Jim West (FIDE 2225) - Saevar Bjarnason (FIDE 2315), New York Open 4/2/1994
1.e4 c5 2.Nf3 Nc6 3.d4 cxd4 4.Nxd4 g6 5.Nc3 Bg7 6.Be3 Nf6 7.Bc4 O-O 8.Bb3 d6
What started out as an Accelerated Fianchetto has transposed into the Dragon variation.
9.f3 Bd7 10.Qd2 Rc8 11.h4 Ne5 12.O-O-O Nc4
One year earlier at the same tournament, international master Elliott Winslow tried the Soltis variation by transposition with 12...h5. That game continued 13.Bg5 Rc5 14.Kb1 b5 15.g4 a5 16.Bxf6 Bxf6 17.gxh5 a4 18.Bd5 e6 19.hxg6 exd5 20.h5 Bg5 21.f4 Bh6 22.Nxd5 f5 23.Qg2 Rxd5 24.exd5 Nc4 25.Rd3 Qf6 26.c3 Rc8 27.Ne6 a3 28.b3 Nb6 29.Ng5 b4 30.Nf7 Kg7 31.Nxh6 Bb5 32.c4 Nxc4 33.Nf7 Nb2 34.h6+ Kf8 35.Rd2, Black resigns.
13.Bxc4 Rxc4 14.h5 Qc7
At the Edgar McCormick Futurity in 1991, international master Mike Valvo played the more usual 14...Nxh5. That encounter proceeded 15.g4 Nf6 16.Nde2 Re8 17.e5 Nxg4 18.fxg4 Bxe5 19.Bd4 Rxd4 20.Nxd4 Bxg4 21.Qh6 Bh5 22.Nf5 Kh8 23.Rxh5 gxh5 24.Rh1, Black resigns.
15.hxg6 fxg6 16.Nde2 Rc8 17.Kb1 Be6 18.Bd4
The book continuation is 18.Nf4 Qd7 19.Nxe6 Qxe6 "and White has trouble holding his queenside together", according to Miles and Moskow.
18...b5 19.Bxf6!?
This move is double-edged, since it unleashes Black's powerful dark-squared bishop. On the other hand, Black's h-pawn can no longer be protected.
19...Bxf6 20.Qh6 b4 21.Qxh7+ Kf8 22.Nf4 Bf7 23.Qh6+ Bg7
On 23...Ke8, White stands clearly better after 24.Ncd5 because the bishop on f6 can be captured with check.
24.Nxg6+ Bxg6 25.Qxg6 Rxc3?
This move should have lost, with correct play on White's part. Critical is 25...bxc3 when White can force a draw by repetition after 26.Qf5+ Ke8 27.Qg6+, etc. If instead Black tries 26...Kg8, the outcome is the same after 27.Qh7+ Kf7 28.Qf5+ since 28...Bf6?? would lose to 29.Rh7+ Ke8 30.Rdh1.
26.Rh7 e6 27.Rdh1?!
I thought this move would force an immediate win, but I overlooked my opponent's brilliant reply. Instead White has a won ending by 27.Rxg7 Qxg7 28.Qxg7+ Kxg7 29.bxc3.
27...Rxf3!
Not only does this move threaten 28...Qxc2+ but it also exposes White's weak back rank to mating themes with ...Rf1+ in lines where White sacrifices an exchange with Rh8+.
28.Rh8+?!
Surprised by Black's last move and in time trouble, I miss a problem-like win after 28.e5!. For example, 28...Rf7 29.Rxg7 Rxg7 30.Rh8+ wins, as does 28...Rf2 29.Rh8+ Bxh8 30.Rxh8+ Ke7 31.Rh7+ Kf8 32.Qh6+ Ke8 (or 32...Kg8 33.Rh8+ Kf7 34.Qh7#) 33.Qxe6+ Kf8 34.Rh8+ Kg7 35.Qh6+ Kf7 36.e6+! Ke7 37.Rh7+ Kd8 38.Qg5+ Ke8 39.Qg6+ Kf8 40.e7+ Qxe7 41.Rh8#.
28...Bxh8 29.Rxh8+ Ke7 30.Rh7+ Kf8 31.Rh8+
Now White has nothing better than a draw by repetition because 31.Rxc7?? allows 31...Rf1# and 31.gxf3?? fails to 31...Qxc2+.
31...Ke7 32.Rh7+ Kf8 33.Rh8+, draw.
{This article originally appeared in Atlantic Chess News in 1994}
