Spanish Exchange
1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Bxc6 bxc6!?
 |
| The other recapture is much more popular, but the text has been played by grandmasters |
*****
*****
*****
*****
*****
*****
*****
5.Nc3!?
This is fourth-most popular in ChessBase's 2025 Mega database, behind 5.d4, 5.Nxe5!? and especially 5.0-0.
5...d6
Stockfish17 and Dragon1 like the rare 5...Qf6!?, albeit reckoning White is slightly better.
6.d4 exd4
 |
| How should White recapture? |
*****
*****
*****
*****
7.Nxd4
The engines prefer the less popular 7.Qxd4!?
7...c5!?
One of first-world-correspondence-champion Cecil Purdy's aphorisms - I do not have the reference to hand - is that it usually pays to advance the front member of a pawn-duo, and here the move is the top choice of Stockfish17, although Dragon1 prefers 7...Ne7!?
8.Nf3
The engines suggest an apparent-novelty, 8.Nc6!? meeting 8...Qd7 with 9.Na5 (not 9.Qd5?? Nf6).
8...Bg4!?
The position after 8.Nf3 is reached six times in Mega25, with six different moves played. The engines do not like the text.
9.h3 Bh5?
The bishop should retreat along the h3-c8 diagonal (even 9...Bxf3?! is better than the text).
10.0-0?!
White has at least the upper hand after this, according to the engines, but stronger is 10.Qd5 Nf6 (10...Bxf3? 11.Qc6+ Ke7 12.Nd5+ Ke6 13.Nxc7+ etc) 11.Qc6+ Nd7 (11...Ke7? 12.Bg5, with Nd5+ to come) 12.g4 (other moves are also good) Bg6 13.Nd5 Ra7 14.Bg6 f6 15.e5!? fxg5 16.e6, with a large advantage, according to the engines.
10...Bxf3?!
The engines suggest 10...Be7 or 10...Ne7!?
11.Qxf3 Qf6?
Hoping to relieve pressure by getting queens off, but the black queen becomes a target on f6 in a way that the white queen does not on f3.
12.Qe2
The engines reckon even stronger is 12.Bf4, with Nd5 or e5 to come.
12...Qd8?!
Black should probably cover the d5 square with ...Ne7 or ...c6.
13.Rd1!?
This seems best as it makes the pawn-push e5 strong.
13...Ne7 14.Be3
Despite Black's last move, 14.e5!? is still best, the idea being to meet 14...d5 by 15.Nxd5! Nxd5 16.c4 c6 17.cxd5 cxd5 18.e6, with a large advantage, according to the engines.
14...Qb8 15.Qd2
The text threatens 16.Bxc5! as 16...dxc5?? allows mate in one, but the engines still like pushing the e pawn.
15...Qb6 16.Bg5 f6 17.Nd5 Qc6
This is probably better than 17...Nxd5 18.Qxd5.
18.Bh4 Nxd5
The engines fluctuate between the text, 18...Ng6 and 18...Kf7!?
19.cxd5 Qd7 20.Re1+ Be7 21.Qe3
21.Re6 allows time for 21...0-0.
21...Kf7!? 22.Re2 Rae8 23.Qd3
 |
| How should Black proceed? |
*****
*****
*****
*****
23...Bd8!
The best move in a difficult position, according to the engines. After 23...Qb5?! (protecting the a6 pawn) 24.c4 Qd7 24.Rae1 Black is in big trouble, eg 24...g6 25.Re6 Bd8 26.Qe3 Rhf8 27.g4 with continuing intense pressure.
24.Qxa6 Rxe2!?
Possibly better is activating the queen with 24...Qf5.
25.Qxe2 Re8 26.Qd3 Qa4 27.Qxh7!?
25.Qxe2 Re8 26.Qd3 Qa4 27.Qxh7!?
This is good but perhaps 27.c4, protecting d5 and cutting the black queen off from the kingside, is simpler.
27...Re4!?
This wins back a pawn since White cannot save both the bishop and c2.
28.Bg3 Qxc2 29.Kh2!?
This is best (Black threatened, among other things, to win the white queen with 29...Re1+ etc).
29...Re2
 |
| White to play and maintain a winning advantage |
*****
*****
*****
*****
30.Qxc2?
This throws away almost all of White's advantage, whereas 30.Qh5+ Kg8 31.a4 Qxb2 32.Rb1! Qxb1 (forced) 33.Qxe2 wins easily, eg 33...Kf8 34.a5 Qa1 35.Qb5 c4 36.a6 c3 37.Bf4 c2 38.Qc6 leaves Black helpless.
30...Rxc2
BV thought for a few seconds after making this move, and then offered a draw.
 |
| How would you assess this rook-and-bishop ending? |
*****
*****
*****
*****
White is, at least temporarily, a pawn up, and has the better bishop, but Black's rook is very active, and Black will be able to free his queenside pawns by playing ...Ke7-d7 and ...c6. The engines at first reckon White has the better part of equality, but Dragon1 comes to call the position completely equal, and Stockfish17 very nearly does.
31.a4!? Rxb2 32.a5 Rb7 33.f4!?
The engines at first do not like this, but Stockfich17 comes to regard it as fine. Dragon1 slightly prefers 33.a6 Ra7 and 34.Bf4 or 34.h4.
33...Ke7 34.Be1!?
Again a move that the engines do not like at first, but it comes to be Dragon1's top choice, at least for a while.
34...Kd7 35.f5?
But the engines strongly dislike this, preferring 35.g4 or 35.h4!?, eg 35.g4 c6 36.dxc6+ Kxc6 37.h4 d5 38.g5 fxg5 39.fxg5 d4 40.a6 Ra7 41.Kg3, reckoning the position is equal, although it would still be difficult (for humans) to play.
The difference between the text and the engines' suggestions at move 35 is, I believe, that the text leaves White needing an extra tempo to create a kingside passed pawn.
35...c6 36.dxc6+ Kxc6 37.g4 d5 38.Kg2 d4 39.Kf3 Kd5!?
This allows the black rook to be trapped, but the resulting position is still winning for Black.
40.a6 Ra7 41.Bg3!?
This seems to be objectively inferior to, for example, 41.h4, but it complicates matters, whereas after 41.h4 c4 Black's winning plan is a lot more straightforward.
41...c4 42.Bb8 Ra8 43.a7
 |
| Black is effectively the exchange down, but Black's advantage is roughly the equivalent of being a rook up, according to the engines |
*****
*****
*****
*****
43...Bb6 44.h4 d3 45.g5 c3 46.gxf6 gxf6 47.h5 Kc4 48.h6 Bxa7!
This piece sacrifice is the only way to win as 48...d2 is met by 49.Ke2, and 48...c2 by 49.Bf4.
49.Bxa7
 |
| Black to play and win |
*****
*****
*****
*****
49...d2?
White has no answer to 49...c2.
50.Ke2 Re8+ 51.Be3 Kb3 52.h7 Kc2 53.Rd1
 |
| Black to play and draw |
*****
*****
*****
*****
53...Rh8?
Three moves draw: 53...Rd8, 53...Kb3 and 53...Rxe3+!?
54.Bxd2!?
Another bishop sacrifice, but there is nothing difficult about this one.
54...Rxh7
Or 54...cxd2 55.Rxd2+ Kc3 56.Rd7 etc.
55.Rc1+ Kb2 56.Rxc3 Rh2+ 57.Kd3 Rh3+ 58.Kc4 Rh4+ 59.Kd5 Rh2 60.Rd3 Kc2 61.Rd4!? 1-0
