In the following game, I found myself in a bleak middle game after being outplayed in the opening by my opponent (rated 2300+). My opponent had created a backward pawn on my queenside and was calmly lining up his heavy pieces on a half open file to exploit the weakness. With his kingside depopulated and several of my pieces pointed at g6, I decided that the most prudent course would be to sacrifice a bishop and rook to achieve perpetual check; I would live to fight another day. But while confirming that it really was a forced draw, I found that it was actually a lucky/brilliant win.
1 d4 Nf6 2 Nf3 e6 3 Bf4
I am preparing for OTB play and so I force myself to play the opening from memory. I have studied enough to deal with 1…d5 (QGD, QGA, Slav, Semi-Slav), but I’m not yet booked up enough to deal with the Gruenfeld, King’s Indian or Queen’s Indian. 3 Bf4 is a London System approach to avoiding the Queen’s Indian. If I am not mistaken, Gata Kamsky used this system recently and scored a surprising victory.
My opponent transposes the game back to the London System proper (1 d4 d5 2 Nf3 Nf6 3 Bf4 e6). Strong players generally find it easy to equalize as Black against the London.
4 e3 Bd6
One of the more reliable approaches for Black is to force White to trade off the f4 bishop. However, Game Explorer master game percentages for this line are only 42-35-23 (White Win Pct-Draw Pct-Black Win Pct).
In Ideas Behind Modern Chess Openings, Gary Lane suggests 5 Bg3 to force black to open up the h-file in order to exchange for the bishop. Lane offers 5 Ne5 as alternative path; I’ve tried it twice and gotten a bad game both times. Game Explorer percentages for this line are 51-29-20, but I’m telling you that it’s not that good .
5…0-0 6 Bd3
Standard placement of the light squared bishop in the London System (and the Colle).
6…c5 7 c3
Since White has not played c4, he can use the c-pawn to maintain a stronghold on d4.
Striking at the weakness on b2 created by the absence of the Bf4.
Protects the b2 pawn, but the white queen is more productive on e2 or f3-h3. If 8 b3, then 8…Qc7 9 Nd2 cxd4. White would prefer to recapture with 10 exd4 (to allow the Ne5 to eventually be supported by a Re1), but has to play 10 cxd4 since the c-pawn is hanging.
Have forced White to put his queen on a suboptimal square, Black’s queen returns to c7 to put pressure on White’s stronghold on e5.
9 Bg3 c4
This is not a good approach for Black if White has played Qe2. White can drop his bishop back to c2 and follow up with e4. Here, with the White queen on c2, it forces the bishop off the b1-h7 diagonal and White can no longer play e4.
10 Be2 Nbd7 11 f4
White protects e5 in a way that discourages Black from exchanging on that square. An exchange would open up the f-file for White and force Black to move an important kingside defender.
But I’m really starting to hate my position here. The position is closed. Black has more space and chances to advance on the queenside than White does on the kingside.
11…b5 12 Nd2 a5 13 0-0 b4
Black makes advanced contact with White’s position on the queenside.
With a pawn on f4, time to give this bishop a little more scope.
14…Bb7 15 Rab1 Rfb8
In hindsight, Black has got nearly all of his forces on the queenside. When White finally strikes back, the black rooks will sit helplessly on b8 and a8, unable to defend the black king. But at this stage of the game, Rfb8 made plenty of sense to me. Both Black’s b-pawn and a-pawn are advance-able and need support.
Rook lift. The rook can put pressure on Black’s king by moving to g3 or h3. And the Ra1 can now be swung over to f1 at some point.
White’s position is critical. (I am gradually adding analysis from my notes...)