Frank Gillespie (1496 ECF/1321 Fide) - Spanton (2019 ECF/1804 Fide)
Spanish Berlin
1.e4 e5 2.Nf3 Nc6 3.Bb5 Nf6 4.0-0 Nxe4 5.d4 a6!?
An invitation to transpose to the Open Variation of the Spanish by 6.Ba4, while avoiding the Exchange Variation, ie 3...a6 4.Bxc6.
6.Bxc6!?
This is an attempt to refute Black's move-order.
6...dxc6 7.Re1 Nf6
Not 7...Bf5? 8.g4!, when 8...Bg6 9.Nxe5 Qf6 10.Qe2 0-0-0 11.Nxg6 hxg6 12.Qxe4 Rxd4 13.Qg2 leaves Black with nowhere near enough compensation for being knight-for-pawn down, according to Stockfish14.1. and Komodo12.1.1. However the engines reckon 8...Nxf2!? 9.Kxf2 Bxg4 is not quite so clear, and might give Black practical chances.
8.dxe5!?
The engines prefer 8.Nbd2!? or 8.Nxe5.
8...Qxd1 9.Rxd1 Nd5
 |
| Much as in the Berlin Wall, White has a kingside pawn-majority with a pawn on e5; Black has the bishop-pair but, unlike in the Berlin Wall, has not lost castling rights |
*****
*****
*****
*****
10.a3?!
This is probably too slow. The engines give 10.c4, intending to meet 10...Nb4 with 11.Nc3!?, eg 11...Nc2 12.Rb1 Bf5 13.Bf4 Nb4 14.Ra1, and if 14...Nc2?! then 15.Rac1.
10...Bg4
Maybe even stronger is 10...Bf5.
11.Rd3?!
The engines give 11.Nbd2, albeit preferring Black.
11...Bf5 12.Rb3 0-0-0 13.c4 Bxb1?!
The start of an interesting combination, but almost certainly better is the engines' 13...Nf4!?, which is possible thanks to the threat of back-rank mate.
14.Rxb1 Nc3 15.Bg5 Nxb1 16.Bxd8
 |
| How should Black proceed? |
*****
*****
*****
*****
16...Bc5?
Correct is 16...Kxd8, eg 17.Rxb7 Kc8 18.Ra7 Kb8 19.Rxa6 Be7, when 20.Rxc6?? Kb7 is curtains for White.
17.Bh4 h6 18.Rd3 g5 19.Bg3
Probably winning is 19.Nxg5!? hxg5 20.Bxg5, when White prevents an exchange of rooks. A sample continuation runs 20...Re8 21.h4 b5 (21...Rxe5?? 22.Rd8#) 22.cxb5 cxb5 23.Bf6 Be7 24.Bxe7 Rxe7 25.f4 c5 (25...Rd7? 26.Rxd7 Kxd7 27.h5 Ke7 28.h6 Kf8 29.f5 Nd2 30.e6 etc) 26.h5 c4 27.Rd1 Nxa3 28.bxa3, giving White what the engines reckon is a winning rook-and-pawn ending. Little of this is forced, but this and other variations look very good for White.
19...Rd8 20.Rxd8+ Kxd8
 |
| The start of a tricky ending that Komodo12.1.1 reckons is dead-equal, but Stockfish14.1. believes slightly favours White |
*****
*****
*****
*****
*****
21.h3
Otherwise the black knight escapes thanks to ...g4, dislodging the white knight.
21...b5
Unclear is 21...h5!? 22.Nxg5 Bd4 23.b3 Nd2 24.Nxf7+ despite White being, at least temporarily, two pawns up.
22.cxb5 cxb5 23.Kf1 Kd7 24.Ke2 Ke6
Komodo12.1.1 prefers 24...Na3!? 25.bxa3 Bxa3, giving White a slight edge. Stockfish14.1 reckons Black should play 24...a5, but also favours White.
25.Kd3 Kd5 26.Kc2
The engines give 26.a4!?, the point being 26...bxa4 27.Kc2 Na3+ 28.bxa3 Bxa3 gives Black what seems to be an inferior version of how the black knight is lost in the game.
26...Nxa3+ 27.bxa3 Bxa3 28.e6!
Preventing Black from attacking on the queenside with three connected passed pawns.
28...fxe6?!
Not 28...c5?? 29.e7. But probably better is 28...Kxa6 29.Bxc7 Bc5, when the f2 pawn can hardly be saved.
29.Bxc7 Bc5 30.Bg3?!
The engines' 30.Ne5 seems much stronger.
30...a5 31.Nd2 a4 32.f3 b4 33.h4 Be3 34.hxg5 hxg5 35.Be1 b3+?!
The game is dead-equal after 35...Bd4 or 35...Kd4, according to the engines.
 |
| How should White proceed? |
*****
*****
*****
*****
36.Nxb3?!
The engines' 36.Kd3 seems strong, eg 36...b2 37.Nb1 Bc5 38.Nc3+ Ke5 39.Kc2 a3 40.Ne2!
36...axb3+ 37.Kxb3 Ke5 38.Kc3 Ke5
 |
| White to play and lose |
*****
*****
*****
*****
39.Bd2??
Within a few moves the position has gone from probably winning for White to definitely winning for Black.
39...Bxd2+ 40.Kxd2 Kg3 41.Ke3 Kxg2 42.Ke4?!
Objectively, White is lost whatever is played, but 42.Ke2 f5 43.Ke3 obliges Black to find 43...Kf1.
42...Kf2 43.f4 g4 0-1
No comments:
Post a Comment