Spanish Exchange
1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Bxc6 dxc6 5.d4 exd4 6.Qxd4 Qxd4
Grandmaster Spyridon Skembris, who is available for postmortems, suggested 6...Bg4!?, after which Black scores an excellent 57% in ChessBase's 2023 Mega database. After the much-more popular text, Black scores 52%, which is less impressive but still very good.
7.Nxd4 c5 8.Ne2 Be6!?
Much more popular is 8...Bd7, but Stockfish15.1 and Komodo14.1 regard the moves as virtually equally good.
9.Nbc3 0-0-0 10.Be3 Nf6 11.f3 h6!?
Presumably played to prevent the pinning Bg5, and to support ...g5 in some variations to open lines, particularly if White castles kingside.
12.Nf4 Bd7 13.Ncd5 Nxd5 14.Nxd5 Be6 15.Nf4 Bd7 16.0-0-0
 |
| Now both sides have castled, how would you assess the position? |
*****
*****
*****
*****
It is a battle of pawn-structure versus bishop-pair - an equal battle, according to the engines.
16...Bd6 17.Nd5 Rhe8 18.Bf4
I rejected getting rid of the bishop-pair by 18.Nb6+ as Black gets strong drawing chances with opposite-coloured bishops, eg 18...cxb6 19.Rxd6 Bc6.
 |
| How should Black proceed? |
*****
*****
*****
*****
18...Bxf4!?
This is Komodo14.1's top choice, until it eventually prefers 18...Be6 by a tiny margin. Stockfish15.1 suggests 18...b5!?, although 19.Bxd6 cxd6 leaves Black with a backward d pawn.
19.Nxf4 f5 20.exf5 Bxf5 21.Rxd8+ Rxd8 22.Re1 g5
H-JM offered a draw, in my time, albeit not long after pressing the clock.
23.Nh5
This is Stockfish15.1's top choice. Komodo14.1 marginally prefers Grandmaster Skembris's suggestion of 23.Re5.
23...g4?!
There is a clever idea behind this, but it is tactically flawed, so Black should probably prefer something like 23...b6.
24.Re5?!
This comes to be Komodo14.1's top choice, but is probably a mistake. If I had seen White's reply, I would have played 24.fxg4. After 24...Bxg4 25.Nf6 Bf5 (an only-move, as 25...Bd7?? lets White have an easily won pawn-ending with 26.Nxd7 etc) 26.Re5 Bg6 27.Rxc5 White has won a pawn while keeping a kingside pawn-majority. However there is still a lot of play left, so 26.Re8!? has a lot to be said for it, even though the engines flash red. They continue 26...Rxe8 27.Nxe8 Kd7 28.Nf6+ Ke6 29.Nh5 Ke5, claiming a slight edge for Black. I find this very difficult to believe as White has the much-superior pawn-majority, and can put the white queenside pawns on a dark-square V formation (a3, b2, c3), where they cannot be attacked by the bishop and where they prevent Black creating a passed pawn. I would be disappointed if I did not win such a position with white over the board, so I will return to this in another post when I have time to study it properly.
 |
| What was Black's idea behind pushing the g pawn? |
*****
*****
*****
*****
24...gxf3!? 25.gxf3
Not 25.Rxf5?? gxf2 etc.
Not 25.Rxf5?? gxf2 etc.
After the text White has a passed pawn, but the white pieces are not well-placed to protect it, and meanwhile White has two isolanis.
25...Rf8 26.Ng3!?
Grandmaster Skembris suggested 26.Rxc5. I rejected it because of 26...Bg6 27.f4 Bxh5 28.Rxh5 Rxf4, which I felt should be a draw. After 29.Rxh6 Komodo14.1 gives White the upper hand, but Stockfish15.1 calls the position equal.
26...Bd7 27.Re3 Bc6 28.Ne4 Bxe4 29.fxe4
 |
| How would you assess this ending? |
*****
*****
*****
*****
White is effectively a pawn up as Black cannot create a passer on the queenside by normal means (I will explain this further later in this post as two players came up to me after the game, thinking they had found a way for Black to win on the queenside). But White has two isolated pawns, and Black can make his rook active. The engines agree the position is completely equal.
29...Kd7
This does not spoil anything, but as Grandmaster Skembris pointed out, Black can play 29...Rf1+ 30.Kd2 Rf2+ 31.Re2 Rf3, after which White cannot make progress.
30.Kd2 Ke6?!
Here 30...Rf2+ transposes to the previous note.
Stockfish15.1 reckons Black is still fine after the text, but Komodo14.1 fluctuates between equality and a slight edge for White.
31.Rh3 Rf6
Now the white rook is active and the black rook is passive - Skembris.
The engines suggest 31...Rf2+!?, and if 32.Kd3 then 32...c4+!?, eg 33.Kc3 Ke5 34.Rxh6, after which White has won a pawn but the black pieces are active.
32.Ke3 b6 33.Rf3 Rg6 34.Rg3 Rf6 35.Rg7 Kd6 36.h4
 |
| Black to play and draw |
*****
*****
*****
*****
36...a5
The engines give 36...Rf1! 37.Rg6+ Ke5 38.Rxh6 Re1+ with what they say is more-or-less complete equality (Komodo14.1 gives White an advantage of 0.02).
37.c4 c6?!
This sets a trap, but almost certainly loses. Again ...Rf1 seems to give excellent drawing chances, as does, according to the engines, 37...a4!?
 |
| How should White continue? |
*****
*****
*****
*****
38.h5!
This is better than going after the queenside pawns with 38.Rb7, as then 38...Rf1 is fairly obvious
and gives good counterplay.
This is better than going after the queenside pawns with 38.Rb7, as then 38...Rf1 is fairly obvious
and gives good counterplay.
38...b5!?
Almost certainly better is 38...Rf1, but it is probably losing.
 |
| White to play and win |
*****
*****
*****
*****
39.Rg6! Rxg6
There is no defence, although two players afterwards suggested to me Black can win with 39...Ke6 40.Rxf6+ Kxf6, followed by using the queenside pawns to create a passer to divert the white king. Black is indeed winning if White takes time out to play 41.b3??, but 41.Kf4 wins easily, as does 41.a4. The point, as will also be seen later in the game, is Black can do nothing on the queenside, eg 41.Kf4 bxc4 42.e5+ Ke6 43.Ke4 a4 44.a3 etc.
40.hxg6 Ke7
Or 40...Ke6 41.e5.
41.e5 Ke6
If 41...Kf8, then, among others, 42.e6 wins.
42.Kf4 a4 43.a3 b4 44.Kg4 bxa3 45.bxa3
H-JM now played a 'spite capture' ...
45...Kxe5
... and immediately resigned.
No comments:
Post a Comment