Allan Savage – Harry Cohen
1 e4 e5 2 f4!
Last round games are special. I had prepared carefully for this one, but I thought there was only a 10% chance I would see this position on the board. It obviously came as a bit of a shock to my opponent. After much thought, he followed the variation recommended for Black by IM Larry Kaufman in his recent book, The Advantage in Black and White. Lucky for me, I had correctly guessed that my opponent would pick this variation.
The King’s Gambit is underrated as an over-the-board weapon. It requires an enormous amount of study to play it with White, however it can be an investment with a high return. Since Black rarely has to face it, he is often ill-prepared to meet it. It is not widely known that Nigel Short was going to play the King’s Gambit against Kasparov in their 1993 World Championship match, had Kasparov answered 1 e4 with 1…e5. Now THAT would have been something to see!
2 … d5 3 ed5 ef4 4 Nf3 Nf6 5 Bb5+ c6 6 dc6 Nxc6
Another popular line is 6…bc6 7 Bc4 Nd5, but theory currently considers that the text move is best.
7 d4 Bd6 8 Qe2+ Qe7
After 20 minutes thought, Black plays the obvious, but unfortunately inferior reply. As Kaufman suggests, Black should play 8…Be6!? 9 Ng5 0-0 10 Nxe6 fe6 11 Bxc6 bc6 12 0-0!, obtaining compensation in development in return for a pawn that he is about to lose. How much compensation is open to debate.
Instead, the text move leads to a queenless middlegame with an unbalanced pawn structure, but one in which the f4 pawn is quite vulnerable. It is known to be better for White.
9 Qxe7+ Kxe7 10 0-0 Nb4?!
Black’s problems are illustrated by 10…Bg4 11 Bxc6 bc6 12 Ne5, but the move played is equally bad. The best try may have been developing with 10…Re8.
11 Re1+ Kd8
Neither do 11…Kf8 nor 11…Be6 solve Black’s problems, which stem from the bad position of his King. This becomes a factor in the coming complications.
12 Na3 Be6 13 c3! Nbd5
Best was 13…a6 14 Bf1 Nc6, but after 15 Nc4 Bc7 16 Nce5, White still holds a clear advantage. On the other hand, the text move leaves Black vulnerable to a subsequent c4 and d5. Note that 13…Nxa2 loses material to 14 Rxe6 Nxc1 (or 14…Bxa3 15 Re1 Nxc1 16 Rxa3 and the Knight is trapped) 15 Rxd6+ Ke7 16 Rxf6!
14 Nc4 Bc7 15 Nce5!
It is difficult to see the defense to the threat of c4 and d5, coupled with the fork threat at f7. White’s central pawn majority in conjunction with Black’s vulnerable King looks imposing. Still, there was a chance to put up stiff resistance by 15…a6 16 Ba4 Bxe5 17 Nxe5 b5 (stopping c4) 18 Bb3 Re8, though White retains the advantage.
15….h6? 16 c4 Nb4 17 d5 Nc2 18 Bxf4!
Also winning was 18 de6 Nxe1 19 Nxf7+ Ke7 20 Nxe1.
18…Nxa1
On 18…Bb6+ 19 Kf1 Nh5 20 Bd2 Nxa1 21 de6 Nc2 22 Nxf7+ Ke7 23 Re5, Black is still lost.
19 de6 Bxe5 20 Nxe5 Nc2 21 Nf7+ Kc8 22 Re5 (?)
In the time scramble, missing mate in four moves, beginning with 22 Rd1. But the text move is adequate since the threat of 23 Rc5 mate forces Black’s reply, as 22…Nb4 23 Rc5+ Nc6 24 Rxc6+ mates next move.
22… Kc7 23 Nxh8 Rxh8 24 Re2+ Black resigns 1-0.