keypusher: 1.d4 Nf6?
My earliest-ever blunder. Having dipped into my modest Game Collection: An Opening Repertoire for Black -- Marovic/Parma, I had planned to play 1….d5 so that I could meet 2.c4 e6 3.Nc3 with 3….Be7, sidestepping the Carlsbad/QGD Exchange Variation. But seated at the board, I plonked out the standard 1...Nf6 without thinking and wound up in a Carlsbad after all.
2.c4 e6 3.Nc3 d5 4.cxd5 exd5 5.Bg5 Be7 6.e3 0-0 7.Bd3 Nbd7 8.Nge2
Marovic and Parma don’t cover this line, though it goes back at least to Botvinnik vs Keres, 1952. I knew generally that I was supposed to 1. trade bishops and 2. not get horribly squashed in the center. Unfortunately I only managed to carry out the first half of this plan.
8….c6 9.Qc2 Re8 10.0-0 Nf8
During the post-mortem, a rather irritating kibitzer expressed doubt about this move, but it’s book. Sadly, the knight on f8 winds up being probably my best-placed piece.
11.f3 Nh5
The kibitzer scorned this move also -- it’s a comfort that I beat him in a later game. Alex King explained in the post-mortem that the normal continuation is 11….Be6, and if White lurches forward immediately with 12.e4?!, then ...dxe4 13….fxe4 Ng4! is strong. After the exchange of bishops Black threatens a fork at e3, and if 15.Qd2 Black flicks in ...c5 to try to gain e5 for his knight. In any event, as King said White can make all sorts of strengthening moves before pushing the e-pawn forward, and it’s challenging for Black to generate counterplay in the meantime.
12.Bxe7 Rxe7
With the vague idea of keeping d4 under attack.
13.e4 Nf6?
After 13….dxe4 14.fxe4 Ne6! 15.d5 Nc5 followed by ...Nf6 (SF), Black is OK! This is pretty dynamic stuff to be playing against someone rated over 500 points higher, but there’s no other way to handle the line for Black. After the text move the SF eval goes immediately to about +1.9.
14.e5 N6d7?
And now it goes higher still. ...Ne8 was a little better.
15.f4 f6
This is SF’s top choice (and the other lines all feature ...f6 in a move or two), which shows how dreadful Black’s position is. The eval is well over +3 at this point. After this King just continued to make solid, strengthening moves until I collapsed. He didn’t always select the engine’s favorite, but that was hardly necessary.
16.Rf3 fxe5?
Excessively cooperative. Note that after fxe5 and doubling rooks on the f-file, the knight on d7 will be soldered in place because of the threat of Rxf8+, and if I move my rook back to e8, the f7 square will be free for the taking.
17.fxe5 b6
Another admission of futility.
18.Raf1 Bb7 19.Nf4 Qe8
As my opponent showed me afterwards, if I’d played 19….Nb8, White has 20.Bxh7+ Nxh7 21.Qxh7+ Kxh7 22.Rh3+ Kg8 23.Ng6 followed by mate. 19….Nb8? Well, given some of the other moves I played, I suppose he had grounds for hope.
20.Rh3 g6 21.Rg3 Rg6 22.h4 Ne6 23.Nxe6 Qxe6 24.Ne4!
As we discussed afterwards, this is the sort of thing that happens to you when you have a really horrible position.
24….Rf8 25.Rxf8+ Kxf8 26.Ng5!
King’s main challenge was that he had several promising possibilities on every move, but this is the best square for the knight.
26….Qe7 27.e6 Nf6 28.Rf3 Ke8
28….Kg8 is not better.
29.Qf2 Ng4
I was going to play ...Ne4 and try eke out a few more moves of misery, but in fact 30.Bxe4 dxe4 31.Rf7 Rxf7 32.exf7+ wins immediately. After the text move, the world ends not with a whimper, but a bang. The funny thing is, in the finale it turns out that I again can’t cover f7 and f8 at the same time.