Monday, January 20, 2014

A Blogger Goes Chessing in Hampstead: Close but no ISE

Author's note:
Hampstead was at Golders Green this weekend.  More on that later in the month. In the meantime, here's something from 2013.

7 Ne5
JMGB - Ali Zarrar, Hampstead u2200 October 2013

I don't have many examples of exchange sacrifices from my own games that I can include in my year of exchange sacs. Neither I nor my opponents have shown much inclination to give up a rook for a knight or a bishop. It's part of the reason for doing this series, I suppose.

I did come closeish at the Hampstead u2200 back in October. One of those games when a couple of amateur hacks blundering about in the dark manage to (very nearly) reproduce the opening moves of a game played in a World Championship match. An infinite number of monkeys and all that.

Close, but no ISE as it turned out.

My scoresheet tells me that I spent three minutes over 7 Ne5. Most of that time, as I recall, was pondering whether I wanted to give Black the chance to give up his rook on a8 in exchange for my Catalan bishop. The two games that I had in mind that were helping me making up my mind couldn't have made a greater contrast: one of my own from the Croydon League five or six years ago and Game 7 of the 2010 Anand-Topalov World Championship match.

The guy I was playing in Hampstead didn't have a rating and struck me as rather inexperienced. I got the impression he played a fair bit on the internet and was just starting to take his first steps playing face to face games. For that reason alone, I felt I didn't need to worry about him giving up the exchange because those kind of guys really don't like to give up material for abstract compensation.

White to play
JMGB v Some Guy, Croydon League ages ago

Pretty early on during those three minutes that I was thinking about playing my knight to e5, that old Croydon League game came to mind. I recalled Black playing 5 … Be6 and then telling me after the game that as soon as he'd released the bishop from his hand he realised that he'd allowed 6 Nxe5 fxe5, 7 Bxb7 "winning the rook".

Black to play

My feeling during the game was that he could simply go 7 … Nbd7 when after 8 Bxa8 Qxa8

White to play

it looked to me that for my tiny material plus I had no development, nowhere obvious to put my pieces and not much in the way of light squares on the long diagonal. When I told my opponent that he shot me the sort of disbelieving frown that I might have expected had I said that I'd made friends with the pixies who live at the end of his garden. He clearly felt that I'd simply missed a chance to get a winning advantage right out of the opening.

Back to Hampstead

So, returning to the Hampstead game, I was fairly sure that I wasn't going to see 7 … b5, 8 Nxc6 Nxc6, 9 Bxc6+ Bd7

White to play

and that Black would in fact just castle or perhaps develop his queen's knight to d7 instead. That proved to be correct. Almost immediately 7 … 0-0 appeared on the board and I breathed a sigh of relief, grateful that I wouldn't have to find out what was going on if he had provoked me into grabbing the rook.

Hampstead variation: 10 Bxa8 Qxa8

Now unlike the Croydon league guy, I believe in this kind of exchange sac. I'm just not equipped to evaluate whether any particular instance of it is any good or not, that's all. It's an improvement of sorts, I suppose, albeit not a massively helpful one.

I felt at the time, and it still seems to me now, that the position I might have got at Hampstead is better for White than the line that I had rejected all those years ago. The relative development of the two sides is about the same in both cases, but White seems considerably more cramped in the Croydon League position.

Hampstead and Croydon

The should have led to me feeling optimistic about taking on a8, I suppose, and yet still the prospect that the game might actually proceed this way rather alarmed me. The reason why was because the position that I might end up with reminded me of a famous game from the Anand-Topalov game. Not so famous that I could recall the actual moves, but I did have a strong memory of Black giving up the exchange and getting an enormous amount of play - not to mention an hour on the clock -  in return.

When Hampstead was all over I discovered that my position and the one from the World Championship were even more similar than I'd imagined. In fact all you have to do is move White's bishop to f4 and have Black castle and you get an exact replica.

u2200 and World Championship

Had I realised this I think I'd have been even more worried than I was at the time although, on sober reflection, the one small difference between Anand-Topalov and JMGB-Zarrar seems to push the position further in my favour. With my queen's bishop still at home I could continue with f2-f3 like Vishy did and Topalov's … Nf6-d5 wouldn't gain a tempo. I suppose the fact that Black's rook is still on its starting square means that it can support a banzai attack with … h7-h5-h4. That would mean leaving his king in the centre of the board, though, and when White eventually gets his pieces out that's going to prove to be problematic, isn't it?

About to demonstrate a slightly higher standard of play than you get from me at Hampstead

So, to recap:-

I never got to grab an exchange in the end, and had I done so I've no idea how the game would have gone, but if I had happened I reckon I would have had an improved version of Anand-Topalov and better than I would have got had I chopped that rook in the Croydon League.

I'm not (quite) as scared of giving up my fianchettoed rook as I once was and if an opponent plays down the same line again I'd certainly give 7 Ne5 another punt. Still, it's early days with this exchange saccery. Anybody who wants to point out precisely how and why I've got it completely wrong is more than welcome to make a contribution to the comments box.

2014 ISE COUNT: 4
TISE Index
Tournament Diaries Index

Hampstead tournament hall from Adam Raoof's Flickr
Vishy & Toppy from Chess Vibes


Anonymous said...

At move 5 in your Croydon League game, you are following the footsteps of Capablanca against Sergeant at Margate in 1936. Neither Sergeant nor the handful of subsequent players have played 5. ..Be6, so the exchange sacrifice is untested in practice. For what it's worth, a computer engine suggests 6. Qb3 as an advantage to Black (6. .. Bc8 being Black's reply). 6. Nxe5 is evaluated in White's favour.

In the Catalan position, once again there are no practical examples of a 7. .. b5 punt. Here again the engine prefers to seek an advantage with thematic queen-side play such as b3 or a4. There's not much in it, as it evaluates taking the exchange as nearly just as good for White.

It's a critical positional point in the Catalan and related positions as to whether Black can play b5 or not.


Anonymous said...

6 Qb3 as advantage to White and 6 Nxc6 as advantage to Black is what I meant to say.


Jonathan Rogers said...

Generally speaking, sacrificing a rook on a8 or a1 for a bishop, and recapturing with the queen, is likely to recur a lot in this series. Hope you have Shirov v Bareev, Biel 1991 or similar, at some point.

Anonymous said...

Erm. I think Black's extra piece in the Croydon League game might also be relevant?

Retired FM

an ordinary chessplayer said...

I once watched a 2100 lose to a 1400 in the English Opening (1.c4 e5 2.g3) in just such a position. Two decades later I don't remember the exact game of course, but Herr Systemssen as white probably played (as he usually does) the sequence Bg2, Nc3, e2-e3, when black simply moved ...Bc8-e6 and was down the Exchange right in the opening. I know it was before white castled. Except that after ...Qxa8, f2-f3, I don't think black had ...Be6xc4, so maybe white had varied from his usual?

Anyway, then they played chess.

Besides the chessic features that matter to the likes of Anand and Topalov, there are a couple of psychological features that favor a lower-rated black here.

(1) Black knows the correct plan. In your average position, that may not be true, but in this type of position, black knows that he must attack. No dithering necessary. Even better, it's the simplest possible plan, so black is likely to play good moves for quite some time.

(2) Black has nothing left to lose. Having already fallen victim to white's one big trick, all black's remaining pieces are safe from counter attack. And any future white threat will be seen from a mile away.

Put those two psychological features together and your typical 1400 player is I daresay more like an 1800 player. Which makes him the most dangerous player of all: the UNDER-RATED player.

As for those chessic features, I recall playing through the game (I was on friendly terms with both players), and concluding that white is not simply winning, it's far from easy. As it went, white just waited around for the next "1400" mistake, which never happened. Instead black obtained firstly definite compensation, secondly a strong attack, and thirdly a bucket of pawns.

John Cox said...

It seems to me that 6 Nxe5 would be a very strange move. I'd be surprised if Black wasn't just much better.

Matt Fletcher said...

I agree - Nxe5 in game 1 looks very odd. You haven't even really got any material for it (R+2 vs B+N) and your light squares are wide open - what do you even play after Qxa8? f3 Nf6 etc just looks horrible. Qb3 looks like a much more respectable option.

In the second game, the exchange sac seems much less sound - it's properly Rook for Knight this time, the unopposed Bishop can't get out as quickly because of e6, plus f3 is a much more sensible move in this position because you can play e4 and block off the long diagonal.

Jonathan B said...

Thanks for the comments chaps.

Shirov-Bareev was the very first ISE, Jonathan. I reckon it was long enough ago that I could get away with rehashing it without anybody noticing. *Stares hard in general direction of a certain national newspaper columnist*