0 (42) Fiske,D -- Marache,N, New York 1857. The game was published in the book of the tournament, edited by Daniel Fiske. 

8...Bd7 was recommended as an improvement by Saint Amant 9.dxc5 bxc5 10.0-0 Qa5 11.b4 Qb6 12.Ba4 cxb4 13.Nxd5 Nxd5 14.Qxd5 1-0 (46) Wallace,A -- Crane,B, Sydney 1893

9.e4!

This exciting move happens to be the engine's third choice.

9.Ne5 is Komodo's second choice. 9...Bd7 10.Nxd7 Qxd7 11.Qa4 Rc8 12.0-0 Be7 13.e4 Nxe4 14.Nxe4 dxe4 15.Rd1 a6 1-0 (35) Jovanic,O (2476) -- Grbac,B (2190), Porec 2011;

9.dxc5 bxc5 10.e4 is Komodo's top choice

9...Bd7

9...dxe4 10.d5 Nxd5 (10...exf3 11.Qxf3 Nxd5 12.Nxd5 Bb7) 11.Qxd5 Bb7 12.Qxe4+ Be7 13.Bf4 and Black resigned 1-0 Lundin,E -- Staehelin,A, Zuerich 1952

10.e5 Qe7 11.0-0

Black to move

11...Ng8

11...Ng4 12.Nxd5 Qd8 13.Bg5 Ne7

12.Nxd5 Qd8 13.Qa4

Other moves might be a little better if the engine should be heeded, but White's advantage is overwhelming in any case.

13.Re1 Be7 14.e6 fxe6 15.Nf4;
13.e6 fxe6 14.Nf4;
13.dxc5 Bxc5 14.Qa4

13...Rc8 14.Bg5 14...Nge7

White to move

There is something about this position that makes me think of Morphy's Opera Game.

15.dxc5

15.Nc3

15...bxc5 16.Rad1 a6 17.Qxa6 Nd4

17...Nb8 18.e6 Nxa6 19.exd7+ Qxd7 20.Bxd7+ Kxd7 21.Nxe7+

18.Nxd4 cxd4 19.Rxd4 1-0

Rubinstein's play was exemplary. I return to this game from time to time and marvel at how Heilmann's game fell apart so quickly.

Staunton Annotates

The game below was the eleventh game in the 1843 match between Howard Staunton and Pierre Charles Fourier de Saint Amant. The match was held in Paris over the course of six weeks in November and December. An earlier match in London had been won by Saint Amant 3 1/2 - 2 1/2.

This game came to my attention as I was studying Rubinstein,A. -- Heilmann,E., Barman 1905.

I transcribed the notes from Howard Staunton, The Chess-Player's Companion (London 1849). Nick Pope also has transcribed these notes, as well as notes on the game from another publication by Staunton, and Saint Amant's notes from Le Palamède (see Chess Archaeology)

De Saint Amant,Pierre Charles Fourier -- Staunton,Howard [D40]
Match Staunton-Saint Amant, 02.12.1843

1.d4 e6 2.c4 d5 3.e3

"It is not advisable for the opening player, in games of this description, to carry his Q. B. over to the K's side, on account of the attack the adversary obtains by playing his Q. to her Kt.'s 3rd." (Staunton)

3...c5 4.Nc3 Nf6 5.Nf3 Nc6

White to move

This symmetrical position has occurred in more than 9000 games since the historic 1843 match. I've had it myself fifty times or more, playing both sides. Symmetry generally slightly favors the player with the move.

6.a3 b6

"The importance and utility of this move, when made at the proper moment, have been indisputably proved, we think, by the earlier games; it is evidently premature, however, and made without reflection here." (Staunton)

6...cxd4 7.exd4 would be my suggestion. But, even 6...a6, maintaining the symmetry, is better than Staunton's move. Heilmann repeated this folly in 1905 and lost a miniature to Rubinstein.

7.cxd5 (5)

The number in parentheses indicates the amount of time Saint Amant spent thinking on this move. Captain Harry Wilson accompanied Staunton on the trip to Paris and kept a log of the time used by each player in games 2 through 15. Staunton published lists of move times of every move where five minutes or more were spent. These times appear in the annotated game here.

7...exd5 8.Bb5

Black to move

8...Bb7

"We agree with M. St. Amant in believing that Black would have avoided the loss of a Pawn by playing the B. to Q's 2nd, instead of the Kt.'s 2nd." (Staunton)

Staunton is referencing Saint Amant's annotations in Le Palamède,

8...Bd7 9.Ne5 Nxe5 10.dxe5 Bxb5 11.Nxb5 Ng8 and White won in 38 moves, Correa,A (2345) -- Van Riemsdijk,H (2435) Brasilia 1994.

9.Ne5 [5] Rc8 [5] 10.Qa4 Qc7 [5] 11.Qxa7 [8] Be7

White to move

Staunton spent 8 minutes on Be7, and commented, "He does not appear to have any better move."

12.Bxc6+

Saint Amant initiates a series of exchanges that seem to me to relieve the pressure on Black's position, but also leaves White a pawn ahead. In a comment on the next game, Staunton references a remark in another of his publications, the Chess Player's Chronicle:

“As M. St. Amant, during the latter games of the match, played mainly to draw, and Mr. Staunton solely to win, the former estimating a remise as a victory, and his opponent looking on it as a defeat, it frequently happened that, while M. St . A. sought eagerly for every opportunity of exchanging Pieces, Mr. S. sacrificed position and occasionally the game itself to prevent him.” (C.P.C., vol. v., p. 44) 

I considered several alternatives for White:

a) 12.Qa4 0–0 13.Nxc6 Bxc6 (13...Ra8?? loses a piece 14.Nxe7+ Qxe7 15.Qd1) 14.Bxc6 Qxc6 15.Qxc6 Rxc6;
b) 12.0–0 0–0 13.Bxc6 (13.Nxc6 Bd6 (13...Bxc6 14.Qxc7 Rxc7 15.a4!) 14.Na4 Bxh2+ 15.Kh1 Bd6 16.Qxb6+–) 13...Bxc6 14.Qxc7 Rxc7;
c) 12.Bd2 0–0 13.Bxc6 Bxc6 14.Qxc7 Rxc7 15.Nxc6 Rxc6 16.f3

12...Bxc6 13.Qxc7 Rxc7 14.Nxc6 Rxc6 15.0–0

Black to move

15...Kd7 (5)

"This is far better than castling, because it brings the King more immediately within the sphere of his Pieces' operation." (Staunton)

16.Rd1 c4 17.f3

"Well played, and a difficult move for White to answer."

Staunton's comment does not make sense. Perhaps he means that it is difficult for Black to answer, or perhaps a printing error appended the comment to the wrong move. Perhaps it should appear in reference to Black's 16...c4. Nick Pope gives only "well played" after White's move.

17...Rd8

White to move

18.Bd2 (6)

"Better, we should have thought, to advance the K. P. at once." (Staunton)

18...h6 19.Na2 g5

Here Staunton praises his own play: "By the vigour and boldness of the attack on this side Black nullities the inequality of force."

20.Bb4 Re6

"With the intent to double his Rooks, when circumstances require him to do so." (Staunton)

21.Bxe7 Rxe7 22.Kf2

Black to move

22...g4 23.Nc3 h5 24.Re1 Rde8 (5) 25.Re2 h4 26.Rae1 Rg8

White to move

27.e4 (8) 27...g3+

"Compelling the King to retreat again." (Staunton)

28.Kg1 dxe4 29.Nxe4 Nxe4 30.fxe4

"Taking the Kt. with the R. would possibly have been better." (Staunton)

30...Rg4 (6) 31.hxg3 hxg3

White to move

32.Re3

"This was an important precaution lest Black should double his Rooks on the K. R's file. It also enables White, at the fitting moment, to place his R. on the K. B's 3rd." (Staunton)

32...b5 33.R1e2 (7)

"To afford his K. an opportunity of passing over to the Q's side." (Staunton)

33.Kf1 f6 (33...Rf4+ 34.Rf3) 34.Ke2?? obviously fails.

33...Re8 15 minutes--Staunton's longest think of the game.

White to move

34.Kf1 15 minutes--Saint Amant's longest think of the game

34...Kd6 35.Ke1 (7) 35...Rf4 (10) 36.Kd2

"By taking the Kt. P. he must have sacrificed his Pawns in the centre." (Staunton)

For instance, 36.Rxg3 Rexe4 37.Rxe4 Rxe4+ 38.Kf2 Rxd4 39.Ke2 and the position seems equal to me.

36...f5

White to move

37.Rxg3 (10)

Better might have been 37.exf5 Rxe3 38.Rxe3 Rf2+ 39.Kc3 Rxg2 when White has a slight advantage.

37...Rfxe4 38.Rxe4 Rxe4 39.Kc3 Kd5 40.Rf3 f4

White to move

41.g3 (5)

"From this point Black has a 'forced won' game." Staunton is correct. Saint Amant just blundered away the game.

41...Re3+ 42.Rxe3 fxe3 43.Kc2

"He has nothing better to play." (Staunton)  

Black to move

43...Ke4

And Staunton has just blundered the game back. 

 "We have here a second instance where a momentary relaxation loses Mr. S. the honours of a well-won victory when just within his reach. To any one who has never undergone the punishment of playing an arduous game of many hours' duration in a densely crowded room, such a lapse as Black's in the present, or in Game IX, may well appear inexplicable, but those players who have, will know the difficulty of keeping up the preternatural tension of the mind required so long, without a disposition to relieve it by one hasty move. The want of reflection in the present cases must be admitted to have been mortifying enough. Of the twelve first games these two, (and these actually given him,) were the only two games scored by the French player, consequently, but for them, he would actually have lost the match without winning a single game!" (Staunton)

Staunton presents the following variations in The Chess-Player's Companion

43...Kxd4 44.Kd1 (44.g4 Ke4 45.Kd1 (45.g5 Kf5) 45...Kf3 46.Ke1 Kxg4) 44...Kd3 45.g4 e2+ 46.Ke1 Kc2 47.g5 Kxb2 48.g6 c3 49.g7 c2 50.g8Q c1Q+ 51.Kxe2 Qc4+

44.Kd1

And now White has a winning ending.

Black to move

44...Kd3 45.d5 e2+ 46.Ke1 Kc2 47.d6 Kxb2 48.d7 c3 49.d8Q c2 50.Qd2 1-0

"And after a few moves Black resigned." (Staunton)

After this game, Staunton led the match 8-2. Game three had been drawn. Over the next ten games, Saint Amant won four, Staunton won three, and four games were drawn.

Textbook Ending

 In today's Morning Membership tournament on Chess.com, two players found themselves in a difficult pawn ending with under a minute remaining. The time control was game in five minutes with a two second increment (5+2). White had 18.5 seconds and Black had 38 when Chris started showing the game on Twitch. Chris Bird is the tournament director and streams these weekly events.

Both players shuffled their kings back and forth until a repetition occurred. When players are premoving, as they seemed to be doing, draw offers are often refused even as they are made. It was not clear from Chris's discussion whether he knew that White had a winning position.

In the screenshot, White has just played 51.Kd4.

There was an alternative.

51.Kb5 wins.

After 51.Kb5!

This move can be frightening with less than twenty seconds left, but it is the correct move. White has stepped out of the square of Black's pawn. If White fails to promote first, or set up a checkmate, Black will also promote.

After 51...Kc8, White has two ways to proceed.

52. Kc6, or 52.a6. I played the latter against Stockfish 11, which replied 52...e3.

White to move

Now, only one move wins for White. All others lose.

After 53.b6+, 53...Kc7! leaves White lost. However, 53...Kb8 allows 54.Kb6 and 55.a7#. In one of the textbook lines beginning with 51...Kb8, Black promotes the pawn just before getting checkmated.

53.a6 allows 53...Kb7 and White can make no further progress, while Black's e-pawn cannot be stopped.

53.Kc6! e2 54.b7+

Black to move

54...Kb8 55.Kb6 e1Q Black promotes first, but White checkmates with pawns. Stockfish chose a longer checkmate.

54...Kd8 55.b8Q+ Ke7 56.Qe5+ and White wins.

While watching the Twitch stream, I saw a familiar pattern. After the tournament, I out that my belief that White was winning was indeed correct. Although as a practical matter, with less that twenty seconds remaining, I, too, might have taken the draw.

The textbook position appears in Dvoretsky's Endgame Manual. Consequently, it is a position that I have played out several times with my students (see "Pawn Ending Flash Cards").

White to move

I know the technique through training with positions from Dvoretsky. Likely other books have the position as well, and it appears in Harold van der Heiden's Endgame Study Database. Heiden credits Reuben Fine, Basic Chess Endings (1941), while Dvoretsky credits the nineteenth century composers J. Kling and B. Horwitz.*

In any case, the first critical position is reached after the moves:

1.Kd4 Kg4 2.h4 Kh5 3.Ke3 Kg4 4.Ke4 Kh5 5.Kf4 Kg6 6.g4 Kg7

White to move

7.g5 is a mistake. White can move the king, but the way forward is 7.h5!

7...Kh6 8.Kf3

Heiden has 8.Ke4 here. Dvoretsky's solution is three moves longer, so seems to offer better defense by Black. Both 8.Kf3 and 8.Ke4 have the same number of moves to mate, according to my chess engine.

Dvoretsky points out the importance of triangulation for White.

8...Kg5 9.Ke4 Kh6 10.Kf4 Kh7 11.g5 Kg7 12.g6 (12.h6 loses) 12...Kf6 13.Ke4 Kg7 14.Kf3 Kf6 15.Kf4 Kg7

White to move

Now, White's confidence comes into play, as it is time to leave the square of Black's passed pawn.

16.Kg5 c3 17.h6+ Kg8 18.Kf6 (see the critical Kc6 in my analysis of what might have been above).

18...c2 (so close to becoming a queen!)

19.h7+ Kh8 

White to move

20.Kf7 c1Q

Black promotes, but White will promote with check.

21.g7+ Kxh7 22.g8Q+ Kh6 23.Qg6# 

This ending is worthy of study. Perhaps if your nerves are calm, you can win with a quarter minute and a two second increment should you find yourself is a comparable position.


*Fine, on my reading presents a similar position that is a draw. He does mention winning chances, though, if the pawns on the left are another row to the right of the position illustrated here. There seems some room for some historical research.