You gotta have faith …

Because George Michael said so!

In my last tournament I had a disappointing finish, a draw in the last round against a class-A player named Roberto Aiello. In this game, as White, I experimented with a variation of the King’s Indian Defense that I have never played before: the Makagonov Variation (1. d4 Nf6 2. c4 g6 3. Nc3 Bg7 4. e4 d6 5. h3). It’s hard to draw conclusions about an opening from one game, but I hated the position I got. It just seemed as if I was a tempo behind in development. I was actually relieved to exchange down into a blocked position where neither side had any chances.

Last night I went over the game, first by myself and then using the computer. I had to settle for the not-so-reliable Shredder instead of Rybka, because my wife is out of town with the laptop that has Rybka installed.  🙁

Anyway, it became clear that I greatly underrated my position, and that was the main reason that I had to settle for a draw. I gotta have faith! (Of course, it’s hard having faith in a variation you have never played before.)

Here’s the position after Black’s 15th move:

Black has just played the rather unusual-looking 15. … Qf6. It is very rare for Black to voluntarily block his f-pawn in the King’s Indian, and my gut reaction was that this move couldn’t be good. My first inclination was to finish my development with 16. O-O-O, preparing 17. … Rdg1 with a kingside attack.

But what happens after 16. O-O-O Nf3? Obviously this is the reason that Black played the funny-looking … Qf6. White doesn’t want to allow the knight to come to d4, so 17. Bxf3 Qxf3 is forced. And here I spent several minutes looking at 18. Rhg1 (a pawn sac) and 18. Rdg1. I didn’t see compensation for the pawn after 18. Rhg1 Qxh3 (especially since the g4 pawn is also under pressure). And 18. Rdg1 looked too slow. It takes two moves for White to evict Black’s queen (Rh2 and Rg3) and meanwhile Black can be organizing pressure on the queenside with moves like … Ba6 and … a4.

However, I don’t think that I seriously analyzed the most obvious move (after 16. O-O-O Nf3 17. Bxf3 Qxf3): 18. Bxh6! I mean, I must have looked at it, but I don’t have a conscious recollection of why I decided not to play it. Most likely after 18. Bxh6 Bxh6 19. Qxh6 Qxf2 I felt that Black’s queen on the seventh rank (my second) would be very dangerous, especially after Black plays … Rb8 threatening checkmate on b2.

But in fact, after 19. … Qxf2 Black is losing! First we play 20. Rhf1 Qg2, which is pretty obvious. Do you see what comes next?

Position after 20. … Qg2 (analysis). White to play and win.

While you’re thinking that over, let me say that this is the sort of position where your attitude makes all the difference. If you have faith in the position, you’ll be looking for ways to (hint!) take advantage of Black’s awkward queen position. If you don’t have faith, you’ll see the queen as menacing, rather than awkward, and you won’t even go into this line. Even last night, three weeks after the game, sitting in my comfortable chair at home with no tournament pressure on me, I still didn’t go any deeper. I had no clue that White was winning until the computer told me so.

The winning move here is 21. Qe3!, and Black’s queen is trapped! The threat is 22. Rf2, and the only way for Black to prevent it is to give up his bishop with 21. … Bxg4, with an obvious huge advantage for White.

By the way, 20. … Qg3 (instead of 20. … Qg2) was no better because White just plays 21. Ne2 chasing the queen to g2 or h2.

Ergo, Black’s 19. … Qxf2?? is just bad. A better try is 19. … Ba6:

Position after 19. … Ba6 (analysis). White to play.

Here again, my pre-computer analysis was really shoddy. I looked at the obvious move 20. h4! and I thought, “Well, after 20. … Qxg4 21. Rdg1 Black gets out of trouble with 21. … Qf4+! 22. Qxf4 ef. I have no attack any more and my c-pawn is weak.”

Which is completely wrong! After 20. h4! Qxg4 21. Rdg1 Qf4+ 22. Qxf4 ef 23. b3 White’s queenside is solid (or at least solid enough) and Black has serious problems holding his kingside together because of the very weak pawns on f4 and g6. Moral of the story: just because queens are off the board, it doesn’t mean the attack is over.

It’s an interesting catch-22. I didn’t have faith in the position because I didn’t see all these tactical possibilities. I didn’t see all of these tactical possibilities because I didn’t have faith in the position.

Now let’s go back to the original diagram. For all the above reasons, I came to the mistaken conclusion that Black was doing well after 16. O-O-O Nf3 17. Bxf3 Qxf3, and so reluctantly I played the move 16. Bxg5 Qxg5 17. Qxg5 hg. Because of the blocked position White has no chances of winning. And if White tries to open the position up at all it will only work to the advantage of Black’s two bishops. So I agreed to a draw seven moves later.