The Interesting Sacrificed Exchange III

Black to play

Movsesian - Kasparov, Sarajevo 2000

Another Tuesday bonus post. Why not? Especially since last Wednesday's post has inspired me to get this series rolling at long last.

For today's TISE we journey back in time to join Movsesian and Kasparov for move 13 of their game at the Sarajevo Grandmaster tournament of 2000. Gazza is about to make a very good move and while I'm sure you can guess what it's going to be, the title of today's post being a bit of a clue, you might want to have a think about why he played it before reading on.

Of all the goodness-knows-how-many games of chess that I played before starting to write about exchange sacrifices last June, I can only think of two in which I gave up a rook for a minor piece myself. Actually, even those don't really count: the first was an accident - I played ... Rxf3 (knight) thinking I'd end up winning material, but I'd miscalculated and had to pretend that I'd done it merely to break up his kingside pawns - and the second wasn't truly a sacrifice.

White to play

JMGB v Bloke, London League circa 2000

Here I 'gave up' the exchange with 21 Rxh5, but really it just leads by force to a position that is concretely better for White. In six moves time I will be queen for rook ahead and any attempt by Black to avoid that fate will only lead to something even worse.

I was pleased to play it at the time, but rook takes knight is, at best, a temporary or ‘numbers’ sacrifice and not at all difficult to grasp.  Even I can understand that 9 is bigger than 5 and that mate ends the game. Much more interesting is what went on over on the other side of the board. White’s doubled c-pawns and the absence of Black’s queen's rook rather suggest that rook takes knight on c3 has been played at some point and the game had indeed opened in a manner familiar to all Dragonistas.

1 e4 c5, 2 Nf3 d6, 3 d4 cxd4, 4 Nxd4 Nf6, 5 Nc3 g6, 6 Be3 Bg7, 7 f3 O-O, 8 Qd2 Nc6, 9 Bc4 Bd7, 10 O-O-O Rc8, 11 Bb3 Ne5, 12 h4 Nc4, 13 Bxc4 Rxc4, 14 h5 Nxh5, 15 g4 Nf6, 16 Bh6 Nxe4, 17 Qe3 and now Rxc3

The more experienced chessers amongst our esteemed readership won't need me to tell them that … Rxc3 is a common idea in the Sicilian Defence in general and the Dragon in particular - and that it is not played with number compensation in mind. Knowing that and understanding it are two very different things, however. I suspect that, like me, my opponent was aware that he was supposed to sac the exchange, but didn’t truly know why. That might explain why he spent 40 minutes on his 17th move and then took another quarter of an hour on his following turn.

Anyhoo, it was stumbling across the scoresheet for this old game (stored in traditional pre-Chessbase 'shoved-in-a-shoebox-under-the-bed' fashion) that reminded me of Movsesian-Kasparov, a game I had come across in Jacob Aagaard's Excelling at Chess.

So, from the diagram at the head of today's blog,

13 … Rxc3, 14 bxc3 Qc7, 15 Ne2 Be7, 16 g5 and now 16 … O-O (! – Aagaard)

turning the battle into one of competing opposite-wing attacks. Movsesian, according to Aagaard, had played this line against Loek van Wely in many internet blitz games, but the Dutch GM hadn't risked castling in a single one of them. How to explain the difference of opinion as to whether it's safe for Black to walk into a White's kingside pawn storm? Let us quote The Voweled One quoting Kasparov:

"From my perspective it’s a matter of chess culture. If you take on c3 and the knight goes to a4, then Black is fine. Black need not look for an immediate approach. You castle, you put your knight on e5 and the queen on c7 or a5, and you have many options. Sometimes you strive for d5 or even for f5. The exchange means very little, since we both have such attacks going, the quantity of pieces is often more important than their quality."

Aagaard himself made precisely this last point when discussing his win against Lindberg in Attacking Manual 1 – the game that inspired this series in the first place. If I understand the argument correctly, the idea is that after … Rxc3, bxc3 Black will sooner or later follow up by swinging his king’s rook over either to c8 or somewhere similar. As a result Black is net +1 for pieces on the queenside, has got rid of White’s best defender and the doubled pawns might well make it difficult for White to transfer reserves across to even up the numbers. In consequence Black must have a decent attack and can castle short safe in the knowledge that the enemy king will always be in as much if not more danger than his own.

Simples? Well, not really. After all, castling short didn't occur to Movsesian and van Wely so I guess it can't be that obvious.

When I first played through Movsesian - Kasparov I found the exchange sac much harder to understand than the one from Aagaard-Lindberg. I'm not sure that's any less true today, but, still, I do feel that by looking at the game my understanding of these sacrifices inched a little bit further forward. Just a tad more comprehension, then, and certainly not a grasp of the 'chess culture' thing that Gazza talks about, but it was enough to help me to go on to play an exchange sacrifice - a real exchange sac - in one of my own games. That story, though, I think I'll save for TISE IV.

Previously:-
TISE
TISE II