Ahmed-McNab, Khanty-Mansiysk 2010.

Another attack from the recent Olympiad. Black to play.

RB If chess pieces could talk, they would be screaming at me something like: “Can’t you take a hint, dummy? We’re all pointing in one direction – at c3!” And I would say back to them: “Yes, thank you, I see that, but I have to tell you that I am always nervous of sacrificing a piece when I can’t calculate a forced mate or win of material.” But let’s try to get the nerves under control and venture the less valuable piece first: 1…Bxc3. What’s to stop White playing 2 bxc3, and if 2…Qxc3 3 Bb3, with threats against f7? That doesn’t seem to help Black much, so what about the rook: 1…Rxc3 2 bxc3 Qxc3 – and, yes, I can see the point of this. White is exposed along the a1-h8 diagonal and would have to play 3 Kc1 (or 3 e5 – is that better?) when Black has 3…Qa3+ 4 Kd2 Bh6+ and all sorts of possibilities to harry the king and pick up material en route, though White, of course, may have a better defence.

DK The rook sacrifice is screaming to be played – and this can be a problem. There is a lot of pressure to go for glory: you see your team-mates looking over at your board with a knowing smile, as if it’s the simplest thing in the world to polish off the enemy king, but they’re not the ones who have to risk giving up the rook.

Let’s look closely: 1…Rxc3 2 bxc3 Qxc3 leads to mate and the variations aren’t too difficult, but 1…Rxc3 2 Rhf1 is a concern, and it’s the kind of defence that is easy to miss, especially by your overexcited colleagues. After 2…Qxh3? White can take the material as the rook on f1 covers some important checking squares. Instead, McNab had prepared a crusher: 2…Rxc2! so that if 3 Rxf3 Rxb2+ 4 Kc1 Rc8+ wins. The game continued 3 Kxc2 Qe2+ 4 Qd2 Qxe4+, and for the cheap price of a rook for bishop, Black had three pawns and a continuing attack that quickly brought success.