Game of possible interest that includes a major piece ending

[Kevin Pacey]

I played the game below in the middle of the 2007 Canadian Open. It involved a line of the Classical Orthodox QGD that may be seen as controversial, depending on how one views Black's chances after 7...dxc4, should he wish to play for more than a draw. Indeed, the Classical Orthodox as a whole often gets chessplayers on their high horses in one way or another...

I had been told just before the game that my opponent had done very well in the CYCC, also played in Ottawa, and thus given that he was a junior who was apparently hot, I had no misgivings at all about sticking with my previously selected solid Black repertoire choices for this lengthy tournament.

Unfortunately, I had planned on winning lengthy games if necessary, but in this case, when the opportunity came, I found that due to a combination of aging, a slight sleep deficit, mediocre fitness (at best), accumulated fatigue from earlier rounds, and a desire to save energy for the next day, I didn't have the will to try to win a very tough endgame, the proper result of which I still do not know. By the end of this event I had accumulated too many draws, including this one, to be able to hope for a prize:

[Event "2007 Canadian Open"]
[Site "Ottawa"]
[Date "2007.07.10"]
[Round "5"]
[White "Zhou H"]
[Black "Pacey K"]
[Result "1/2-1/2"]
[ECO "D63"]
[WhiteElo "2016"]
[BlackElo "2271"]
[Annotator ""]
[Source ""]
[Remark ""]

1.d4 e6 2.c4 d5 3.Nc3 Be7 4.Nf3 Nf6 5.Bg5 0-0 6.e3 Nbd7 7.Rc1 dxc4 8.Bxc4 a6 9.a4 c5 10.0-0 cxd4 11.exd4 Nb6 12.Bb3 Bd7 13.Ne5 Bc6 14.Nxc6 bxc6 15.Re1 Nfd5 16.Bxe7 Nxe7 17.Ne4 Nbd5 18.Qd3 Nb4 19.Qd2 a5 20.Rc3 Nf5 21.Rd1 Qh4 22.Ng3 Nxg3 23.Rxg3 Rfd8 24.Rh3 Qe4 25.Re3 Qf4 26.Qc3 Nd5 27.Bxd5 cxd5 28.Rc1 Qb8 29.Qa3 Qb6 30.Qc3 Qb7 31.Qa3 Rdb8 32.Rec3 h6 33.Rc7 Qb4 34.Qf3 Rf8 35.Qf4 Qxb2 36.h3 Rab8 37.Ra7 Rb4 38.Rxf7 Qxc1+ 39.Qxc1 Kxf7 40.Qc7+ Kg8 41.Qxa5 Rxd4 42.Qb6 Rxa4 43.Qxe6+ Kh8 44.Qxd5 Ra1+ 45.Kh2 Rxf2 46.Qd8+ Kh7 47.Qd3+ Kg8 48.Qc4+ Rf7 49.Qc8+ Rf8 50.Qe6+ 1/2-1/2

Here is the final position:



After doing a database search for this material balance, with h- and g-pawns, I discovered more than a handful of examples in my biggest database of games, with either a win for the rooks, or else a draw, resulting (no wins for the queen, as one could likely expect). The wins for the rooks more often than not involved charging the king out of its shelter all the way down deep into the the other side's camp, so that queen checks would run out. However a win often seemed to depend on the pawn structure being certain ways. In any case, the side with the rooks always played on for any number of moves.

I don't happen to own an Encyclopedia of Chess Endings volume that might cover this type of ending, nor do I have other books that comment on it. Perhaps someone would know how to assess it, in the case of the above game, or in general. Regretably (or charmingly?) a tablebase can't be used, since more than 6 pieces are involved.

[Kevin Pacey]

Re: Game of possible interest that includes a major piece ending

To try to put the endgame in my previous post into some more perspective, the (old) book Basic Chess Endings by Reuben Fine states the following when it comes to positions with queen vs. two rooks:

Queen and Pawn are normally equivalent to two Rooks. This means that Q+P vs. 2 R's is drawn, but Q+2 P's vs. 2 R's is won, while with even Pawns or one Pawn to the good, the two Rooks will win.

Without Pawns the game is as a rule drawn. If the enemy King is confined to the edge of the board, the Rooks may win; if the Rooks are not adequately defended the Queen may win.

However, on the next page, later on follows:

With Pawns on the board, two R's and one Pawn always win, unless there are so few Pawns left that a perpetual check is possible. Thus when there is nothing else, two R's+P only draw; two P's are required to win.

A bit later follows:

The minimal position where 2 R's+P vs. Q win is shown in No. 593 (Steinitz-Pillsbury, Nuremburg, 1896) Yet even here Black cannot advance his Pawn by any straightforword series of moves: he must combine the possibility of the advance with a threat against White's K-side Pawns.

No. 593: Black wins (White to move)



A little later there follows:

The superiority of the two Rooks when the Pawns are even is seen quite clearly in No. 594 (Tchigorin-Janowski, Carlsbad, 1907). Here White's Pawns are isolated and cannot be supported by the [Queen].

No. 594: Black to play wins (Black to move)




All the above when taken together is a bit confusing, though example number 593 makes me strongly suspect that Fine would have supposed the ending reached in Zhou-Pacey (within my previous post) ought to be drawn. Yet I want to emphasize again that Basic Chess Endings is a (very) old book (the copy I have gives it as published 1941), even though much of it apparently has held up well over time (barring the occasional flaws, even with examples given throughout the book, that people have noted). However, I once overheard a Toronto-based IM saying that it was written as a rush job, which is a bit disconcerting.

[Kevin Pacey]

Re: Game of possible interest that includes a major piece ending

...the Classical Orthodox as a whole often gets chessplayers on their high horses in one way or another...

For those who are unaware, a case in point is the naming of the QGD "Orthodox" Defence by Dr. Tarrasch, who was making fun of QGD systems involving ...e6, ...Nf6 and ...Be7 (the good doctor naturally prefered his own method, know afterwards as the Tarrasch QG).

In the H.Zhou-Pacey game above, note that if White plays 13.a5, Black can play 13...Nbd5, the only move recommended by the books I've seen, but then White can virtually draw at will by trading off minor pieces on d5, establishing a practically even position with a Black pawn on d5, and with a slightly bad B vs. Kt, giving Black no real winning chances whatsoever.

The winning attempt that some people have tried in the event of 13.a5 is 13...Nc8, planning to recycle the Kt, though it may be debatable whether Black is then clearly slightly worse. The strong player Graf (formerly named Nenashev) has played this way in a similar position (i.e. with a R on a1 instead of c1, which if anything would be helpful to White since then a5 is guarded already).

I somewhat recently played the 7.Rc1 dxc4 line against Robert Hamilton, who followed the moves of H.Zhou-Pacey until move 12, preferring the less common 12.Bd3, a move that doesn't make my books. With both of us sleep deprived, Black's task as usual was the harder, and I quickly slipped up badly by allowing a d4-d5 break. In the post-mortem, Robert mercilessly bad-mouthed the QGD Classical Orthodox as giving Black no winning chances . Though I would have been satisfied with a draw in this particular game, I somehow didn't feel like giving him the satisfaction of hearing it.

[James Williamson]

Re: Game of possible interest that includes a major piece ending

I played the game below in the middle of the 2007 Canadian Open. It involved a line of the Classical Orthodox QGD that may be seen as controversial, depending on how one views Black's chances after 7...dxc4, should he wish to play for more than a draw. Indeed, the Classical Orthodox as a whole often gets chessplayers on their high horses in one way or another...

I had been told just before the game that my opponent had done very well in the CYCC, also played in Ottawa, and thus given that he was a junior who was apparently hot, I had no misgivings at all about sticking with my previously selected solid Black repertoire choices for this lengthy tournament.

Unfortunately, I had planned on winning lengthy games if necessary, but in this case, when the opportunity came, I found that due to a combination of aging, a slight sleep deficit, mediocre fitness (at best), accumulated fatigue from earlier rounds, and a desire to save energy for the next day, I didn't have the will to try to win a very tough endgame, the proper result of which I still do not know. By the end of this event I had accumulated too many draws, including this one, to be able to hope for a prize:

[Event "2007 Canadian Open"]
[Site "Ottawa"]
[Date "2007.07.10"]
[Round "5"]
[White "Zhou H"]
[Black "Pacey K"]
[Result "1/2-1/2"]
[ECO "D63"]
[WhiteElo "2016"]
[BlackElo "2271"]
[Annotator ""]
[Source ""]
[Remark ""]

1.d4 e6 2.c4 d5 3.Nc3 Be7 4.Nf3 Nf6 5.Bg5 0-0 6.e3 Nbd7 7.Rc1 dxc4 8.Bxc4 a6 9.a4 c5 10.0-0 cxd4 11.exd4 Nb6 12.Bb3 Bd7 13.Ne5 Bc6 14.Nxc6 bxc6 15.Re1 Nfd5 16.Bxe7 Nxe7 17.Ne4 Nbd5 18.Qd3 Nb4 19.Qd2 a5 20.Rc3 Nf5 21.Rd1 Qh4 22.Ng3 Nxg3 23.Rxg3 Rfd8 24.Rh3 Qe4 25.Re3 Qf4 26.Qc3 Nd5 27.Bxd5 cxd5 28.Rc1 Qb8 29.Qa3 Qb6 30.Qc3 Qb7 31.Qa3 Rdb8 32.Rec3 h6 33.Rc7 Qb4 34.Qf3 Rf8 35.Qf4 Qxb2 36.h3 Rab8 37.Ra7 Rb4 38.Rxf7 Qxc1+ 39.Qxc1 Kxf7 40.Qc7+ Kg8 41.Qxa5 Rxd4 42.Qb6 Rxa4 43.Qxe6+ Kh8 44.Qxd5 Ra1+ 45.Kh2 Rxf2 46.Qd8+ Kh7 47.Qd3+ Kg8 48.Qc4+ Rf7 49.Qc8+ Rf8 50.Qe6+ 1/2-1/2

Here is the final position:



After doing a database search for this material balance, with h- and g-pawns, I discovered more than a handful of examples in my biggest database of games, with either a win for the rooks, or else a draw, resulting (no wins for the queen, as one could likely expect). The wins for the rooks more often than not involved charging the king out of its shelter all the way down deep into the the other side's camp, so that queen checks would run out. However a win often seemed to depend on the pawn structure being certain ways. In any case, the side with the rooks always played on for any number of moves.

I don't happen to own an Encyclopedia of Chess Endings volume that might cover this type of ending, nor do I have other books that comment on it. Perhaps someone would know how to assess it, in the case of the above game, or in general. Regretably (or charmingly?) a tablebase can't be used, since more than 6 pieces are involved.

e-mail Pal Benko
pbenko(AT)uschess(DOT)org

[Kevin Pacey]

Re: Game of possible interest that includes a major piece ending

Thanks for the email address, though I wonder if Benko answers all such chess inquiries for free.

Benko co-authored [edit: with Hochberg] an interesting book Winning with chess psychology some years ago [edit: 1991], though I thought some of his pronouncements on Black's winning chances in a number of openings were either dated or debatable (e.g. French and Caro-Kann were deemed as permitting too much simplification if White wished, or, conversely, the Q-exchange line of the Classical KID was thought to have ample winning chances for Black, which is disputable by 2500+ vs. 2500+ player database results, or in the opinion of a number of modern GMs).

[Alan Baljeu]

Re: Game of possible interest that includes a major piece ending

Chiefly it's a matter of finding a viable strategy for black to attack, while shielding the king from perpetual.

Black wins if ever the rooks get traded for queen and pawn.

So hypothetical plan, put the K on h7, R on g6, and R on x2, forcing g4.(But where is white's king then?) I feel like white's going to fall apart after that. White's queen makes the whole process awkward, but that's it. Unless I'm missing some key idea.

[Kevin Pacey]

Re: Game of possible interest that includes a major piece ending

Chiefly it's a matter of finding a viable strategy for black to attack, while shielding the king from perpetual.

Black wins if ever the rooks get traded for queen and pawn.

So hypothetical plan, put the K on h7, R on g6, and R on x2, forcing g4.(But where is white's king then?) I feel like white's going to fall apart after that. White's queen makes the whole process awkward, but that's it. Unless I'm missing some key idea.

Even assuming that the hypothetical plan above is a key idea for Black to try to win, I suppose, without looking at a real board, White could try (after 50...Kh8) 51.Qe4, to prevent ...Kh7 for a least a while. And/or play h3-h4, introducing the possibility of h4-h5 if it's needed. However I haven't gone over this game for some time, so I too may be missing something in return.

[edit: Also, remember that not every 2P vs. 1P king & pawn ending that may arise is a win for the side with the extra pawn]

I may later post game(s) played between strong players (if any) with this endgame material balance, for comparison, after I track some down again.

[Kevin Pacey]

Re: Game of possible interest that includes a major piece ending

...
Here is the final position:



After doing a database search for this material balance, with h- and g-pawns, I discovered more than a handful of examples in my biggest database of games, with either a win for the rooks, or else a draw, resulting (no wins for the queen, as one could likely expect). The wins for the rooks more often than not involved charging the king out of its shelter all the way down deep into the the other side's camp, so that queen checks would run out. However a win often seemed to depend on the pawn structure being certain ways. In any case, the side with the rooks always played on for any number of moves.

My apologies, my memory played a bit of a trick on me, as some time has passed since I did the database search refered to above.

Often it was actually the defender's king who was caught well up the board, outside of its shelter, in the cases where the Rooks won, in games I looked up again in my database. There was one outstanding example of the side with the rooks running his king well up the board, but only to retreat all the way back into its shelter(!), and that game (Krnan-Stripunsky, Montreal(!) 2004) was 150 moves long, even though one of White's pawns dropped via a possible mistake by move 139.

Cases where the rooks were able to overpower White's g-pawn were not that common (in one case, a drawn pawn up king and pawn ending resulted, and another such pawn ending was lost), and draws were the most common result, with the players sometimes agreeing to them after not too many moves at all: Alekhine, and later modern GM Rowson, were involved in two such cases.

[Louis Morin]

Re : Game of possible interest that includes a major piece ending

Perhaps someone would know how to assess it, in the case of the above game

It seems to me that White can draw with eyes closed after 50...Kh8 51.h4 and advancing all his pawns as soon as possible. Just an example: 51...Raf1 52.h5 R1f5 53.g4 Rf2+ 54.Kh1 Rd2 55.g5 hxg5 56.h6 etc.

Of course I am not saying that this is best play for White, but only the fact that he can still draw after such silly moves goes a long way to show how this endgame is drawish.

[Kevin Pacey]

Re: Re : Game of possible interest that includes a major piece ending

It seems to me that White can draw with eyes closed after 50...Kh8 51.h4 and advancing all his pawns as soon as possible. Just an example: 51...Raf1 52.h5 R1f5 53.g4 Rf2+ 54.Kh1 Rd2 55.g5 hxg5 56.h6 etc.

Of course I am not saying that this is best play for White, but only the fact that he can still draw after such silly moves goes a long way to show how this endgame is drawish.

Critical for any Black winning attempt after 51.h4 could be to try to avoid a quick exchange of pawns, say by trying 53...Rg5 in the variation you gave.

I'll next post the possibly interesting (but lengthy!) game Krnan-Stripunsky that I noted in my previous post (which I edited a bit to add a couple of corrections, errors due to my medication-impaired memory combined with an obsolete computer that has my chess software on the other side of its partition :(:o).

[Kevin Pacey]

Re: Game of possible interest that includes a major piece ending

Here's the gamescore I promised in my previous post, along with a few diagrams that I added (after the gamescore, in case folks want to cut & paste it :)):

[Event "2004 It B"]
[Site "Montreal CAN"]
[Date "2004.11.08"]
[Round "7"]
[White "Krnan, T"]
[Black "Stripunsky, A"]
[Result "0-1"]
[ECO "B42"]
[WhiteElo "2331"]
[BlackElo "2553"]
[Annotator ""]
[Source ""]
[Remark ""]

1.e4 c5 2.Nf3 e6 3.d4 cxd4 4.Nxd4 a6 5.Bd3 Qc7 6.0-0 Nf6 7.Re1 Bd6 8.h3 Nc6 9.Nf3 h6 10.Nbd2 e5 11.Nc4 Bc5 12.a4 d6 13.c3 Be6 14.b4 Ba7 15.Be3 0-0 16.Rc1 Bxe3 17.Nxe3 Rad8 18.Nh4 Ne7 19.c4 a5 20.Nd5 Bxd5 21.cxd5 Qb6 22.bxa5 Qxa5 23.Qb3 Rb8 24.Qb5 Qd8 25.Nf3 Nd7 26.Nd2 b6 27.Nb3 Nc5 28.Qc4 f5 29.Nxc5 bxc5 30.exf5 Nxf5 31.Bxf5 Rxf5 32.Rb1 Ra8 33.Rb5 Qf6 34.f3 Rg5 35.Qe4 Rf8 36.Kh1 Rf5 37.Qc2 Qf7 38.Qd3 c4 39.Qe2 Rf4 40.Rc1 Qc7 41.Rc3 Rd4 42.Rb4 Rc8 43.Qc2 Kh8 44.Qc1 e4 45.Qg1 Rxd5 46.Rbxc4 Qxc4 47.Rxc4 Rxc4 48.fxe4 Rxe4 49.Qb1 Rde5 50.Qb8+ Re8 51.Qxd6 Rxa4 52.Qc6 Rea8 53.Kh2 R4a5 54.h4 Rf8 55.Qd6 Raf5 56.Kh3 R5f6 57.Qd5 Re8 58.Qd4 Rfe6 59.Qd7 Re3+ 60.Kh2 R3e4 61.Kh3 Rb8 62.Qd5 Re3+ 63.Kh2 Reb3 64.Qc6 R3b5 65.Qf3 h5 66.Qe2 Rf5 67.Qe4 Rbf8 68.Kh3 g6 69.Qd4+ Kd7 70.Qd7+ R8f7 71.Qe6 Rf4 72.Qe8 Rb7 73.Qe6 Rff7 74.Qd5 Rbe7 75.Qd6 Re3+ 76. Kh2 Re4 77.Kh3 Ra7 78.Qc5 Rd7 79.Qc2 Re1 80.Kh2 Re3 81.Qf2 Re4 82.Qf6 Re3 83.Qf4 Rde7 84.Qf8 Rb7 85.Qf4 Ree7 86.Qf6 Rf7 87.Qe6 Rbe7 88.Qc4 Rc7 89.Qe6 Rf4 90.Qe1 Rd7 91.Kh3 Rg4 92.Qf2 Kg7 93.Qe3 Rgd4 94.Qe5+ Kf7 95.Qh8 R7d6 96.Qh7+ Kf6 97.Qh8+ Kf5 98.Qf8+ Ke4 99.Qf3+ Ke5 100.Qg3+ Kf5 101.Qg5+ Ke4 102.Qe7+ Kd3 103 Qe1 Kc2 104.Qe2+ Kc3 105.Qe1+ Kb2 106.Qe2+ Kb3 107.Qb5+ Kc3 108.Qc5+ Kd2 109.Qa5+ Kd1 110.Qa1+ Ke2 111.Qb2+ Ke3 112.Qc1+ Kf2 113.Qc5 Kg1 114.g3 Kf1 115.Qc1+ Ke2 116.Qc2+ Ke3 117.Qc1+ Ke4 118.Qc2+ Ke5 119.Qc5+ Kf6 120.Qc3 Rd7 121.Qf3+ Kg7 122.Qe3 Rd2 123.Qe5+ Kh7 124.Qf6 Re2 125.Qf3 Rde7 126.Qf6 Re1 127.Qf8 Rd7 128.Qf6 Ree7 129.Qf3 Rf7 130.Qe2 Rde7 131.Qd2 Rf1 132.Kg2 Rfe1 133.Kf2 Rb1 134.Qd5 Rb2+ 135.Kf3 Ra7 136.Qc4 Ra3+ 137.Ke4 Rb7 138.Qc6 Rab3 139.Qf6 Rxg3 140.Kd5 Rg4 141.Kc5 Rgb4 142.Qd8 Rf4 143.Kc6 Rbf7 144.Kb6 R4f5 145.Kc6 R7f6+ 146.Kb7 Rf8 147.Qd1 R5f6 148.Qd5 Rg8 149.Qc5 Rg7+ 150.Kb8 Rff7 0-1

After 51...Rxa4:



After 113...Kg1:



After 138...Rab3:

[Wayne Komer]

Re: Game of possible interest that includes a major piece ending

Game of possible interest that includes a major piece ending

April 6, 2018

Final position in Zhou-Pacey




Krnan-Stripunsky has an error in the score, 69...Kh7 is the correct move

Position after 51..Rxa4



After 113...Kg1



After 138...Rab3



________

Basic Chess Endings Reference

No. 593 Steinitz-Pillsbury, Nuremberg, 1896



Black wins

No. 594 Tchigorin-Janowski, Carlsbad, 1907



Black to play, wins