Still, when it came to putting my FIDE rating on the line at the World Open in Philadelphia, I decided to "play it safe" with the Sicilian defense. At that stage in my chess career, I lacked confidence in the PCG against the top-notch players. In round six of the World Open, I faced FIDE master Brian Hartman of Canada (who has since become an International Master).
At that point, I was playing the best chess of my life. In late June, I had finished in first place at a futurity in Elmwood Park that included Stoyko, Feuerstein, Lahoz, Radomskyj, Klemm, and Friend - to name a few. With the $250 in prize money burning a hole in my wallet, I drove out to Philadelphia where I plunked it down as my entry fee.
After losing to GM Larry Christiansen in the first round (on the elevated stage at board one), I had not lost another encounter, including a victory over IM Ed Formanek. At the end of the tournament, my FIDE rating stood at 2275, the highest it has ever been.
To this day, I wonder if it was mere coincidence that Atlantic Chess News was temporarily out of circulation at the time following Glenn Petersen's rise to the editorship of Chess Life. For a brief "window of opportunity", I was able to focus my attention squarely on my over-the-board play without having to worry about journalistic responsibilities.
In any event, this game with Hartman was my "last hurrah" with the Najdorf Sicilian in serious play. As the reader will soon see, the middlegame position was every bit as wild and unclear as any position that has ever arisen out of the PCG in my games. But don't hold your breath waiting for anyone to opine that the Najdorf variation is unsound!
Brian Hartman (FIDE 2345) - Jim West (FIDE 2210), World Open 1990
1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 a6 6.Bg5 e6 7.f4 Be7 8.Qf3 Qc7 9.O-O-O Nbd7 10.Bd3 b5 11.Rhe1 Bb7 12.Qg3 O-O-O
So far, this is Spassky-Fischer, Reykjavik 1972. Now my opponent avoided Spassky's 13.Bxf6 winning a pawn (although Black eventually drew) in favor of what was then considered to be the refutation of 12...O-O-O, first played by Velimirovic against Kazzaz.
13.Bxb5! axb5 14.Ndxb5 Qb6 15.e5 Nc5!
Kazzaz had played the weaker 15...d5.
16.exf6
Stronger may be 16.Qh3. Now my original suggestion of 16...Ba6!? 17.exf6 gxf6 18.Nd5 Qb7 with an unclear position is smashed by GM John Nunn's 19.Nxe7+ Qxe7 20.Qc3! fxg5 21.Nxd6+. So Black must try 16...dxe5 17.fxe5 Nd5 18.Bxe7 Nxe7 19.Nd6+ Kb8 20.Nxf7 Rxd1+ 21.Rxd1 Rf8 unclear.
16...gxf6 17.Bh4 Rhg8 18.Qe3 Rxg2 19.Qd4
After the blunder 19.Bf2?, I had won a few games with the neat trick 19...Rxf2! when 20.Qxf2 allows 20...Nb3+ 21.axb3 Qxf2. But 19.Kb1 Nd7 (19...Na4!?) or 19.Bg3 f5 (with the idea of ...h5, ...h4) are worth considering.
19...Bf3?!
I should have played 19...Nd7 20.Nxd6+ Bxd6 21.Qxd6 Qxd6 22.Rxd6 Rxh2 23.Bg3 Rg2 24.Re3 Rg8 25.Be1 f5 with an approximately equal position, although Nunn thinks that 26.Nb5 is still +/=.
20.Rd2 Rxd2 21.Qxd2
Black has some compensation (although probably not enough) after 21.Kxd2 d5 22.Kc1 (22.Bxf6?? Bxf6 23.Qxf6 Ne4+ -+) Nd7 23.Qd3 Be4 24.Nxe4 dxe4 25.Rxe4 Kb8.
21...d5 22.Bf2 Qa6 23.Bd4 Be4 24.Qe3 Na4 25.b3!?
In his new book The Complete Najdorf 6.Bg5, Nunn agrees with FM Rudy Blumenfeld's analysis in Informant 50 of 25.Nxe4 dxe4 26.c4, giving it as winning for White, no doubt because Black must lose material due to the threat of 27.b3. But 26...f5 27.b3 Nb6 28.Bxb6 Rd3 29.Qf2 Rf3! (29...Bh4? 30.Qc5+ +-) 30.Qg1 Qxa2 31.Be3 Qxb3 32.Qg8+ Kb7 33.Qxf7 Qxc4+ 34.Kd1 Qd3+ leads to a draw by perpetual check.
25...Nxc3 26.Qxc3+ Kd7 27.Qc7+ Ke8 28.Rg1! Bg6
Blumenfeld refutes 28...Qxb5? by 29.Rg8+ Bf8 30.Rxf8+ Kxf8 31.Qxd8+ Kg7 32.Qxf6+ Kf8 33.Qd8+ Qe8 34.Bc5+ +-.
29.Bc5! Bxc5 30.Qxc5 Qxa2
This was the end of the first time control. I got up from the table to get a drink of water, looked at my wrist watch, and seriously considered resigning. But the hour was not yet late, and I decided to play on for a few more moves. One of the spectators took a look at my position, shook his head sadly, and left the tournament hall. By this point, I was berating myself for not having played the PCG. It could hardly have been any worse!
31.f5 d4!
This is the only move. Now Black threatens 32...Qa1+, winning White's rook.
32.Rg4
Instead 32.Kd2 d3! 33.fxg6 dxc2+ 34.Nd6+ Rxd6+ 35.Qxd6 c1=Q+ 36.Kxc1 Qa1+ 37.Kd2 Qxg1 38.gxh7 Qg2+ 39.Kc1 Qg1+ 40.Kb2 Qg2+ 41.Ka3 Qa8+ draws (42.Kb4? Qe4+ and 43...Qxh7). But until Brian McCarthy left a message on my voice mail in June 1999, the move 32.Rd1! had never occurred to me. Apparently Blumenfeld missed it, as well. Certainly my opponent did. McCarthy's idea of retreating the rook to d1 is to shore up the defense of White's king while simultaneously threatening 33.Nc7+ Kd7 34.Rxd4+ which is curtains. Black can do no better than 32...e5 33.fxg6 fxg6, but his two pawns are insufficient compensation for White's knight.
32...Qa1+ 33.Kd2 exf5! 34.Rg2
Only now did my opponent realize that his intended 34.Nc7+ Kd7 35.Rxd4+ would fail to 35...Qxd4+! 36.Qxd4+ Kxc7-+.
34...d3 35.Nc7+ Kd7 36.Nd5 Qe5 37.Qb5+ Kd6 38.Qb6+ Kd7 39.Qb5+ Kd6 40.Qb6+ Kd7
Here I offered a draw, feeling certain that my opponent would accept since he had less than a minute to make the next time control. Much to my amazement, he declined.
41.c4? f4!
Now Black is winning, having a flight square for his king on f5.
42.Nxf6+ Ke7 43.Nd5+ Rxd5! 44.cxd5 Qb2ch 45.Kd1 Bh5+ 46.Ke1 Qc1+!
This is stronger than 46...Qxg2.
47.Kf2
And here, with no time left, White resigned in the face of 47...Qd2+ 48.Kf1 Qd1+ 49.Kf2 Qe2+ 50.Kg1 Qe1#.
As I drove back to my motel after this game, the thought crossed my mind that I had just played "an Informant game". And, indeed, I was right!
Approximately three months later, at the Kasparov-Karpov world championship in New York City, I encountered the same spectator who had shaken his head sadly in Philadelphia. The first question he asked is how I had managed to win my game against Hartman. I just shrugged my shoulders and grinned.
*{This article originally appeared in the July-August 1999 issue of Atlantic Chess News}
