Spanton (1972) - Anders Lundbäck (1897)
Spanish Cozio
1.e4 e5 2.Nf3 Nc6 3.Bb5 Nge7 4.0-0
White has many ways of meeting Cozio's Defence, but castling is easily the most popular.
4...Ng6!?
This gives a strange impression - Black has spent two tempi moving the king's bishop to g6, when it could have gone to the more-central f6 in one move. Normal is 4...g6, which Alexey Dreev in Anti-Spanish: The Cozio Defence (Chess Stars, 2014) calls "no doubt the most accurate." But the text is preferred by the analysis engines Stockfish14.1 and Komodo12.1.1, and has been tried by very strong players, including Adolf Anderssen, Bent Larsen, Nigel Short and Vassily Ivanchuk. AL said it was recommended in a chess course he had taken, and contains sharp traps.
5.d4 exd4 6.Nxd4 Bc5 7.Nb3
Most popular in ChessBase's 2022 Mega database is 7.Be3, after which 7...Nxd4 8.Nxd4 Bxd4 9.Qxd4 Qg5!? introduces one of the aforementioned traps.
 |
| Position after 9...Qg5!? - how should White respond? |
*****
*****
*****
*****
10.Qd5 is the commonest move in Mega22, after which Alonso Zapata (2555) - Michael Rhode (2575), World Open (Philadelphia) 1993, continued 10...h6!? 11.Bc4 0-0 12.Nc3 d6 with what the engines reckon is a roughly equal position ( ½–½, 54 moves). My instinctive response, when AL showed me the position, of 10.Be2? runs into 10...Nf4 11.Bf3 Qxg2+! 12.Bxg2 Ne2+ etc. Even worse is 10.Bc4?? Nh4, while 10.Nc3??, which has been played by a 2219, also loses to 10...Nh4. Best, according to the engines, is 10.e5, after which they reckon 10...Nf4 11.g3 Ne6 12.Qe4 f5 13.Qd3 f4 14.Nc3, as has been played in two games in Mega22, is slightly better for White.
However, while 7.Be3 leads to interesting play, the engines reckon better are the text and 7.Nf5!? 0-0 8.Be3.
7...Bb6
The bishop is open to harassment on this square by the white queen's knight coming to d5, but then, as AL pointed out in the postmortem, the same is the case after 7...Be7.
8.Nc3 0-0
Black can retain the bishop-pair with 8...a6, although after 9.Bc4 0-0 10.Nd5 White was slightly better, according to the engines, in Daniel Jacobsen Kovachev (2243) - Josef Ask (2183), Rilton Cup (Stockholm) 2009-10 (but 0-1, 50 moves).
9.Nd5 Re8 10.Nxb6 axb6 11.f3 Qh4 12.g3!?
This weakens the white king's position but restricts the possibilities of the black king's knight. The engines are not keen on it, but cannot agree on what should be played. Stockfsh14.1 likes 12.Nd4 for a long time, but later prefers 12.a4 or 12.Rf2, while Komodo12.1.1 also rates 12.a4, as well as 12.Be3.
12...Qf6 13.c3 h5!?
Very interesting is Stockfish14.1's suggestion of 13...d5!?, meeting 14.exd5 with 14...Rd8, eg 15.c4?! Bh3 16.Rf2 Nce5 gives Black lots of play for a pawn. Better seems to be the defensive 15.Be2, but 15...Nce7 again gives Black decent play.
14.Nd4
Even stronger seems to be the engines' 14.f4!?, meeting 14...Rxe4 with 15.Qxh5, and 14...h4 with 15.e5.
14...Ra5?!
This is probably a waste of time. Consistent is 14...h4, when White has a slight edge at best.
15.a4 h4 16.b4 Ra8 17.Kg2
Stockfish14.1 gives the strange-looking 17.Ra3!?, eg 17...hxg3 18.hxg3 Nxd4 19.cxd4 c6 20.Be2, claiming White is on top, although Komodo12.1.1 reckons Black is equal.
17...Nce5?
A serious waste of time. The engines like 17...Re5, threatening 18...Nxd4 19.cxd4 Rxb5, or 17...d6!?, when 18.Nxc6 bxc6 19.Bxc6 Qxc3 20.Bxa8 Qxa1 21.Bc6 is unclear. Also better than the text is 17...Nxd4, preparing to free Black's light-square bishop.
18.f4 Nc6 19.e5 Qe7 20.Qd3
The engines like, among other moves, 20.Kg1!?, which means ...h3 will not come with check, and there is no knight check at f4 or h4.
20...Nxd4 21.cxd4
21.Qxd4 Rd8 is equal, according to the engines.
21...c6
Black is losing after the greedy 21...Qxb4? as both 22.f5 and 22.Ba3 are very strong.
22.Bc4 d5
 |
| How should White proceed? |
*****
*****
*****
*****
23.exd6?!
Opening lines seems natural, but the engines do not like it, reckoning White is only better after 23.Bb3, after which White can hope to exploit his kingside pawn majority.
23...Qxd6 24.f5 Nf8
The engines prefer 24...Ne7!?, the point being to meet 25.f6 with 25...b5! (25...Nd5 may also be fine for Black) 26.Bb3 Nd5 27.fxg7 Bg4, when White' king seems to be the one more in danger. Instead White should probably play 25.Bf4 with unclear play.
25.f6?!
Probably better is 25.Bb3, when 25...b5!? 26.a5 Nh7 is slightly better for White, according to Stockfish14.1, although Komodo12.1.1 calls the position equal.
25...gxf6
Black is fine after this, according to the engines, but they prefer 25...b5!? 26.fxg7 bxc4 27.gxf8=Q+ Qxf8 28.Qxc4, claiming Black has more than enough play for a pawn.
26.Bf4
The engines prefer 26.gxh4!?
26...Qd7
Black is better after 26...h3+ 27.Kg1 Qxb4 or 26...Qxb4 27.gxh4 Be6, according to the engines.
27.gxh4!?
The white king can find relative safety on h1.
27...Qg4+?
Best seems to be 27...b5!? 28.axb5 Rxa1 29.Rxa1 cxb5, eg 30.Bxb5 Qd5+, when the engines reckon the white king is the more endangered, or 30.Bb3 Qg4+ 31.Qg3 Ng6 32.Qxg4 Bxg4 with roughly equal chances.
28.Kh1 Ng6??
This makes matters a lot worse, whereas after 28...Qf5 or 28...Kh8 the game carries on, albeit with White much better.
29.Rg1 Bf5
Not 29...Qxf4?? as White mates in two, starting with 30.Qxg6+.
30.Rxg4 Bxd3 31.Bxd3 (1-0, 38 moves).
No comments:
Post a Comment