I did well, scoring 3.5 points out of six games, for a performance rating above 2100. In the last round I was paired with a young opponent. We were both going for a fight, since the winner would receive a price for the best result in our Elo interval. A sharp position ensued, on which I had the upper hand.
Here is the position which arose shortly after a Keres defence, which surprised white who played it inaccurately. I am black, and black is to move.
You can easily see that this position is already strategically won for black: white has a king stuck in the center, is looking forward to a lag in development, and is about to lose material. However, accuracy is required. The immediate 9....Nxa1 can be followed by 10.exf5 Bb4 11.Qxa7 and white is likely to eventually recapture the black knight in a1, with a worse position but no immediate loss in sight. However, refusing to snatch the a1 rook seems unnatural...
After some 10' of deep thought I opted for 9....Nb4! With this apparently unpretentious move black aims at keeping the status quo of threats to the white king, solving at the same time the weakness of white squares on the queenside and averting the threats of checks to the black king by the white lightsquared bishop.
After 9.... Nb4! the game continued 10.Bb5+ Bd7 11.Bxd7 (here 11.Qa4? would have lost immediately to 11....Rxb5! followed by a pin of the knight) Qxd7 12.Qxa7?! (better would have been 12.Qa4, which could have been followed by 12....Qxa4 13.Nxa4 dxe4 with a winning game for black) 12... dxe4+ 13.Bd2 and black won the game quickly.
After the game, I gave my faithful Fritz 8 a chance to analyze deeply the position of the diagram. And I was very happy to see that even the materialistic silicon brain prefers by far the move I chose.