In this game, I played Paul for the first time, a Class player who has often been rated 1900+ and has been rated as high as Expert in 2009.
I spent 8-10 minutes on the move where I drop a pawn, by then I was only deciding between Bb7 and a6, and seeing my blunder right after I move. Disturbing part of that is that I analyzed so much other stuff on that turn that I eventually needed to move to be sensible on the clock. The thing that I need to change is that I need to see this sort of threat right away (at G/90). Sure, I had almost gotten to it, but it’s too weird that I actually analyzed a lot of stuff before finally working my way back to an immediate threat.
I thought that I could attack my way out of it with his king still not castled, but Crafty seems to indicate that I am simply never getting back to even on pawns. Originally, my idea was to play Bb7 instead of Rd7, which was an over-finesse, and win the g-pawn, but even then Black has Qb3 and Qxb6 pawn.
Once he castled, I knew the jig was up. I drop a piece at the end and immediately resign, but I knew that I didn’t have time to play a losing ending either. 3 1/2 minutes to his 17. If I were playing a lower-rated player, I would have saved time for the endgame in any event, but here I was simply gambling on my attack. I had thought perhaps …Bb7 was a threat to trap his queen at first, but then saw that it was a mirage. Even though I had castled early, it turned out that I was the one who was always needing to factor in those back-rank mate threats, and after …g6 my attack would evaporate.
My last move blunder, ironically finally gave my king the luft it was looking for, but out of frustration. A typical ending, without even looking at it, could have gone rook and queen trade, then rook (while threatening both pawn chains) trades itself for bishop and queenside pawn, leading to won king and pawn ending. He knew the ending was lost as well.
I had seen the stuff that Crafty was looking at, that …dxc4 was probably Black’s best move, that h6 looked like an option as well, and had figured he would play cxd like he did. My point is, it seems like I enjoy seeing how higher-rated players will handle weird positions, instead of simply playing straight-forward best moves against them. I didn’t want my bishop on that b7 diagonal, even though I had predicted he would play Bb5, so if I had noticed that it was forced, I wouldn’t have let him trade pawns first like that.
One thing that might have thrown me off was that my opponent, who writes a column for the local club, had one recently where he said the perfect time-control was G/30. After playing him, I realized it must have been a sarcastic comment based on his results, but at the time all I knew was that he plays a lot of quicker time-control events, so I wasn’t sure if he was as strong as his rating indicated. Yep, he really is that strong.
Here is a continuation of how I should have played that attack. What is strange is that I had to pick out some of the moves for Crafty (i.e., they were my suggestions), namely Be4 (instead of Qxh4), Nd3+, g6 and even Qc3 at the end, which seems like an easy to find draw. But isn’t is just obvious to a human’s eye that if I had simply _developed_ my bishop to that diagonal as I should have, forget about attack just then, compare development. His king is in the center and all my pieces are developed and infiltrating, back rank is secure. There is a certain amount of “intuitively obvious” sort of reasoning that can be had there. Not necessary to do a nervous, sweaty material count on every move. Actually, in the game, he seemed nervous but I wasn’t.
In the post-mortem, after …Bxg2, Rg1, I proposed ..Be4 and he said Rc1 then, noting that if I play Nd3+ fork, that he would just take it with his queen, but really even that leads to an even position as his Rc8+…Qf8 has given me back the exchange, with even pawns. And if …Ra8 instead of …Nd3+, as Crafty notes, then the fork is back on and Black actually has the advantage. Crafty gives Ke2 followed by Qxh4. Then actually after Qxb6 and …Nd3 (my move, although Crafty finds another) Rcf1 g6 Black is able to take it down to an even ending or forced draw, based on White’s replies. So, it was indeed drawable after I had lost the pawn. I’m sure Fritz stands a good chance of finding better, but based on what was likely to happen on the board, that would have been sufficient enough.
I think what happened in that post-mortem is that I had postulated moving the Qf6, which allows for that QxNd3, then Rc8 mate. That is the problem when I do post-mortems, I put the ideas out of variations. There is no variation where the queen ends up on f6 in the Bb7 variation. I must have left it there over from analyzing the …Rd7, losing variation, that was actually played.
The way it actually would have gone after Rc1 …Ra8 is that he wanted to play Qc3 to threaten g8 (go figure). So I would have played ..g6 and had a slight advantage of 2 pawns for the exchange after a few more moves (-.80). So, he did still have a chance to work the clock, in any case. Time could have become a big factor, so I should have played Bb7 and moves like that quickly, especially as I had spotted them all along, was only spending all that extra time in case there was something “better”. Unfortunately, that is what lead to my …Rd7 also, too cute a move. Threatening his d4 knight was not a big deal as I had either a back rank issue, until Bb7, or it was simply an extra unnecessary temo, ..Rd7, which let him get out of the pickle. IOW, he has time for Qb3 and Qxb6 then.