63...Be8; 66...RxR

That's all he has to do to get the ball rolling. Instead he goes after the K w/ ...Qf2+

This is not brilliance, its: one guy has no clue.

*****

All right, forget the winning plan for black he obviously has no clue.

But the game was still not lost, so the next question is:

It will help the student much more to try to show where the game was lost.

1) The queen sac by Shirov is not sound. After 22) xh4 Rybka eval is -1.14. But ...Qb6 is a lot better than the move played.

2) Black maintains an edge in the eval of at least .3 for 20 moves. 43) ..f2 is not that good though, the game is now about even. Better was Rg6 or ka8 (dono't ask me, Rybka move).

3) About move 54 Rybka first evaluates the position as advantage white. Now for the next 8 moves, 54-62, the game is either even or about +.1 for white.

You are not the only one.

First let's rewind to the early middlegame after white has played 18. Kh1

click for larger view

Black's plan seems fairly clear. He is going to castle on the queenside, park his rooks menacingly on g8 and h8 and throw absolutely everything at the white king. Moves like Ne7/g6, Ng4 and Qg7 or Qh7 spring to mind. Double or even triple heavy pieces on a half open file and bad things will happen. Very bad things indeed.

Black doesn't mind if white grabs the h4 pawn. It's only getting in the way of his kingside attack. This is crude caveman chess, all bare hairy chest, gold medallion, cheap aftershave and seventies glam rock. Knight Rider (first time around), anyone? But horribly effective for all that.

For now, black's problem child is his Bd7, which is hemmed in by the black pawns on e6 and d5. In other words, the archetypal bad bishop. But everything else has a part to play in the formulaic black kingside attack.

And what has white got by way of compensation? Well, he has a sort of queenside attack, but it is a pawn short of a breakthrough. We'd like to sink a knight or two onto d4, but this is going to need some preparation. All in all, the black stonewall pawns are really cramping our style in this position.

So what kind of plan could white come up with? The standard idea of attacking on the queenside doesn't seem to promise much. Black will just lock queenside pawns and continue hammer away on the kingside.

A central attack also seems doomed to failure. The strong bulwark of the black stonewall seems way too strong.

Shirov's plan is simply breathtaking. He swaps his queen for a black rook, knight and pawn. Few of us mortals would have the cojones to do this, even though all the beginner books say that this is a fairly level trade - a queen worth 8 or 9 points for a rook (5), knight (3) and pawn (1).

But look what Shirov achieves with this trade. He gets rid of two of the black pieces which were poised for the attack and by doing this accentuates the weakness of the Bd7. The fewer pieces there are on the board, the weaker a bad bishop usually becomes. Shirov also gives himself a plan - push the h pawn and swap off any black pieces into the endgame.

This is not a decisive advantage, but it stops black's plan (nearly always a good thing) and unbalances the game. And that gives Shirov something to work with rather than sit back and watch black throw rocks at him. Sure it takes many many moves to do this, but all the time black is having to defend. And suddenly black is the one without a plan.

Well, it's a theory...>

And not a bad one at that. By my reckoning most of it is very plausible. The bit I don't quite buy is <Shirov also gives himself a plan - push the h pawn and swap off any black pieces into the endgame.>

Looking back at the game, it seems likely that Shirov had two plans in mind. The first was to gang up on the weak e6 pawn and the second was to push h4. I suppose that's pretty classical stuff around playing against two weaknesses. I suspect that Shirov didn't know which of these would come good, but at least it gives him something to play for.

But in this game, the trade is more balanced - a queen for a rook, knight and pawn. And here it is much more tricky for the side with the queen. If he just swaps off all of the pieces, white emerges with an extra passed pawn, which ought to be enough to win. And it is not easy to engineer a fair trade if you have a queen against lots of minor pieces, as you need to get two or three minor pieces in return for the queen.

Two rooks, in particular, are often more than a match for a single queen - notice that Shirov is careful not to allow black to exchange a pair of rooks. Shirov is also careful to keep his pieces protected as much as possible. In queen versus lots middlegames, the queen excels at picking off loose pieces through fork, pin and skewer tricks.

Shirov vs Vallejo-Pons, 2002.
YOU ARE PLAYING THE ROLE OF SHIROV.
Your score: 118 (par = 110)