This is the game that I played against Dana Mackenzie during the labor day tournament, on which he has also commented in his blog (http://www.danamackenzie.com/blog/).
The opening is a symmetrical English/Botvinnik formation with similar qualities to a previously annotated game (against Rudy Vrana). This is a fascinating formation in which nothing really seems to happen on the surface, but there is a lot of long term tension in the game, with slow pawn storms and fights for square control. Indeed, matters often come to a head later in the game. As Karpov remarks in his book on the English opening, early calmness lead to middle game complications.
In a tense middle game (where black was obviously trying to do his kingside pawn storm) I sac the exchange to reroute the LSB to an outpost spot on the kingside, which serves the dual purpose of delaying (but not stopping, as I later found) the kingside attack and paralyzing an entire complex of light squares. I think my opponent could have found improvements in his play - there were things that I did not like for me. But he fails to react the way I expected and shuffles around, eventually to go for his kingside pawn storm again after considerable prep. I could have stopped that here, but I reckoned that I was going to set up a blockade with my own pawns. But this was faulty. I saw that this blockade could have been broken with a piece sac. But again, I missed his continuation which was more logical and simple. I think both these continuations gave him good advantage. Instead of setting up my leisurely queenside bind, I should have given some thought to his possibilities (which were no less leisurely). But then, I felt that my leisurely bind was necessary, as he could have made things difficult for me otherwise. All this goes to show that my exchange sac was a little to presumptuous.
In the game, I muddy the waters somewhat by sacrificing a piece to put his king in a mating net. The moves that followed were all nearly forced - my opponent gave back the piece to save his king, and then forced me to sac a second exchange. We had this funny material balance of two exchanges -2 P for him. He then defended and slowly took the initiative with his exchanges. In zeitnot, queens were traded and he gave back one of the exchanges. And then, in the time scramble, he blundered a pawn which again gave me winning chances, and offered a draw. Sadly, the time situation was so bad for me that I couldn't really do anything with my new found material (we were back on 'even' material count, although he had a very active rook, and I had very advanced pawns). But then again, these delicate positions require master level technique, and although I was quite optimistic of my chances, it was mostly just greed. Certainly, greed shouldn't play a role while playing difficult endgames, as even if either side were to win such a game, it would, by it's very nature, be faulty and tarnish the spirit of the struggle, which is to play good chess (at least in my opinion) and not turn it into a blunderfest, regardless of the result.
I don't know what happened at this point - his rook ate some pawns and a cheap shot of mine was foiled (which I thought was key), and it was game over.