..laughs last, or at least sometimes anyway.
I played this game yesterday, up an exchange and simply moving fast, making obvious moves. Luckily, it exposed a hole in my thinking that I have been coming across a lot of lately, and that is following a capture-trail down to the last capture rather than stopping at the first capture.
Notice at the end of the game, I am capturing his knight first, but he is capturing my rook last.
Another example, SAME game. Move 29.Be4, why is this move wrong? Looks sensible, but, the trade is freeing Black’s queen which is currently completely out of play and blocked in by own bishop. Same principle as the last example, the correct move is 29.g4! (if he captures pawn first, I capture knight next). Move 31, again, same theme, I played the pedestrian 31.Re5. Correct is 31.g4 #Wi-nning! as Charlie Sheen might say. g4, doesn’t that drop the rook? Yes, but taking the rook drops the king with g5 mate!
In both instances capturing the first piece is not nearly as profitable as capturing the last piece. A lot of times this theme comes up in the endgame with pawn-races where the side which queens a move or two later still manages to win by force. These examples go against intuition because, after all, who doesn’t want to capture an en-prise knight, or who does want to leave their rook en-prise? So the principle is to follow the capture-trail down to the last capture, and not stop at the first capture. 😉
Incidentally, where I blundered with 23.g3, either c4 or Qf7 would have been plenty good enough to maintain the win.