A Trap in the Winawer

This morning I looked through a couple of games in Irving Chernev, The 1000 Best Short Games of Chess (1955). One of these was Booth--Fazekas, London 1940. The somewhat unusual manner of trapping Black's queen was fresh in my memory two hours later when I had the opportunity in a similar position. Naturally, I was curious whether this sort of trap appears often. It is not terribly common, but with so many games recorded daily in online play, it happens often enough. My annotations give a sense of the frequency.

Booth -- Fazekas [C18]
London, 1940

1.e4 e6 2.d4 d5 3.Nc3 Bb4 4.e5 c5 5.a3 Bxc3+ 6.bxc3 Qc7 7.Nf3 Nc6

Third most popular move.
7...Ne7 is most common.

8.Bd3

Black to move

8...cxd4

This capture does not get played by masters, but there are 23 games between lesser players in ChessBase Mega 2020. Also, there are 123 games in Lichess database, including several players of Black rated over 2300.

8...c4 9.Be2 seems appropos.

9.cxd4 Nxd4??

Loses the knight. ChessBase Mega has 18 games with this error; Lichess has 67.

10.Nxd4 Qc3+

White to move

Black has forked knight and rook and White can only defend one of them.

11.Qd2

Naturally, there are a number of games where White played Bd2, opting to give up the knight.

11...Qxa1? 12.c3!

The rook was poison. Five games in ChessBase Mega 2020 go this far. The highest rated Black player is 2153. Four games on Lichess--all players over 2000 on that site. Black appears to have resigned here as the queen is trapped. A game on Lichess continued 12...Qxc1+ 13.Qxc1 and then Black gave up.

1-0

In my game, my opponent did not blunder the knight. I still trapped the queen, but with less advantage. Alas, I moved the knight to the wrong square and let the queen back out.

Stripes -- Internet Opponent [C17]
Live Chess Chess.com, 22.09.2022

1.e4 e6 2.d4 d5 3.Nc3 Bb4 4.e5 c5 5.Nf3

5.a3 was Booth -- Fazekas.

5...a6

A rare move: 145 games on Lichess; 16 in ChessBase Mega 2020

6.a3 cxd4

6...Bxc3+ 7.bxc3

7.Nxd4 Bxc3+ 8.bxc3 Qc7

22 games between Lichess (20) and ChessBase (2) have reached this position.

9.Bd3

This move was a novelty in the position, but played in recognition of the trap.

9...Qxc3+

White to move

The position differs in particular ways from the London miniature. Compare the two diagrams.

10.Qd2!=

Booth's Qd2 gave a clear advantage

10...Qxa1??

10...Qxd2+ 11.Bxd2=

11.c3!

The queen is trapped.

11...Nc6 12.Nc2??

Lets the queen escape.

12.Nb3+-

Black to move

12...Qa2-+

I could have resigned here, but it was a rapid game and I was able to make some threats. My opponent needed several moves to get the queen back with its compatriots. In the end, facing a desperate check, my opponent moved the king to the wrong square handing me a mate in three. I did not fail a second time in this game.

Progressing Through

After announcing in "Try, Try, Again" my resolve to play through the whole of the games section in Chess Informant 152, I have felt a sense of responsibility. Initially, I was racing through the games as Jeremy Silman appears to advocate, but I slowed down (see "Quality"). My routine has developed into one that satisfies me. I play through the games quickly, but with enough attention that I can write a few words in the margins of the print text for later review. I have also extracted key positions from a number of games that became lessons for my students.

This morning I compiled some data on my progress.

The numbers on the spreadsheet are approximate and low. They were compiled mainly through ChessBase's history feature. They do not include games that I played through on Lichess, Chess.com, or Chessgames.com. I used all of these sites, as well as FICS Database while writing my article yesterday on a Greco game that I know well (see "Following Greco").

Since 7 September, I have been through the first 80 games (there are 201) in Informant 152. With annotations that are full and partial games, that includes at least the opening phase of 113 games. In addition, I have been through 133 other games from books or database search. I have also looked at a considerable number of my own games, although the vast majority was only a specific endgame position in the game.

The principal reason I have not successfully completed the project of playing through all games in any given Informant in the past is that other projects crowd in. The past few days I have again begun a project of playing through all of the games in Irving Chernev, The 1000 Best Short Games of Chess (1955).

Last night I downloaded a PGN file of the games in Chernev's book for playing through on my iPad. I play through a game and then read what Chernev wrote. I went through 40 games in this manner last night and another 10 this morning. I expect to create this week's chess lessons for my students from Chernev's collection off miniatures.

After starting with Chernev, I spent half an hour or so going through games 76-80 in Informant 152.

I go through each game without seeing the game score or annotations. Often I am trying to predict the moves. For some games, I then go through the game again with notation visible and play through some or all of the analytical variations. After recording a thought or two in the margins of my print copy, I move on to the next game.

The game Inarkiev -- Sakaev, Russia 2022, 152/80 received my comment, "reminds me of a line by Steinitz". I'm thinking of his comment that the king can take care of himself. The game was drawn.

After 27.Kf3

Following Greco

In the headnote to the game, Irving Chernev mentions four later publications, each with different players, that follow an eight-move miniature that he presents in The 1000 Best Short Games of Chess (1955).* It would seem that this particular Gioachino Greco composition frequently occurs in actual play.

The earliest extant record of this game appears in Greco's MountStephen manuscript (1623). Chernev claims that it appeared in a book published by Greco in 1619. Greco never published a book. Nonetheless, the 1619 date appears with this game in many places. Greco's work is known through slightly more than two dozen extant manuscripts, as well as a few that were once known and were copied before they disappeared. Greco's sojourn in Rome is known through four manuscripts that were produced in 1619 or very early in 1620. But, this game appears only in later MSS: MountStephen, mentioned above, and Paris (1625), as well as a few documents derived from Greco's MSS.**

The rest of Chernev's note withstands scrutiny. The note, preceding game 23, reads:

    In 1891 the Deutsches Wochenschach published a pretty game won by Deichmann. The same game appeared in the Deutsche Schachzeitung of 1917 with K.J. as the winner. Chess Review printed the score in 1935 with Pearsall the victor over X. In his book Lehrreiche Kurzpartien, Herr Benzinger gives himself credit for the win.
    These claimants and a few others will have to step aside though and make room for a fellow known as the Calabrese, Gioachino Greco.
    Away back in 1619 he published a book of his games. This is one of the brilliants which graced this early collection. (9-10)

The game score appears in descriptive notation in Chernev's text. Here I present it in algebraic.

1.e4 b6 2.d4 Bb7 3.Bd3 f5 4.exf5 Bxg2 5.Qh5+ g6 6.fxg6 Nf6 7.gxh7 Nxh5 8.Bg6#

I often present the checkmate in two to beginning students.

Three of Chernev's four references have been digitized and are readily accessible. The game appears in Deutsches Wochenschach (December 1891), 412. An extra set of moves is included that are absent from Greco's version.

The Deutsche Schachzeitung was also easily located (158-159). It shows the initials of the player of White as K.Sz.J. A note explains that the game had been played in Budapest and was copied from a Hungarian publication. In Chess Review (July 1935), Arnold Denker comments, "Very, very old, but always amusing; the Black player had ideas" (155). The player with White, Allen G. Pearsall, is sufficiently well-known that chessgames.com has six of his correspondence games and a win against Frank Marshall from a 20-game simul in San Diego. Game 31 in Chernev is another Pearsall miniature.

Josef Benzinger, Lehrreiche Kurzpartien (1938) has not been digitized and presented on the Web, so far as I have found, but a copy exists in the Cleveland Public Library.

The "few others" referenced by Chernev must number in the thousands. Several commentators on the game at chessgames.com claim they have played this miniature. 3...f5 appears 1393 times in the FICS Database. In the overwhelming majority, however, 6...Bg7 was played instead of moving into mate in two with 6...Nf6. After 6...Nf6, ten games have 7.g7+ (an error) and 19 have 7.gxh7+. That is still 19 instances of the miniature on the Free Internet Chess Server.

An astounding 22,410 games on Lichess have the nearly fatal 3...f5 and a much higher percent (~20%) of those that continue to follow Greco's line play 6...Nf6. Alas, 7.g7+ is played nearly as often as 7.gxh7+. A total of 818 games on that site conclude with Greco's 8.Bg6#.

If the Chess.com explorer allowed full archive access in that manner afforded by Lichess, surely there would be at least another thousand or so instances of this game. One of my youngest students memorized Greco's game earlier this summer, so the repetitions should continue.


*Please see "My First Chess Book" (2012) for a description of the importance of this book to my early development as a chess player.
**My source for the early records of this and other games is Peter J. Monté, The Classical Era of Modern Chess (2014), "Part II. Openings and Games of the Classical Era of Modern Chess".

Outside Passed Pawn

In Jeremy Silman, Silman's Complete Endgame Course (2007), he uses the term "the fox in the chicken coop" for the concept that I learned as "outside passed pawn". It is a skill that I routinely teach beginning students and require for completion of my Knight Award. The core idea that Silman explains--abandoning a passed pawn to create another--appears in the endgame book that I have had since my youth. Irving Chernev, Practical Chess Endings (1961) presents this position. Perhaps that is where I first learned the concept many years ago.

White to move

Chernev explains, "White abandons his passed pawn. Capturing it will keep Black busy on one side of the board, while White gets time to win on the other" (39). He gives the moves:

1.Kf5 Kh6 2.Ke5 Kxh5 3.Kd5 Kg6 4.Kc6 Kf6 5.Kxb6 Ke7 6.Kc7

When young players learn this simple idea, they have a requisite skill for navigating a great many pawn endings that will occur in their own games.

Chernev gives a second solution that employs a fundamentally different set of principles, but that leads to a faster checkmate. Hence, the position is less than ideal for my purposes. Even so, I had the exact position in a blitz game in July 2021 and played it according to book.

Silman's illustration is cleaner. While there are several winning moves that can be played, his solution appears to follow the top engine moves.

White to move

1.b5 Kb7 2.Ka5 Ka7 3.b6+ Kb7 4.Kb5 Kb8 5.Kc6 Kc8 6.Kd6 Kb7 7.Ke6 Kxb6 8.Kxf6 Kc7

White to move

9.Kxg5

Silman writes, "the rest is mindlessly easy" (67).

I would play 9.Ke7, which wins faster.

When teaching or testing students, I usually compose positions on the spot. But, perhaps it would be useful to have a printed sheet in my chess bag with some of the most instructive positions from books and game. Paul Keres, Practical Chess Endings (1974) offers one that is excellent. His term is "distant passed pawn" (55). As he notes, without the passed pawn that will be sacrificed (or exchanged for Black's passed pawn), White would be lost. 

White to move

1.f5 Ke5 2.f6 Kxf6 3.Kxd4 Ke6 4.Kc5 Kd7 5.Kxb5 Kc7 6.Ka6

Reviewing my own internet games stretching back to the late-twentieth century produced many positions where the basic idea of using a passed pawn to deflect the opponent's king from the main scene was required. In many cases, the correct move was not played.

I played it correctly in this position from the Internet Chess Club in 2001.

Black to move

38...b3! was the only winning move. My opponent resigned after 39.Kxb3 Kxd5 40.Kc3 Ke4 41.Kd2 Kf3

Thinking I was applying this idea, I managed to throw away the win in this game from 2001.

White to move

I played 45.h4 and succeeded in forcing Black's king to the h-file while gobbling Black's two remaining pawns with my king. However, Black's king was able to get back to the a-file in time. This position shows that calculation is necessary in pawn endings and sometimes there is more than one idea in play.

1.Kf4! is the only winning move.

If Black then goes after the a-pawn, White needs to win a queen vs. pawn ending where it is important that Black's pawn is two squares from promotion. If the Black king tries to get the the h-file, then the distant passed pawn is far enough away that White will be able to promote the a-pawn.

Queen vs. Three Pieces

Continuing with my resolve to work through all of the games in Chess Informant 152, I have settled into a pattern of 4-8 games plus those embedded as annotations in a morning session of 30-60 minutes. This morning, however, a single game fully captured my attention. The game was played in the second event of the 2022 FIDE Grand Prix just over a week after Russia's invasion of Ukraine.

The war affected the game and was the focus of the post-game discussion. Dmitry Andreikin won with Black against Alexander Grischuk. Andreikin said that he opted for the Sicilian Defense because he "would fall asleep" if the game reached a long endgame. The war disrupted his travel to Belgrade. Grischuk made a critical blunder and focused in the press conference on the situation in Ukraine and his disapproval of Vladimir Putin's decision to invade.

I have embedded a video of the press conference at the bottom of this article.

Grischuk's blunder made sense superficially, but provided Andreikin an opportunity to exchange his queen for a rook and two minor pieces. Bringing his remaining pieces to a position where they were all working together brought the game to an end. Milos Perunovic annotated the game for Chess Informant.

Grischuk,Alexander (2764) -- Andreikin,Dmitry (2724) [B48]
Beograd 152/55, 2022

1.e4 c5 2.Nf3 e6 3.d4 cxd4 4.Nxd4 Nc6 5.Nc3 a6 6.Be3 Nf6 7.a3 d6 8.f4 Qc7 9.Bd3 Be7 10.0-0 0-0 11.Kh1 Re8 12.Qf3 Bd7 13.Rae1 Rac8

White to move

It seems to me that both players have developed logically to reach this position and was surprised that it occurs in only one previous game in my databases: Radoszta,A. -- Giardelli,S, Villa Martelli 2013. In that game, however, it was with Black to move.

14.Qg3 Nh5 15.Qf3 g6 16.f5 Ne5 17.Qh3

Black to move

Grischuk stated in the post-game interview that his blunders in the game included missing Andreikin's move from this position and his move 20 as well.

17...Qd8! 18.fxe6

Perunovic offers some analysis of 18.Be2

18...fxe6 19.Nxe6 Qa5 20.Nd5 Bd8

White to move

21.Bh6??

Grischuk thought after the game that White was better after 21.b4. In Perunovic's Informant annotations, the main line is given as unclear, but White is better in all the other lines. There are sharp tactics.

On the surface 21.Bh6 seems a sensible way to defend the pinned knight with a checkmate threat, but Andreikin found the refutation.

21...Qxe1! 22.Rxe1 Bxe6 23.Qe3 Ng4 24.Qd2 Nxh6

Black has three pieces--rook, knight, and bishop for a queen and pawn. This imbalance provoked me to spend more time on this game.

25.Be2 Ng7-+ 26.Rf1 Nf7 27.Nf6+ Bxf6 28.Rxf6 Rc5 29.h3 Nh5 30.Bxh5 Rxh5 31.Qf2 Re5 32.Qb6 Re7 33.Rf4 g5 34.Rf2 Rxe4 35.b3 Bd5 0-1

It is an instructive game.

Quality

When I was just starting out (I was an instant chess addict!), I used game collections, tournament books, and eventually Chess Informants to zip through anywhere from 200 to 500 games a day, every day. That’s not too amazing when you consider that I only used 20 to 40 seconds a game.
Jeremy Silman, "Studying Master Games and Berkmaster's First Over-The-Board Tournament Battle," Chess.com (21 January 2014)

When I started my effort to quickly go through every game in the back half of Chess Informant 152, I went through the 39 games and game fragments classified as ECO A in two sessions of 20-30 minutes. That is much slower than Jeremy Silman claims to have been his practice as a young player, but much faster than my pace so far through ECO B. This morning I went through game numbers 40-45, which, including annotations, was nine whole games and one fragment. My wife says I was at my computer about an hour. Six minutes per game is still very fast and superficial, of course.

As I went through games 40-45 in Informant 152 this morning on my computer, I made notes in the margins of the print edition. Every game has at least one note. Two tactical positions are marked as possible material to use with my students.

From Lupulescu,C.--Nanu,C., Romania 2022, 152/44

White to move

From Zhang,Zhong--Li,Di, Hangzhou 2022, 152/43

White to move

I marked Erdos,V.--Babula,V., Corte 2022 152/41 as "pins and forks". The next game, Esipenko,A.--Moiseenko,A., Deutschland 2022, "use every piece".

Silman asserts that quantity of games develops knowledge of patterns--"positional patterns, tactical patterns, structural patterns, piece placement patterns, timing patterns."* He suggests that to become an IM or GM, a player needs to look at 100,000 games. 

My ambitions to become a titled player, if I ever had them, dissipated a few years prior to my 60th birthday. Nonetheless, I enjoy the process of learning and seek a range of approaches to this process. I know that I have played through many thousands of games over the years, many very quickly and others that I have studied for many hours. A few years ago, I played through every available game on chessgames.com played by Rezső Charousek, of which there are only 171. The process took a week. The biggest impact on my play came from detailed study of a game that he lost. It inspired me to try my hand with the opening in that game (see "Losing My Virginity with the Ponziani" [2014]). This study of Charousek's games was influenced by Silman's claims.

I have played through a substantial percentage of Paul Morphy's games and have used at least thirty regularly in my teaching. I have been through more games credited to Gioachino Greco than can be found in any database, including my own because my hard drive crashed a few months after entering the 168 variations in William Lewis, Gioachino Greco on the Game of Chess (1819). Several years later, I entered all of the games from Francis Beale, The Royall Art of Chesse-Play (London 1656), which contains 94 games, many of which are not in the databases most people use.

As I work through these games with the objective of getting through all soon enough that Informant 153 does not arrive first, I am pulled one way by Silman's assertions that quantity and speed are beneficial. At the same time, slowing a little seems to offer better prospects of quality.

*Jeremy Silman, "Snarky Silman Presents: Reader's Questions," Chess.com (14 January 2014).

Try, Try, Again

Reading chess books is an almost daily activity for me, but finishing one is rare. In 2021, my embrace of ebooks, especially Everyman Books in Chessbase format, altered an old pattern. Prior to 2021, I could count on my fingers the number of chess books that I had read from cover-to-cover. The first four months of that year, I ran out of digits. This was not due to amputation, but rather to finding that I could look at every move in every chess game in a book much faster and with more enjoyment by viewing it on my computer screen.

Nevertheless, I still have not played through every game in any issue of Chess Informant, even though I have been getting the ebook version since Informant 113, and the ebook and print edition together since 124. In "Determination" (2016), I wrote about my resolve to go through all of Informant 128. I failed.

A couple of days ago, I worked through all of the games classified ECO A that were published in the games section of Informant 152. I ran through the games quickly in two sittings and made few notes in the print edition. I noted in the margin next to Mamedyarov -- Keymer, Berlin 2022 152/22 that the game contained "interesting imbalances".

Starting the B section, I am moving slower. The first game in the section, Ali Marandi -- Hungaski, Saint Louis 2022 152/30 sent me to The Week in Chess for more games with 4.b4 against the Scandinavian. I went through 28 of the 141 games in that database this morning, then a few more in ChessBase Mega 2020. Emanuel Lasker appears to have been the first player to try the move, employing it in two simuls in 1905 and 1908. The only prior game published in Informant with 4.b4 was a rapid game played on chess24, Nakamura -- Firouzja 144/24, which White won in 16 moves.

After relishing the tactical blow that ended Maurizza -- Pelletier 152/35, I took a moment to create this post. Now, it is time for me to resume the effort to succeed where I have failed so many times in the past.