Short-Gulko, New York 1994

Boris Gulko

It is always useful to devise something new for a match, to undermine your opponent's home preparation. Accordingly I prepared the Caro-Kann Defense for my games with the black pieces and switched from my usual 1 d4 to 1 c4 as White. The latter is not a very big change for me--but it did deprive my opponent of the Queen's Gambit Accepted, which had served Short so well in his match against Karpov.

As I see it, my match with Short had two symmetrical parts. The first four games were tense. We exchanged victories in the second and third games; in both, Black collapsed quickly after an unsuccessful opening. During the next four games I was only trying to survive--but not in the chess sense of the word! My temperature had risen close to 104 degrees Farenheit (40 degrees Celsius) by the evening of the fourth game. The doctor told me, "It's nothing special, just a virus. In seven days, the virus will die and you'll be okay!" But that meant that my fever wouldn't end until the day of the playoff--if there was to be a playoff.

During the next few days, my temperature was between 102 and 104 degrees F (39-40 C) in the evenings and mornings, though it decreased during the day. I made two short draws with the white pieces in the sixth and eighth games, when Nigel accepted my offers in the openings. But the fifth and seventh games were fierce battles. In the seventh game I forced a draw in the position even though I had two separate forced wins. It was the sixth hour of the game, near 9 PM, and I urgently needed to get to bed.

In the playoff, Nigel missed a winning opportunity in the first game, as I did in the second. But Nigel played the third game very strongly, and for the fourth game I was already exhausted. The weakness after a fever is sometimes even more dangerous to one's chess than the fever itself.

Short-Gulko, New York (m/1) 1994
Caro-Kann Defense B17

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nd7 5 Nf3 Ngf6 6 Nxf6+ Nxf6 7 Bc4 Bf5 8 0-0 e6 9 h3!? Be7 10 c3!? N

The first surprise in the match: I used the Caro-Kann Defense, which I had played only once before--28 years ago. The second surprise: Nigel answered with a new idea, playing 10 c3 (instead of 10 Qe2 0-0 11 Rd1 =, Reshevsky-Smyslov, Palma de Mallorca 1970). It seems that his idea was to place the knight on e5 and attack with g2-g4 and f2-f4.

10 ... 0-0 11 Qe2 [D 1]

11 ... c5?!

This reaction was not the best, although I was relying on my vast experience with the French Defense. If 11 ... Qc7?!, White continues 12 Ne5, with the idea 13 g4 Bg6 14 f4 +/-, so perhaps best was Smyslov's plan from the above-mentioned game against Reshevsky: 11 ... b5!? 12 Bb3 Be4 with an unclear position. Now Short plays consistently and increases his opening advantage.

12 Bf4 Be4 13 Rad1 Bxf3?!

Perhaps better is 13 ... Qb6!? with an unclear position.

14 Qxf3 Qb6 15 dxc5 Bxc5?!

An improvement is 15 ... Qxc5! 16 Bb3 Qb6 +/=.

16 Rd2 Rfd8 17 Rfd1 Be7 18 b3! +/- Rxd2 19 Rxd2 a6 20 g3 h6 21 Be5 Rd8 22 Rxd8+ Bxd8 23 Bf1 Qa5 [D 2]

24 Bxf6?

Instead of cashing in prematurely, Short could have retained a significant advantage with 24 Bd6! (Razuvaev's recommendation). Now the position is drawish on account of the opposite colored bishops.

24 ... Bxf6 25 Qxb7 Qxc3

Not 25 ... Qxa2? 26 Qc8+ Kh7 27 Bd3+ g6 28 Qf8 +-.

26 Qxa6 Bd4 27 Kg2 Qd2?

Costing Black two tempi, but luckily the draw is still there. More accurate was 27 ... g6.

28 Qa8+ Kh7 29 Qf3 Kg8 30 a4 g6 31 Bb5! Qc3

Not 31 ... Kg7? 32 Be8.

32 Qxc3 Bxc3 33 Kf3 Kf8 34 g4 Ke7 35 Ke2 f5 36 Kd3 Be1 37 f3 fxg4 38 fxg4

If 38 hxg4, Black holds with 38 ... h5 39 g5 h4! =.

38 ... h5 [D 3]

For those who don't know it, the method of defense against the two connected passed pawns in the opposite-colored bishop endgame is instructive. Black's king has to be on c7 when his white colleague is on b5.

39 Kc4 hxg4 40 hxg4 Kd6 41 b4 g5 42 Be8 Bd2 43 Bf7 Be1 44 a5 Bd2 45 Kb5 Kc7!

Black must not try to keep his e-pawn: 45 ... Be1? 46 Kb6! Bxb4 47 a6 +-.

46 Bxe6 Be1 47 Bf7 Bd2 48 Be8 Be1 49 Ka4 Kd6! 50 b5 Kc5 51 b6 Bd2 52 b7 Bf4 53 Kb3 Bb8 54 Kc3 Kd6 55 Bb5 Kc7 56 a6 1/2-1/2


Gulko-Short, New York (m/2) 1994
English Opening A29

1 c4 e5 2 Nc3 Nf6 3 Nf3 Nc6 4 g3 d5

As far as I know, this is the second time Nigel used this system, the first time being 10 days before this game against Timman in the Euwe Memorial. But I was prepared for it. My old friend Razuvaev, who helped me enormously during not only my preparation but the match itself, predicted it would be Nigel's choice for several reasons. First of all, the system matches Nigel's style. His trainer, Robert Huebner, is an expert in this system. Besides, Nigel had to prepare something new against 1 c4 for his match with Garry Kasparov. He had been very unsuccessful against Garry's English Opening before the match and had to change his old openings. But Garry didn't play 1 c4 in the match, so the surprise was left for me.

5 cxd5 Nxd5 6 Bg2 Nb6 7 0-0 Be7 8 a3

Here the most popular move is 8 Rb1 as Timman had played in the above mentioned game. After Korchnoi's 8 ... g5, I think the position looks unclear.

8 ... 0-0 9 b4 Re8 10 d3 Bf8 11 Bb2 a5 12 b5 Nd4 13 Nd2 a4 14 e3 Ne6 15 Nf3 [D 4]

I'm sure that in preparing for the match, Nigel analyzed the game Azmaiparashvili - I. Novak (Haifa 1979). There Black answered 15 ... f6, but after 16 Qc2 Nc5 17 d4, White had an advantage. Why not 15 ... Nc5? both Short and I had asked ourselves. But only I had answered this question in my analysis!

15 ... Nc5?! 16 Nxe5! Rxe5

Neither of us had spent more than about five minutes in the opening, but this move took 40 minutes of Nigel's time. and he did find an interesting idea. Bad would be 16 ... Nb3 17 Nxa4 Nxa1 18 Nxb6 cxb6 19 Qxa1 f6 20 Nc4 with a huge advantage for White.

17 d4 Rg5! [D 5]

A surprising move. Normal continuations were unattractive, e.g.: 17 ... Nc4 18 dxe5 Qxd1 19 Nxd1 Nb3 20 Ra2 and White's material advantage is more significant than the awkward position of his pieces; or 17 ... Re8 18 dxc5 Bxc5 19 Ne4 with a clear advantage. The move in the game has a psychological advantage compared to 17 ... Rf5 18 dxc5 (18 e4 Nb3 19 exf5 Nxa1 20 Bxa1 Bxf5 is not promising) 18 ... Rc5 with the same position as in the game: Short gave me a choice between the continuation in the game and the intermediate 18 h4.

18 dxc5

I rejected 18 h4 because of 18 ... Nb3 19 hxg5 Nxa1 20 Qxa1 Qg5 with an unclear position. But stronger for White is 20 Bxa1! Qg5 21 Qd3 or 20 ... Bxa3 21 Qh5, with an initiative in both cases. Black had to play 18 ... Rf5 19 dxc5 Qxd1 20 Raxd1 Rxc5 21 Rd8 Be6! with the same possibilities as in the game. I overestimated the significance of penetration on d8, so I decided not to deviate from the main line.

18 ... Qxd1 19 Raxd1 Rxc5 20 Rd8 [D 6]

20 ... f6?!

During the game neither of us considered the pawn sacrifice 20 ... Be6! 21 Rxa8 Nxa8 22 Bxb7 Nb6, as recommended by Yasser Seirawan in Inside Chess. It looks as though Black would have reasonable compensation after 23 Bc6 Nc4 (23 ... Bb3 24 Ne4 Rc2 25 Be5 Nd5 26 Ra1) 24 Ba1 Nxa3 25 Nxa4 Rc4.

21 Rfd1 Kf7 22 R1d4! Ke7?

I couldn't understand this move during the game, and I don't understand it now. In a samurai's life, it would be a preparation for hara-kiri. A very interesting position could emerge after 22 ... Be7 23 Rh8, when the tempting 23 ... Bf8 can be met with 24 Rdd8 Be7 25 Rdg8! and Black is paralyzed. If 23 ... f5, then 24 Nd5! retains an advantage. But Black can keep struggling with 23 ... h6, as 24 Nd5? doesn't work: 24 ... Rxb5 25 Nxc7? Rxb2 26 Nxa8 Rh1+ 27 Bf1 Bh3.

23 h4 Nc4?

The realization of the faulty idea begun with 22 ... Ke7? Necessary was 23 ... Ra5, but after 24 Bf1! Black's game is extremely difficult.

24 R8d5! [D 7] 1-0

Short-Gulko, New York (m/3) 1994
Caro-Kann Defense B17

The next game began a discussion about 12 Nh3, finally to be resolved, I believe, in the 7th game. After White's bishop occupied d4, his position became more active. But Black was not compelled to play the suicidal 18 ... 0-0. It was possible to play 18 ... 0-0-0 or 18 ... Bc5 (Razuvaev's recommendation). After 18 ... 0-0, Nigel was able to demonstrate his attacking skill.

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nd7 5 Bc4 Ngf6 6 Ng5 e6 7 Qe2 Nb6 8 Bb3 h6 9 N5f3 a5 10 c3 c5 11 a3 Qc7 12 Nh3 Bd7 13 0-0 Bd6 14 dxc5 Qxc5 15 Be3 Qc7 16 Bd4 Ng4 17 Nd2 Nf6 18 Kh1 0-0 19 Bxf6 gxf6 20 Qg4+ Kh7 21 Ne4 f5 22 Nf6+ Kh8 23 Qh4 Kg7 24 Rad1 Be5 25 Nh5+ Kh7 26 f4 Bh8 27 Ng5+ [D 8] 1-0

Gulko-Short, New York (m/4) 1994
Nimzo-Indian Defense E54

1 c4 c6 2 e4 d5 3 exd5 cxd5 4 d4 Nf6 5 Nc3 e6 6 Nf3 Bb4 7 Bd3 dxc4 8 Bxc4 0-0 9 0-0 b6 10 Bg5 Bb7 11 Re1 Nbd7 12 Rc1 Rc8 13 Bd3 Re8 14 Qe2 Bxc3 15 bxc3 Qc7 16 Bh4 Qd6!? N [D 9]

Short's opening was more accurate than in Gelfand-Karpov, Linares 1994, where White had a strong initiative, but less so than in Portisch-Andersson, Buenos Aires 1978, which continued 16 ... Nh5 and was drawn after two more moves.

17 Ba6 Bxa6?!

17 ... Qd5! equalizes.

18 Qxa6 Qb8 19 c4 Qa8 20 a4?! [D 10]

20 Bxf6 Nxf6 21 a4 +/=.

20 ... Ne4

The game could have become wild had Black played 20 ... Nc5!? 21 dxc5 Rxc5. If then 22 Re5?! Nd5! 23 cxd5 Rxc1+ 24 Re1 Rxe1+ 25 Nxe1 Qxd5 =/+ or 24 Ne1 exd5 25 Qe2 d4 =/+. Instead, White would have had to sacrifice the queen: 22 Bxf6! gxf6 23 Nd4 Ra5 24 Qxa5 bxa5 25 Bxf6 gxf6 26 c5, when the strong passed pawn compensates for White's material deficit.

21 Qb5 Rc7 22 Qb1 Nd6 23 Bg3 Rc6 24 Bxd6 Rxd6 25 a5 Rc8 26 Qb5 Qc6 27 Qb4 Qc7 28 axb6 axb6 29 h3 Rc6 30 Qb5 h6 31 Rc2 Qd6 32 Ra1 R6c7 33 Ra6 Qf4 34 Ra1 1/2-1/2


Short-Gulko, New York (m/5) 1994
Caro-Kann Defense B17

After Black's inaccuracy 18 ... Bc6, White obtained an initiative. Better for me was 18 ... Bb5! 19 c4 Ba4, solving all the problems. I had to defend throughout the game, but I don't believe Nigel had a forced win at any moment.

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nd7 5 Bc4 Ngf6 6 Ng5 e6 7 Qe2 Nb6 8 Bb3 h6 9 N5f3 a5 10 c3 c5 11 a3 Qc7 12 Nh3 Bd7 13 0-0 cxd4 14 Bf4 Bd6 15 Bxd6 Qxd6 16 Rad1 0-0 17 Rxd4 Qc5 18 Rfd1 Bc6 19 Ne5 Bd5 20 c4 Bc6 21 Qe3 a4 22 Bc2 Ra5 23 Nd3 Qe7 24 Nb4 Qc5 25 Nd3 Qe7 26 Re1 Nbd7 27 Qd2 Raa8 28 f4 Rfd8 29 Nhf2 Nb6 30 Ne5 Qc7 [D 11]

31 Rd1 Rxd4 32 Qxd4 Rc8 33 Qd6 Be8 34 Qxc7 Rxc7 35 Rd4 Kf8 36 Nfd3 Nfd7 37 Nxd7+ Rxd7 38 Rxd7 Bxd7 39 Ne5 Be8 40 c5 Nd5 41 g3 b6 42 cxb6 Nxb6 43 Kf2 Ke7 44 Ke3 f6 45 Nd3 Kd6 46 Kd4 Bg6 47 Bd1 Be8 48 h4 Bc6 49 b3 1/2-1/2


Gulko-Short, New York (m/6) 1994
Slav Defense D10

This was scarcely a real game because of my illness.

1 c4 c6 2 d4 d5 3 cxd5 cxd5 4 Bf4 Nc6 5 e3 Nf6 6 Nc3 a6 7 Bd3 Bg4 8 Nge2 e6 9 f3 1/2-1/2


Short-Gulko, New York (m/7) 1994
Caro-Kann Defense B17

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nd7 5 Bc4 Ngf6 6 Ng5 e6 7 Qe2 Nb6 8 Bb3 h6 9 N5f3 a5 10 c3 c5! 11 a3 Qc7 [D 12]

This position, which was first introduced to the world in the game Kasparov-Karpov (Linares, 1994), had suddenly become the main topic in our match. Here Kasparov played 12 Ne5?!, which gave Black the opportunity 12 ... cxd4 13 cxd4 Bxa3! Instead Short played a simple developing move.

12 Nh3 Bd7 13 0-0 cxd4

I was afraid to make this natural move in the third game because of 14 Nxd4?! Bd6 15 Nf5 Bxh2+ 16 Kh1, even though Black is OK after 16 ... Kf8 (16 ... 0-0-0 is worse because of 17 Ne7+ Kb8 18 g3 e5 19 Nf4! and the complications are favorable for White). But Short had a different way of approaching this position.

14 Bf4 Bd6 15 Bxd6 Qxd6 16 Rfd1

An improvement on the 5th game, when Short tried 16 Rad1 0-0 17 Rxd4 Qc5 18 Rfd1 (worse for White is 18 Ne5 Bb5) and now 18 ... Bb5 (instead of 18 ... Bc6, as played) would have solved all Black's problems.

16 ... 0-0 17 Rxd4 Qc5 18 Nf4 [D 13]

18 ... Qb5!?

This move reduces the tension in the position. A more complicated situation would emerge after 18 ... Bb5. I was afraid of 19 Nxe6 fxe6 20 Qxe6+ Kh8 21 Ne5, but this aggression could be refuted by means of 21 ... Be8. And since neither 19 Qe5 Nbd7 nor 19 Qe1 Bc4 20 Bc2 Nbd7 promised anything for White, it seems as though Short's whole opening concept, beginning with 12 Nh3, was harmless.

19 Qxb5 Bxb5 20 Re1 Nbd7

This time I escaped a real danger. After the natural 20 ... Rfd8? Short had planned 21 Nxe6! fxe6 22 Rxe6 Nc4 23 Re5 Rxd4 (impossible is 23 ... Ba6? 24 Bxc4+) 24 Nxd4 Ba6 25 Rc5 b5 26 Nxb5 Bxb5 27 Rxb5 Rc8 28 Re5 and the pawns are certainly stronger than the knight.

21 Rdd1 Bc6 [D 14]

The overly preventive play of a sick man. The active 21 ... Nc5 was clearly stronger, and I had difficulty persuading myself not to play it.

22 Nd4 Nc5

Incorrect was 22 ... e5? 23 Nxc6 bxc6 24 Ng6 Rfe8 (or 24 ... Rfb8 25 Bc4 anyway) 25 Bc4! a4 26 Nh4 and 27 Nf5, when Black will regret having moved the pawn from e6.

23 Bc2 Be4 24 Bxe4 Nfxe4! 25 f3

25 b4? does not work because of 25 ... Nc3 26 bxc5 Nd1 27 Rd1 e5 and Black wins.

25 ... Nf6 26 Nd3 Nxd3 27 Rxd3 Rfd8 28 Re5? [D 15]

28 ... Rd5?

Another spiritless move. Why should Black exchange the awkwardly positioned rook on e5? Indeed, why did Short place it there? After the stronger 28 ... Nd5!, White must cope with two threats: an attack on the queenside as well as a breakthrough along the d-file. For example, 29 a4 Nf4 30 Rde3 f6 31 Rb5 e5 32 Ne2 Rd1+ 33 Kf2 Nd3+ 34 Kg3 f5 35 f4 g5! and Black wins, or 29 Ne2 b5! when Black has an advantage on the queenside.

29 Rxd5 Nxd5 30 Nb5

There is no time to support the b5-square: after 30 a4 Nb6! 31 b3 Rd8, White has problems along the d-file.

30 ... a4! [D 16]

With this move, my mood changed. Before, I was thinking only about finishing the game, reaching bed and taking my medicine. But now I recalled that winning this game would practically ensure a victory in the match for me, since I would have White in the last game. Now White's queenside looks miserable.

31 Rd4 Kf8 32 Kf2 Ke7 33 Ke2 g5!

Now White can lose on the kingside as well.

34 g3 f5 35 Kd3?

This cedes the kingside to Black without a fight. I expected 35 g4 with some counterplay.

35 ... h5! 36 c4

Forced, otherwise 36 ... g4 37 f4 h4 would allow Black to penetrate along the h-file, winning the g3 pawn and the game.

36 ... Nf6 37 Rd6 Nd7 [D 17]

38 Kc3

Taking the e-pawn is unattractive: 38 Rxe6+ Kxe6 39 Nc7+ Kd6 40 Nxa8 Ne5+ recovers the pawn favorably.

38 ... Rc8!?

38 ... g4! 39 fxg4 hxg4 was even better:

a) 40 Rxe6+ Kxe6 41 Nc7+ Kd6 42 Nxa8 Kc6 43 Kb4 Ne5! (certainly not 43 ... b6 44 Kxa4 Kb7 45 Nxb6) 44 c5 Nf3 (44 ... Nd3+ 45 Kc4 Nxc5 doesn't work on account of 46 Nc7) 45 Nb6 Nxh2 and White loses on the kingside.

b) 40 Kb4 Rh8 41 Kxa4 (or 41 Rd2 Ne5 42 Kxa4 Nxc4 43 Rc2 Nb6+ 44 Kb3 Nd5) Rxh2 42 b4 Rh3 43 Rd3 f4 is bad for White.

39 Kb4 Ne5 40 c5 Nxf3!?

The sharpest move. I could also have played 40 ... Nd7!? 41 Kxa4 Rxc5 42 Rd2 (White loses the pawn endgame after 42 b4 Rxb5 43 Rxd7+ Kxd7 44 Kxb5 e5 45 Kc5 g4 46 fxg4 hxg4 47 a4 f4 etc.) Re5 with a clear advantage.

41 Rb6 Ne5 42 Rxb7+ Kf6 43 Kxa4 f4!? [D 18]

After 43 ... Rxc5 44 Nd6, White would have sufficient counterplay. Now White's problem is that his rook can not return to the first rank to stop Black's f-pawn.

44 gxf4

White must exchange pawns to free the square g3 for his knight.

44 ... gxf4 45 b4

This move looks like a waste of time. Better was the immediate 45 Nd6.

45 ... f3 46 Nd6 Rg8! [D 19]

Black takes control of the g3 square. It's unclear if 46 ... Rd8 wins (the threat is 47 ... Rxd6 48 cxd6 f2). For example 47 Ne4+ Kg6 48 Re7 Rd4 49 Nf2 Kf6 (bad is 49 ... Rd2 50 Rxe6+ Kf5 51 Rf6+) 50 Re8 Kf7 51 Rc8 Rd2 52 Ne4, intending to sacrifice the knight for the f-pawn with 53 Ng5+.

47 Kb3

During the game, I thought that Nigel could escape with 47 c6 Nxc6 48 Rf7+ Ke5 49 Nc4+ Kd5 50 Nd2 Nd4 51 b5. But Black could still win with 51 ... e5!, as 52 Nxf3 is met by 52 ... Ke6.

47 ... Rg4!

This move is much stronger than 47 ... Kg6 48 Ne4 Kf5 49 Nd2 f2 50 c6! with an unclear position.

48 Rb8 [D 20]

48 ... Kg7

My first intention was 48 ... Rf4 49 Rf8+ Kg5. Here 50 h4+ doesn't work because of 50 ... Kg4 51 Rg8+ Kxh4 etc. And after 50 Rg8+ [D 21], there are two ways to win:

a) 50 ... Kh4 51 Rg1 Kh3 52 b5 Kxh2 53 Rc1 f2 54 c6 f1=Q 55 Rxf1 Rxf1 56 c7 Rc1 57 b6 Nd7 58 c8=Q Rxc8 59 Nxc8 h4 and the h-pawn will soon queen;

b) 50 ... Ng6 51 c6 f2 52 c7 f1=Q 53 c8=Q Qb1+ 54 Ka4 Qd1+ 55 Ka5 Qxd6 with an extra knight.

49 Rb7+ Kf6 50 Rb8 Kg7 51 Rb7+ 1/2-1/2


Gulko-Short, New York (m/8) 1994
Slav Defense D10

In Game 8, just as in Game 6, I exchanged my right to the first move for the opportunity to stay in bed.

1 c4 c6 2 d4 d5 3 Nc3 Nf6 4 e3 a6 5 Nf3 b5 6 b3 Bg4 7 Be2 e6 8 Ne5 Bxe2 9 Qxe2 Be7 10 0-0 0-0 11 Bb2 Nfd7 12 Nxd7 1/2-1/2


Editor's note: After eight games, the score was tied at 4-4. The match rules provided for ties to be broken by a series of two-game mini-matches under rapid chess conditions. Spectators pressed against the glass wall of the "fishbowl" playing room to watch the exciting finish. The first two overtime games (the 9th and 10th of the match) were both drawn, but Short was able to win the 11th and draw the 12th to emerge victorious.

Short-Gulko, New York (m/9) 1994
Caro-Kann Defense B12

1 e4 c6 2 d4 d5 3 e5 Bf5 4 Nf3 e6 5 c3 c5 6 a3 cxd4 7 cxd4 Ne7 8 Be3 Nec6 9 Bd3 Bxd3 10 Qxd3 Be7 11 Nc3 Nd7 12 0-0 0-0 13 Ne2 Re8 14 Nf4 Nf8 15 g3 Ng6 16 Nh5 f6 17 exf6 Bxf6 18 Rad1 Rf8 19 Kg2 Be7 20 Nf4 Nxf4+ 21 Bxf4 Bf6 22 Rfe1 Qd7 23 h4 Rae8 24 Re2

24 ... Nxd4 25 Nxd4 e5 26 Bxe5 Bxe5 27 b4 Bxd4 28 Rxe8 Rxe8 29 Qxd4 Rd8 30 Re1 a6 31 Re5 h6 32 Kh2 Qf7 33 f4 Kh8 34 h5 Qf6 35 Kh3 Qf7 36 g4 Qd7 37 Kg3 Qf7 38 Qb6 Qf8 39 Qxb7 d4 40 Qe4 d3 41 Rd5 d2 42 Qd3 Rxd5 43 Qxd5 Qg8 44 Qxd2 Qb3+ 45 Kh4 Qf3 46 a4 Qh1+ 47 Kg3 Qg1+ 48 Kf3 Qh1+ 49 Qg2 Qa1 50 Qe2 Qxa4 51 Qe4 Qb3+ 52 Kf2 Qb2+ 53 Kg3 Qc3+ 54 Kh4 Qd2 55 g5 hxg5+ 56 fxg5 Qd8 57 Qe5 Kg8 58 h6 gxh6 59 Qe6+ Kf8 60 Qxh6+ Kf7 61 Qh7+ Ke6 62 Qe4+ Kf7 63 Qf5+ Kg7 64 Qe5+ Kg6 65 Qe6+ Kg7 66 Qxa6 1/2-1/2


Gulko-Short, New York (m/10) 1994
Slav Defense D10

1 c4 c6 2 d4 d5 3 Nc3 Nf6 4 e3 a6 5 Nf3 b5 6 b3 Bg4 7 Be2 e6 8 h3 Bxf3 9 Bxf3 Be7 10 0-0 0-0 11 Qc2 Nbd7 12 a4 Rc8 13 axb5 axb5 14 c5 e5 15 Bd2 Ra8 16 Na2 Qc7 17 Nb4 exd4 18 exd4 Rfe8 19 Be3 Nf8 20 Qb2 Qb7 21 Rxa8 Rxa8 22 Ra1 Ne6 23 Ra2 Kf8 24 Qa1 Rxa2 25 Qxa2 Ke8 26 g3 g6 27 Kg2 Bd8 28 Be2 Nd7

29 Bd3 Bf6 30 Nc2 Nb8 31 f4 Qa6 32 Qxa6 Nxa6 33 f5 Ng7 34 g4 g5 35 Kf3 h6 36 Ke2 Bd8 37 Bd2 Kd7 38 Ne1 f6 39 Nf3 Ne8 40 h4 Kc8 41 h5 1/2-1/2


Short-Gulko, New York (m/11) 1994
Caro-Kann Defense B12

1 e4 c6 2 d4 d5 3 e5 Bf5 4 Nf3 e6 5 c3 Ne7 6 Nh4 Bxb1 7 Rxb1 c5 8 a3 Nbc6 9 Be2 Ng6 10 Nxg6 hxg6 11 Be3 cxd4 12 cxd4 Be7 13 g3 Qd7 14 b4 f5 15 exf6 gxf6 16 h4 f5 17 Qd2 Bf6 18 Kf1 Qg7 19 Bb5 g5 20 Rc1 Rc8

21 Rg1 gxh4 22 gxh4 Qf7 23 Bg5 a6 24 Bxc6+ Rxc6 25 Rxc6 bxc6 26 Qf4 Kd7 27 Bxf6 Qxf6 28 Rg5 Qh6 29 Qg3 Re8 30 Ke2 Qf6 31 Kd3 f4 32 Qg4 Rh8 33 h5 Rf8 34 h6 Rf7 35 Rg6 Qf5+ 36 Qxf5 Rxf5 37 Rg7+ Kd6 38 h7 Rh5 39 Ke2 Rh3

40 Ra7 e5 41 Rxa6 f3+ 42 Ke3 exd4+ 43 Kxd4 Rxh7 44 b5 Rh4+ 45 Ke3 Kc5 46 bxc6 Kb5 47 c7 Rc4 48 Rd6 Rxc7 49 Rxd5+ Kc4 50 Rd4+ Kc5 51 Rf4 Rb7 52 Kxf3 Rb3+ 53 Kg4 Rxa3 54 Rf6 Kd5 55 f4 Ra1 56 Rb6 Ke4 57 Re6+ Kd5 58 f5 Rg1+ 59 Kh5 Rg2 60 Kh6 Rg1 61 Re2 Rf1 62 Kg6 Rg1+ 63 Kf7 Kd6 64 f6 Kd7 65 Kf8 Kd8 66 Rd2+ Kc7 67 f7 Kc8 68 Rd5 Kc7 69 Ke7 Re1+ 70 Kf6 1-0


Gulko-Short, New York (m/12) 1994
Queen's Gambit Declined D35

1 c4 e6 2 d4 d5 3 Nc3 Nf6 4 cxd5 exd5 5 Bg5 c6 6 e3 Bf5 7 Qf3 Bg6 8 Bxf6 Qxf6 9 Qxf6 gxf6 10 Kd2 Nd7 11 Bd3 Nb6 12 b3 Ba3 13 Nge2 Kd7 14 Ng3 Nc8 15 h4 Nd6 16 h5 Bxd3 17 Kxd3 Bb2 18 Rab1 Bxc3 19 Kxc3 Rhg8 20 Kd3 a5 21 a4 f5

22 Rh2 Ne4 23 Nxe4 fxe4+ 24 Ke2 f5 25 f3 Ke6 26 Rf1 Rg3 27 fxe4 fxe4 28 Rf4 Rag8 29 Kf2 h6 1/2-1/2

Boris Gulko is the only man to have won the championships of both the USSR (1977) and USA (1994). He emigrated in 1986, and from 1987-1989 he was Grandmaster in Residence at Harvard University. His award-winning article "The Mystery of Bad Bishops" appeared in ACJ 2 (1993), pp. 63-71, and was later published in translation in the Russian magazine 64. This article was published for the first time on ACJ Extra, in 1997.