Spanton (1904) - Aakarsch Boswan (1982)
Caro-Kann Tartakower
1.e4 c6 2.d4 d5 3.Nc3 dxe4 4.Nxe4 Nf6 5.Nxf6+ exf6 6.Be3!?
This is fourth-most popular in ChessBase's 2026 Mega database, but is liked by Stockfish17.1, along with the commonest continuation, 6.c3. Dragon1 likes 6.c3 and 6.h3!?
6...Bd6 7.Bd3 0-0 8.Nf3 Re8 9.0-0
 |
| Now both sides have castled, how would you assess the position? |
*****
*****
*****
*****
Appearances can be deceptive. White has the only pawn on the two central files, which is also the only pawn on a fourth rank, and White, unlike Black, has an undamaged pawn-majority, but the engines reckon the position is equal. Wikipedia tries to explain this paradox by explaining that ...exf6 "offers Black rapid development," which is true, but there is no reason why Black's development should be any more rapid than White's.
9...Nd7 10.c4 Nf8 11.Re1 Bg4 12.Be2!?
 |
| This position occurs 13 times in Mega26, but in each case with White to move, when the engines reckon White is slightly better |
*****
*****
*****
*****
12...Qd7
White's loss of a tempo gives Black equality, according to the engines, and the text is Stockfish17.1's top choice, although Dragon1 suggests 12...Ne6 or 12...Bb4!?
13.Qd2 Ng6 14.Rad1 Qf5
 |
| How should White proceed? |
*****
*****
*****
*****
15.c5!
The engines agree this is the only way to avoid a disadvantage.
Disastrous would be 13.h3? Bxh3! as 14.gxh3? Qxh3 wins for Black thanks to the threatened ...Nh4. Also losing is 15.Bd3? Qh5, when White has no good answer to 16...Bxf3, while the immediate 15.d5?!, which, like the text, also aims for central counterplay, is well-met by 15...c5.
15...Bc7 16.d5 cxd5 17.Qxd5 Qxd5 18.Rxd5
 |
| Now the queens are off the board, how should Black proceed? |
*****
*****
*****
*****
18...Nf4
Black still has kingside pressure.
The main alternative is 18...Be6, when 19.Rd3 Bxa2!? wins a pawn, but 20.b3 traps the bishop. However, after 20...Rad8 21.Rxd8 Rxd8 22.Bc4 Black can try 22...Ne5 or 22...Be5, with an unclear position that Stockfish17.1 reckons gives equal chances, but Dragon1 reckons favours White.
19.Rd2 Nxe2+ 20.Rdxe2 Bxf3!? 21.gxf3
 |
| How would you assess this late-middlegame? |
*****
*****
*****
*****
White's pawn-minority has been smashed, but that is probably not as serious as it looks as White will in any case be defending on the kingside. Meanwhile White has a healthy farside pawn-majority, which for quite some time Dragon1 reckons gives White an edge, but the engine comes to agree with Stockfish17.1 that the position is equal.
21...Re6 22.Bd2 Rae8 23.Re4 f5 24.R4e3 Be5 25.b4 Bd4 26.Rxe6 Rxe6 27.Kg2?!
Best may be 27.Kf1.
27...Kf8?!
The engines give Black the upper hand after 27...Rxe1 28.Bxe1 Kf8.
28.Rxe6 fxe6
 |
| What should White play? |
*****
*****
*****
*****
29.f4!
This puts a fifth white pawn on a dark square, which is normally not a good idea in an ending when both sides have dark-square bishops. However, the engines agree f4 is the best move in the position, one point being 29.Kf1 is well-met by 29...e5.
29...Ke7 30.Be3!?
 |
| How should Black proceed? |
*****
*****
*****
*****
30...Bxe3!?
The engines reckon this results in a completely equal pawn ending, whereas 30...e5!? gives Black at least a slight edge, eg 31.fxe5 Bxe3! (31...Bxe5 is only equal, according to the engines) 32.fxe3 Ke6 33.Kf3 Kxe5, when Black's more-active king makes the black position easier to play, but the game should probably be drawn.
31.fxe3 Kd7?
The engines suggest 31..a6 or 31...b6.
 |
| White has one winning move - everything else loses |
*****
*****
*****
*****
32.b5!
The plausible 32.Kf3 Kc6 33.a4 loses to 33...a5 (33...Kd5 also wins), while 32.a4 Kc6 leads to similar results.
32...e5!?
Standing pat with, say, 32...Ke7 33.Kf3 Kd7 lets the white king manoeuvre to the queenside and force through a passed pawn, eg 34.Ke2 h6 35.h4 g6 36.Kd3 etc.
33.fxe5 g5
Or 33...Ke6 34.c6 bxc6 35.bxc6, when Black cannot capture on e5.
34.Kf3 h6 35.h3 h5 36.a4 a6!?
Black is lost anyway, eg 36...g4+ 37.hxg4 fxg4 38.Kf4 Ke6 39.a5 Kd7 40.c6+!? bxc6 41.b6 etc.
37.c6+ bxc6 38.bxa6 1-0
The match was drawn 4-4.
My Battersea Season 2025-6
Date Event Colour Rating Opp's Rating Score Season's Perf
16/9/25 CLL W 1936 1952 0 1552
18/9/25 CLL W 1936 1797 = 1675
16/10/25 CLL W 1889 2047 0 1665
23/10/25 CLL B 1889 1823 1 1805
18/11/25 LL W 1891 1980 = 1840
20/11/25 CLL W 1891 1866 1 1911
25/11/25 BCC B 1891 1817 1 1955
9/12/25 LL W 1904 1982 1 2008
CLL: Central London League
LL: London League
BCC - Battersea club championship
Nice win! For me its still annoying when Black plays 4.....Nf6.And especially if black recaptures with the e pawn.White has to take on f6.other moves are less.I think you only can avoid this variation when you know what your opponent plays in the Caro-Kann.
ReplyDeleteThe first 11 times I played 5.Nxf6+, four opponents replied 5...exf6, and seven chose 5...gxf6. In the 19 games since then, 18 opponents have played 5...exf6, and only one 5...gxf6. Books used to criticise ...exf6 for giving White a built-in endgame advantage, while praising ...gxf6 for offering Black dynamic chances. Now the latter is criticised for being weakening, and the former praised for giving Black active play. My results when facing the two moves have been just as starkly contrasting: I have scored 96% against ...gxf6, 43% against ...exf6.
DeleteThat is a very high score on gxf6.
ReplyDeleteProberbly is the reason for this that white got more tactics play?
Maybe I shall try 6.g3 after 5...exf6 and see how that will be developed.
Every opponent bar one was lower rated than I was at the time. Even so, I cannot complain about scoring +7=1-0.
DeleteAfter 5...exf6 I have tried, in decreasing order of frequency, c3, Be3, Nf3, Bf4 and Bd3. Let me know how 6.g3 fares ...