Thursday, February 14, 2008

Hey Mister Tambor Man

And, behold, the veil of the temple was rent in twain from the top to the bottom; and the earth did quake, and the rocks rent

Matthew 27:51

Los tambores de Estadilla

Javier Martínez Catalan (2247) v Justin Horton, Aragón Team Championship 2008, La Aurora (Estadilla) v Casino Jaque (Huesca), board one, as promised on Tuesday:

1.d4 e6 2.c4 Nf6 3.Nc3 Bb4
I hadn't played the Black side of the Nimzo-Indian for more than two years. Partly because nearly everybody (except for me) seems to play 1.e4 now, partly because, after the last time, I spent some time experimenting with 1...d5 on the grounds that I like to play more solidly against stronger opponents. Among my reasons for returning to the Nimzo is simply that I hold it in such high regard. For me, it's the best opening for Black there is.
4.Qc2 O-O 5.a3 Bxc3 6.Qxc3 b6 7.Bg5 Bb7 8.e3 d6 9.Ne2?!
This isn't much good: it's too slow. White needs to deal with the b7-bishop either by putting the knight on f3 or by playing f3 and Bd3.

[Note: within less than three hours of writing this, I bought the latest issue of Peón de Rey which on page 22 gives the game Morozevich-Beliavsky, European Championship (Crete) 2007, in which White demolishes Black using the line I had been badmouthing a short time earlier. I should say that this does not change my opinion that 9.Ne2 is a sub-optimal move and nor does my subsequent discovery that it has been played recently by Bareev, Van Wely, Shirov, Ivanchuk, Dreev, Ivan Sokolov, Stohl, Nakamura, Sasikiran, Rowson, Flear, Cox, Vladimir Kramnik and Garry Kasparov. I add that if Alex wants to take me on in this line then I am ready for his challenge.]
9...Nbd7 10.Qd3 e5 11.Nc3 h6 12.Bh4

Black is fine here and although, immediately after exchanging on d4, I thought I should maybe have waited with 12...Re8, it's OK as it is.
12...exd4 13.Qxd4 Ne5 14.Be2?! c5!?
White criticised this move after the game: I think I'd stand by it, although it may not be the best, Black is still doing OK. 14.O-O-O is probably better than the bishop move.

Rybka, on this move and the previous one, prefers ...g5. Obviously I was reluctant to do that when White was yet to castle, but given that Black plays it anyway in the game, he may as well do so now. However...
15.Qd1 g5?! it happens, Rybka doesn't like ....g5 when it's finally played, and although fickle, it's probably right. Play might have continued instead 15...Bxg2! 16. Rg1 Bc6 17.Qc2 Ng6 and not only is White's compensation for the pawn less than clear, but it's the sort of position I'd be happy defending. I don't remember considering omitting the move of the g-pawn: I wish I knew why.
16.Bg3 Bxg2
I'd realised a few moves back that I was going to have to take this pawn, opening up the g-file in front of my king. However I decided that I might have good practical chances based, partly, on the uncastled white king: and one might as well have a pawn as insurance.
17.Rg1 Bc6
Somewhere round here, my opponent - who speaks very quickly, far too quickly for me to follow - asked me something. I didn't understand, so he asked me again - and again. Eventually, our board two, who speaks a little English, said "in the middle of the squares" and I realised he was just asking to adjust his pieces.

I have a little book on the Linares 2003 tournament which I've used to try and teach myself Spanish chess terms: but I never thought of this one (and I can't, to tell the truth, remember it today). Possibly because I've always used the French j'adoube, which I thought was universal!

I considered 17...Bh3 but rejected it on the general grounds that I didn't want to allow 18.Nb5. Had I looked a little further I might have concluded that the knight was a bit awkward after 18...Qe7 19.Nxd6 Rad8 20.O-O-O Qe6. But had I looked a bit further than that, I might have concluded that 21.f4 was likely to be terminal. It's a dangerous thing, to rely on judgement rather than analysis, but sometimes it works out...
18.f4 Ng6 19.fxg5
This is where White misses a trick. Rybka points out 19.h4!

If 20.h4 Re8 21.h5 Ne5, now that the f4-pawn is no longer present. In the previous line Black would have been obliged to put the knight on f8, a much less powerful position. Its subsequent role in the game perhaps demonstrates better than any variations how much better this would have been for White than the game continuation.

I think we both thought White was better here, with two Black pawns about to drop, but not only does Black get one back on e3 (restoring material equality) but taking the pawns puts White's pieces in vulnerable situations, of which Black can take advantage.
Maybe 20.Qxd6 but after 20...Qxd6 21.Bxd6 and then a similar continuation to that actually played, Black seems to me to be perfectly comfortable.

At some point, possibly round about here, the lights went out. I'm not sure, but I think they went out in the town generally and not just in the playing room. Some players got up, some stayed at the board, much like Fischer in Buenos Aires, until they were requested to arise. After a few minutes light was restored and play continued.

Readers would be forgiven for concluding from the standard of play that the lights had never been on in the first place.
20...Re8 21.Rxg5 Rxe3 22.Bf4
Which brings us to the situation described on Tuesday.

After White had evaded the ...Rxc3 trap, I decided to play it anyway, getting a pawn and compensation for the exchange. In truth this was both too optimistic and unnecessary: the rook could have retreated and Black would still be fine.
22...Rxc3 23.Qxd8 Rxd8 24.bxc3 Ne4

Much better than 25.Rg4, after which 25...Kh7 creates the threat of 26...f5 and it's surprisingly difficult for White to save the bishop without giving back the exchange. The move played allows the bishop to be defended from f1 if necessary and means ....f5 no longer gains a move. Black's fighting for a draw: but I'm not sure White realised this.
25... Nxc3 26.Rc1
At the time I didn't like this and preferred 26.Bh5 - which is also a good move, but no better. I think I'd envisaged getting my knight back to e4 and dominating the board from that post, but in truth it's not going to happen that way. However, the position's messy, and where it's messy there are always chances.
26...Nxe2 27.Kxe2 Rd4 28.Bg5
I saw 28.Bb8! after which it's hard to prevent both White consolidating and taking the queenside pawns.
At this point, as they say at Wimbledon, play was suspended.
Not, however, because of rain. There had been a drumming noise from outside which had become increasingly disturbing, probably because it had become increasingly loud. This, in turn, was because it had come increasingly close to the playing room, which was above a bar. Indeed, the drummers had stopped in the street directly outside the bar, this being the place where they concluded their performance. Clocks were stopped and we went downstairs to watch them finish.

Drummers - tambores - are a very old part of Aragonese culture, and although most associated with Lower Aragon (the province of Teruel) there are plenty of tambores in Upper Aragon (Huesca province) too. People form troupes of drummers, wearing uniforms, and parade through the streets. This particularly happens during Semana Santa (Holy Week) but, as became clear on Saturday, at other times as well. (I have, for instance, seen drummers at SD Huesca matches. And heard them.)

I am far from sure how old the tradition is - which, apparently, seeks to recall the noise of the earthquake that is supposed to have happened at the death of Christ. (The sound of an earthquake is not generally considered to constitute normal playing conditions. Though there was the time I played at Bury St Edmunds early in November and there was a firework display outside.) They are mentioned in a sixteenth-century account of a visit to an Aragonese parish and it's clear that it was not a new practice even then. Given that modern chess was introduced to Spain not long before this time, it's possible therefore than Aragonese chess has been disturbed by tambores for the entire half-millenium-plus of its existence.

I didn't much like my position when we left it: I thought I might be hanging on because I could try and attack the c-pawn, but in truth I shouldn't get time to do that. Fortunately, on our return my opponent chose a questionable plan.
As yet this doesn't spoil anything, but only because, being too keen to get my king off the g-file, I didn't play 29...Bg4+! after which the knight comes to e5 and Black should be OK again. However, I think I'd decided that White had a further mistake in mind, and he surely did.
29...Kh7?! 30.h5?
30.Bf6! probably wins, as 29.Bf6! would have done. By creating threats, White forces Black to remove his pieces from the present good positions, and thereby wins time and space to get his pieces into play.
30... Ne5
Now Black gets the second pawn for the exchange and his pieces are freer. He's probably not winning, yet, but it's a struggle for White, made harder by the knowledge that he's basically thrown the second pawn at Black without his opponent having to do anything to win it.
31.Rc3 Bg4 32.Kf2 Bxh5 33.Rh3 Kg6
It took a little nerve to play this. More importantly, it took a little time - Black now had only a couple of minutes left on the clock.

If there's a point where White's position becomes lost, it might be here: at least 34.Bf4+ allows the bishop to go to somewhere like b8 and make trouble whereas from e3 it can only retreat or be exchanged. I am quite sure my opponent didn't think he was winning the rook: but he too was short of time, he was the one with a deteriorating position and he had to find a move, so he chose one. Very difficult circumstances.
I was absolutely sure I was going to win this game now.
35.Ke1 Re4?!
And then I became unsure. Isn't the bishop on h5 a bit loose? Might it drop if the rooks get at it?

Calculating variations isn't my forte at the best of times: with practically no time left (even allowing for our extra thirty seconds a move) my capacity to think concretely disappears almost completely. I can, often, stay out of trouble and avoid putting anything en prise: it's a way of avoiding defeat. But it's no way to finish off won games.

35...Rxc4! just grabs another pawn - there's no need to pin the bishop which isn't going anywhere. 35...Rd3 would also be strong. With time to work out the answer to the question how do I go about winning from here? I'd no doubt have seen both. As it was, I only had time to think of a safe move, briefly check it and then play it.

But it's not spoiled yet.
36.Kd2 f5?
Now it is. 36...Rxc4 is still the right idea.
37.Kc3! is the move, securing the c-pawn. Black's kingside pieces can hardly move.
Losing my nerve with just a few seconds left. 37...Nxe3 picks up at least one more pawn and with it, the game. There was nothing to be scared of in 38.Rxh5 as 38...Nxc4+ gets the a-pawn as well. In fact, if anything White is handicapped by taking the bishop as it means his h1-rook is temporarily tied to the defence of its colleague at h5.

38...Bf3! is still winning. It frees the bishop from the rooks' attack, which means the knight can rejoin the fight for c4 (via d7 and e5). It wins time to do this by attacking the h1-rook and if 39.Rg1+ then it can drop back to g4 and the rooks are no longer doubled.

It's often in the nature of games where wins are missed that they get gradually harder to see once you've missed the first one. The c4 capture is harder to see when you've already not played it the first time it became available: the bishop move is harder still, it being an aggressive, attacking move when what you're thinking of doing is keeping it tight and not dropping a piece with your clock about to run out.

But now the win is gone and White has the upper hand again.
39.Rxh4 Bg4 40.Rh8
This is exactly the sort of thing I needed to avoid. Now White's rook gets among the queenside pawns, while Black's passed f-pawn is curiously static.
40...Bf3 41.Rc8 Ne4 42.Kd3 Nd6
I'd been planning the blunder 42...Bg2 with the idea of 43...Bf1+, but fortunately I saw White had 43.Rg8+ with the idea of 44.Rxg1. But now Black's going backwards.
Natural, but Rybka thinks White is winning with 43.Rg8+ first, as after 43...Kf7 44.Rd8 Black can't hold: either 44...Ne4 45.Rd7+ and the a-pawn falls, or 44...Ke6 45.Bf4 Nf7 46.Re8+ Kd7 47.Rg8, the point of which is that White will play Bb8 and Rg6 and win a pawn on the queenside.
In fact Black has a bizarre means of escape involving 43...Be4+ 44.Kc3 Bd5!! 45.Rd7 Nxc4!! but who's likely to see that?
44.Ra7 Bg2 45.Bxc5
I thought this was an error at the time (possibly because I still didn't appreciate that I was losing) but my preferred 45.Rxa6 isn't so good because the rook gets into trouble after 45...Be4+ 46.Kc3 (or 46.Ke2 Nxc4 47.Bxc5 Bxb7 48.Ra7 bxc5 49.Rxb7 Nxa3) 46...Nc8 47.Bxc5 Bb7 48.Rxb6+ Nxb6 49.Bxb6, when Rybka reckons White is about 0.60 better, and I do not.

No! 46.Ke3! Nxc4 47.Kf2 bxc5 48.Rxa6+ Kg5 49.Kxf1
Now White has just a pawn for the exchange but the c-pawn must surely fall. Is he winning? He has all the chances, but if you want a definitive judgement you'll need to ask a better player than I.
46...Nxc4 47.Rxa6 Ne3 48.Bxe3
and a draw was agreed, on nobody's particular proposal. I was furious with myself, afterwards, for not having won the game. But I cheered up a bit after I'd realised that I was actually losing at the end.

Not a great game. Not even a good one. But a dramatic one, for sure.

No comments: