Analysis by Vasik Rajlich

What happened in this game is just stunning. How is this possible? Well, there is a simple explanation. As Lasker used to say, chess is a struggle. In even the most lost position, resistance can be put up, even more so if there remain complications. And if you leave yourself little time to win the won position, then A) you are taking a risk, no matter how you look at it B) you absolutely need to do what you can to eliminate this risk. I'll elaborate on this in the notes.

V. Rajlich - E. Salanki
[B08/01] Pirc: Two Knights
4 Jan 2000

1.e4 g6 2.d4 Bg7 3.Nc3 d6 4.Nf3 c6

In my opinion these systems are simply inferior. Black heads for a kingside-queenside race where he just doesn’t have that much on the queenside. Compare his play there to a Yugoslav Dragon.

5.Be3 Nf6 6.Qd2 b5 7.Bd3 0-0 8.Bh6 Qb6 9.0-0-0 c5

This is one of the worst blunders that I have ever seen an 2300+ FIDE make.

10.Bxg7 Kxg7 11.e5 dxe5 12.dxe5 Ng4 13.h3 Nh6 14.Bxb5 Be6

Aside from being up a pawn, White also in my opinion has the better position. There is absolutely no doubt in my mind that it is more than winning. Furthermore, I had spent almost no time to reach this position. The rest is just mop-up duty, right?

Well, it’s a little bit more complicated than that. Of course White would love to just chop off all of the material, but there is no easy way to do that, and with the kings on opposite wings it is inevitable that White’s king will come under some attack. I think that defending against Black’s attack while trying to nurse the extra pawn would have been the wrong thing to do. I decided that White should attack as well, even at the cost of complications. If well-played, White’s attack will be no weaker than Black’s and will eventually force what White really wants: simplification.

I had a game with this theme against Balogh in Budapest. I had a passed h-pawn, and any simplification would have just led to an awesome ending for me. So, I tried to just defend my king against his attack. Unfortunately, he was able to build up his attack, and even though the game ended with a blunder by that point the attack was quite strong. It turns out that I should have counterattacked. The counterattack would have been easily defused, but only at the cost of simplification, after which my h-pawn would have become the main feature in the position. I decided that a similar thing was going on here.

Of course there are surely other ways for White to keep a winning position, but I feel that the method I chose was in fact best.

15.h4 Ng4 16.h5 h6

Now, White must calculate very carefully before proceeding. If he gives Black the d-file and the d4 square without a breakthrough on the kingside, then he will have wrecked his position.

17.Rh4 Nc6 18.hxg6 fxg6 19.Rdh1 Rad8 20.Rxh6

This was in fact correctly calculated, and it does win. It also seems to show the correctness of White’s energetic play starting with 15.h4. However, I consumed a tremendous amount of time on this and my 17th moves, and after I played this I only had about ten minutes left. I don’t necessarily think that getting so low on time was a mistake, such moves need to be checked carefully, and as I mentioned above I think that it was correct to complicate the position even after Black’s blunder in the opening. However, such a path entails a risk, as the remainder of the game clearly shows.

20...Rxf3

My opponent didn’t speak English (and I didn’t speak any language after this game). I’d be curious to see how much he saw of the 20...Rxd2 variation, I thought that it was pretty complicated. It runs 20...Rxd2 21.Rh7+ Kg8 22.Ng5 (only move, 22.Rh8+ Kf7 23.Ng5+ Ke8 isn’t enough) and now:

A) 22...Ncxe5 (no reason to prefer 22...Ngxe5) 23.Rh8+ Kg7 24.R1h7+ Kf6 25.Nge4+ Kf5 26.Rxf8+ Nf6 27.Nxd2 (or even 27.Kxd2) and White will win.

B) 22...Rxc2+ (to rescue the f8 rook first) 23.Kxc2 Rxf2+ 24.Kb1 Ncxe5 25.Rh8+ Kg7 26.Nce4 and mate can’t be stopped.

Of course after 20...Rxf3 Black is lost, but chances exist. What happens between now and move 40 is a chessplayer’s worst nightmare.

21.Rh7+ Kf8 22.Rh8+ Bg8 23.Bc4 e6 24.Qg5 Ncxe5 25.Bxe6

Mistake #1. This move is way too fancy, White should just play 23.gxf3 Nxc4 24.Qf4+, ending up a clear exchange ahead, when he really can just await Black’s resignation.

25...Rf7 26.f3

This is just an oversight, I missed Black’s next move. Best seems just 26.Nd5, with an easy win.

26...Rb8

This woke me up! But, White is still winning.

27.Bb3

With this move White misses his last chance for an outright win. As it turns out, White can survive Black’s counterplay after 27.Bxf7, ie. 27...Qxb2+ 28.Kd1 Kxf7 (28...Nxf7 29.Qxc5+ and 30.fxg4) 29.fxg4 Nc4+ 30.Kd3 and Black can’t put the finishing touch on his attack.

White is still better after 27.Bb3, but it is no longer winning.

27...Qf6 28.Ne4

Objectively speaking here White had a choice between 28.Qxf6 Nxf6 29.Bxf7 and 28.Ne4 Qxg5 29.Nxg5 c4 30.Ba4 Rb4 31.b3 Nf6 32.Nxf7, etc. Both of these endings will be better for him, but neither will be anywhere near winning.

28...Qxg5 29.Nxg5 c4 30.Ba4 Rb4 31.b3 Nf2

Black attempts to show that 28.Ne4 loses outright, but in fact this move itself is a blunder which should lose.

32.Rf1 Rf5 33.f4

After this Black has extricated himself and keeps his extra piece. However, White could have won here with 33.Ne6+ Kf7 34.Rxf2.

33...Neg4 34.Ne6+ Kf7 35.g3 Rf6 36.Ng5+ Kg7 37.Rh4 axb3 38.Bxb3 Bxb3 39.axb3 Rc6 40. Kb2 a5

My opponent managed his time much better than I did, but he too made his fortieth move with just a few seconds to spare. I imagine that his first reaction after reaching the time control was the same as mine: White is lost. In fact, however, White has some play which seems to enable him to hold the balance. I imagine that some improvement somewhere between move 35 and 40 for Black would have led to a win.

41.Rh7+ Kg8 42.Re7 a4

After 42...Rb8 43.Re2 I don’t think that White is any worse. So, my opponent decides to give up an exchange.

43.Re8+ Kg7 44.Rfe1

At first I thought that I might actually stand better and be able to play for a win myself after all, unfortunately right after giving up the exchange Black can immediately win a pawn, so I think that it is still White who is aiming to draw. But it seems that he has no major problems.

44...Rb7 45.Ne6+ Rxe6 46.R1xe6

If 46.R8xe6 then 46...axb3 wins the pawn back immediately.

46...axb3 47.cxb3 Kf7 48.Re2

Forced, else 48...Nf6.

48...Nd3+ 49.Kc3

If 49.Kc2 then 49...Nc5. I decided to return the exchange rather than lose the b-pawn, the knights might be pretty good if the play is localized to the kingside. But, my opponent has other ideas:

49...Nc1 50.Re1 Rc7+ 51.Kb4 Nd3+ 52.Ka5 Nf6 53.R1e3 Nxf4

My opponent wants a pawn after all, even if it’s just the f-pawn. I’ll just give the rest of the game without comment, I think that White is pretty comfortable throughout.

54.Rd8 N4d5 55.Rd3 Ke6 56.Rd4 Rc3 57.b4 Ke5 58.Rd2 Rxg3 59.Re2+ Kf5 60.Rf2+ Ke4 61.Re2+ Kf5 62.Rf2+ Ke6 63.Re2+ Kf7 64.Rd2 Ra3+ 65.Kb5 Nc3+ 66.Kc6 Ra6+ 67.Kb7 Re6 68.R2d6 Re1 69.Rc6 Ncd5 70.b5 Rb1 71.b6 g5 72.Rcd6 Nxb6

If I were Black I’d probably torture White a little bit longer myself, but I guess it’s been a long game. Also, it is not inconceivable that Black will lose - several blunders more than bad enough to lose this have been played in this game alone.

73.Rxb6 ½-½.