Spanton (1900) - Sam Davies (1705)
King's Indian Attack
1.e4 e6 2.d3 c5 3.Nc3 Nc6 4.c3!?
This semi-waiting move, avoiding committing to fianchettoing the light-square bishop, has been played by Svidler, Rapport and Bologan.
4...Nf6 5.Nbd2 Be7 6.g3
Kingside fianchettoing in the KIA is generally considered better when Black has committed to developing his dark-square bishop on the f8-a3 diagonal rather than at g7.
6...0-0 7.Bg2 d5 8.0-0 b6!?
Three moves - the text, 8...Qc7 and 8...b5 - are of roughly equal popularity in ChessBase's 2019 Mega database.
9.e5 Nd7 10.Re1 Qc7 11.Qe2 Ba6
This is both more popular and more successful than 11...Bb7, which would seem to fit in better with ...b6.
12.Nf1 b5!?
This two-stage push of the b pawn has clearly cost Black a tempo, but it scores well and was played by Korchnoi in a 1958 Soviet championship win over Gurgenidze. I guess it confirms what is generally accepted, ie the KIA is not a serious attempt by White to get an opening advantage - merely one way of playing for an interesting middlegame.
13.Bf4 b4 14.c4 Nb6
Korchnoi preferred 14...dxc4 15.dxc4 Nb6.
15.b3 Rad8 16.Rad1 Bb7 17.h4 a5 18.Ng5?
Premature. Clearly better was 18.Ne3 or 18.N1h2.
18...Nd4
I had not considered this obvious reply.
19.Qh5 Bxg5!?
Also good is 19...h6 20.Nf3 g6 21.Qg4 (21.Qxh6?? Nf5 traps the queen) h5.
20.Bxg5
20.hxg5!? is possible, but White's attack down the h file will probably take too long to organise.
20...Ra8
Stockfish10 and Komodo9 prefer 20...Rd7.
21.cxd5?
Giving Black the d5 square for his pieces is positionally hopeless. It still made sense to get the f1 knight into play.
21...Bxd5 22.Bxd5?
I had originally hoped to play 22.Re4??, but 22...g6, or 22...Bxe4 23.Bxe4 g6, is a complete answer.
The engines give 22.Qg4 Bxg2 23.Kxg2 Qc6+ 24.Kh3, when 24...Qf3!? puts paid to White's attack, leaving Black with a pleasant edge.
22...Nxd5 23.Nh2
At last!
23...Nc3 24.Rd2 a4 25.Qg4 Kh8 26.Nf3 Ncb5 27.bxa4 Nxf3+ 28.Qxf3 Nd4?
Hitting the queen like this looks so natural, but it is a mistake. The engines want Black to get on with it on the queenside with 28...Rxa4, the point being that 29.Bf6 is easily met by, among other moves, 29...Rg8.
29.Qf4??
Still trying for a phantom kingside attack. Correct was to redeploy in the centre with 29.Qe4 Rxa4 30.Be3, and if 30...b3, then 31.Bxd4 bxa2 32.Qe2 holds.
29...Rxa4
SD's postmortem suggestion of 29...f5 is also strong, but by now almost anything wins for Black.
30.Bf6 Nf5?
Seeing threats that are not really there. Simple was 30...Rg8 but, as the engines point out, even 30...Rfa8!? is winning as White does not have a perpetual.
31.g4 gxf6
Black is better after 31...Ne7! 32.Bxe7 Qxe7, but it was necessary to see that 33.Bxg7+?? loses to Kxg7 34.Qf6+ Kg8 35.Qg5+ Ng6 36.h5 Qe7.
32.gxf5 Rg8+ 33.Kh2 fxe5 34.Qxe5+ Qxe5+ 35.Rxe5
 |
| At a glance, White might seem slightly better, or at least equal, but Black has a move that wins a pawn |
*****
*****
*****
*****
35...b3! 36.Re4
The only decent defence against Black's threat of 36...Rxh4#.
36...Rxa2 37.Rxa2 bxa2 38.Ra4 exf5 39.Rxa2
Black has won a pawn, but I do not think many people would be surprised if the game ended in a draw, although the engines dutifully give Black a slight edge.
39...Rg4 40.Kh3 Rd4 41.Ra3 h5?!
I cannot see any advantage to Black in playing this, and it is always possible Black might miss the reserve tempi his h pawn could give him. However, the natural 41...Kg7 does not seem to promise Black anything either; play would be very much as in the game.
42.Rc3 Kg7 43.f3 Rd5 44.Kg3 Kf6 45.f4 Ke6 46.Kf3 Rd8
Neither does 46...Kd6 47.Ke3 Kc6 get Black anywhere after, for example, 48.Ra3.
47.Rxc5 Rxd3+ 48.Kg2 Rd8 49.Kf3 Rh8
49...Rg8 is met by 50.Rc6+ Ke7 51.Rh6.
The remaining moves were:
50.Rc6+ Ke7 51.Rc5 Kf6 52.Rc6+ Kg7 53.Rc5 Kg6 54.Rc6+ f6 55.Rc7 Ra8 56.Rb7 Ra1 57.Kg2 Re1 58.Ra7 Re2+ 59.Kg3 Re1 60.Kg2 Re4 ½–½
No comments:
Post a Comment