In hindsight 23. Nxg6 is losing. I thought there was more to my chances in that position. I guess I was too confident.
A loss (of course) against Fritz
Sort:
Your opening was very timid. Refusing the push in the center. But I guess it's reasonable against Chess engines.
Prbobably 7.Qg4 is already an error.
Why? It's against the opening principles to get the queen out early and there's nothing to show an exception here:
There's no attack in site. The queen is a very bad attacker on it's own, when there aren't many weaknesses already in the opponent's camp. f5 won't help too much (doesn't threaten anything if black takes. So there's no effective way to make the bishop enter the attack with tempo... Even if there was, two pieces are usually not an attack at all).
13.g4 looks very weakening. But after it's played, WHY play 14.g5? it's inconsistent. Now there will never be an attack on the kingside (you've closed that side of the board, and black can do as he pleases on the other side of the board (and since you didn't castle kingside early, your king is in serious danger of getting mated).
In the next few moves, fritz lets go of several opportunities to execute white, which I guess is because it played this part in limited strength.
It even gives back the momentum to white (temporarily). On full strength, it would have executed you already.
Still,23.Nxg6? is just a blunder. You sacrifice a piece. That means - if you can't force your opponent into mate or get the material back, you're basically in a lost position.
The game is over at that point. you thought you HAD to break through? that's a bad way to think in Chess (not only in Chess, actually). Must be realistic. The position is roughly balanced and should be about equal if the game proceeds normally (Ng2 instead).
Nxg6 is taking an equal position and turning it into a lost one. A draw would have been a fine result.
There's no reason to expect an advanced pawn will win over a piece. usually it loses. If you don't see the exception as a forced line, it's a bad idea. In here, it was worse, since the h and g pawns were ALREADY blockaded and couldn't advance. so... no point in even looking at that.
In the meantime, black is two pieces up. even without the passed a pawn, he's winning easily.
Going back a bit, after 14.g5?, black should have Qa5+ and this attack is not a one piece attack. Things are crashing through quickly after:
15.Kc1 Qa3+ 16.Kd2 Nc6 (and after that to b4) and you've already got a queen and two knights swarming over your king, with most of your potential defenders locked on the other side of the board.
So, in conclusion:
1) don't bring the queen out early in the game without a concrete reason (Qg4).
2) when your only play is to attack on a certain side, try not to close it 100% (g5).
3) Don't win the game FOR your opponent (Nxg6). Patience. You can't always win. Trying when it's not possible tends to lead to more losses and to being a worse player.
4)a piece is a LOT. A single piece will usually in and of itself be sufficient to win the game even against a passed pawn, so before giving one, make sure you can FORCE an exception to the rule.
5) securely blockaded pawns are a lot less dangerous than they look.
Well, I think I played quite well. I'm not sure where exactly I went wrong. (Please note, the first half of the game was played with Fritz lowered in strength, but then he crashed and I had to resume in full strength).