Axel Smith - Stephan Giemsa (annotated by Axel Smith)
10.04.2011
When the team risks to relegate you have to play!
1.c4 Giemsa always plays the same openings: Grunfeld vs d4 and French vs e4. He seems to know this systems quite well, so it was clear that I should chose some Anti-system. For Grunfeld players, 1.c4 is most difficult to meet, and I noticed that he didn't score so well there.
1...Nf6 2.Nc3 d5 He had played this once in the normal Megabase, but also in a recent game that Stefan sent me.
3.cxd5 Nxd5 4.g3 In Grunfeld black allows white to build up a big pawn center, and then starts to undermine it. I have not occupied the center with pawns, so black's strategy will hardly work.
4...g6 5.Bg2 Nxc3 5...Nb6 is a solid alternative.
6.bxc3 Bg7 7.Rb1! White benefits from the fact that Nf3 isn't played. Now the threat isn't 8:Bxb7? Bxb7 9.Rxb7 Qd5! but 8.Rxb7!
 |  |  |  |  |  | | |
 | | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| |  | | | | | |
| | | | | |  | |
|  | |  |  | |  | |
7...Nc6 The most aggressive move. 7...Nd7 also defends b7 due to 8.Bxb7 Bxb7 9.Rxb7 Nb6, but is slighty passive.
8.h4 Planning h5-h6.
8...h6 If 8...h5, white gets a nice square on g5 for the knight.
9.Qa4 This was a novelty in Wijk Aan Zee 2010, when Nils Grandelius played it. I was there as is second...
9...O-O 10.h5! g5 11.f4 I have to admit that white's frenetic attack doesn't look so good. Where will the king hide? Well, the idea is that the king will stay in center behind the two pawns.
11...e5 The most natural move, gxf4 and g4 beeing alternative.
12.fxg5 12.Ba3 was actually what Nils played in the first game. A few moves later the strategy backfired when black managed to open the center. After winning by checkmate black even got a beauty price for the game.
12...Qxg5
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
13.g4!! Now it looks very crazy! This was my planned improvement after Nils loss in Wijk. We told Magnus Carlsen about the move, and he didn't really believe in white's position, but at the same time he couldn't refute it. The problem for black is that he has almost no moves! The bishop on c8 can't move because b7 is hangning, Rb8 loses a piece and a6 + b5 can be met by Rxb5. White's plan is either Nh3-f2, to get even more control, or d3 + Be3 + Nf3 + g5.
13...Nd4? 13...Rd8 was played by the World's Junior Champion, Eltaj Safarli, against Nils in the Olympiad in Sibiria last autumn. He didn't find any plan at all and a few moves later he made the desperate piece sacrifice Nd4. Nils won a game that attracted a lot of attention. I had already annotated the move 13.g4!! in four various sources, and will gladly write about it a few times more...
14.cxd4 Why not?
14...Qxg4 15.Bf3 Qg3+ I thought that both Kf1 and Kd1 were winning. My argument for playing Kf1 was that Qf2 was stopped. If I later manage to attack the queen, she has to retreat and I am able to develop with Nh3.
16.Kf1 exd4 17.Rb3
I also thought 17.Ba3 was winning, but there is also some risk of being too gready.; 17.Bxb7 is possible as well. a) 17...Qf4+ 18.Bf3 with a winning position. b) 17...Rb8 18.Rb3! and white removes the pin with tempi. c) 17...Bxb7 18.Rxb7 was why I avoided 17.Bxb7. Black can try to win the b-file and combine ideas with d3 and Rb1. Giving it more thought, I don't see why I should be afraid of this. I have Rh3 with tempo and can meet a later Rb8 with Rb3, keeping the file closed.
17...Be6! The best try.
18.Rxb7!? 18.Bd5 exchanges the bishops and keeps the rook on the third rank, but at the same time it improves black's pawn structure with fxe6.
18...Bf6 Planning Bh4 and Qf2#, but it can easily be met with Rxh4. [18...d3 19.e3! and white is completely solid.]
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
19.d3! going for the h6-pawn. If black defends with Kg7, he can't play Bh4 later due to Qxd4+.
19...Bh4 20.Rxh4 Qxh4 21.Bxh6 Rfe8 Here I wanted to play Qa5 or Qb5, to control f5 and to be able to exchange queens with Qg5+ if it's necessary. I saw that Qa5 allowed Rab8, but then I saw that the check on b1 isn't dangerous. 21...Bh3+ 22.Nxh3 Qxh3+ 23.Bg2 Qf5+ 24.Kg1 and there are no more checks.
22.Qa5 Rab8 22...Bh3+ 23.Nxh3 Qxh3+ 24.Bg2 and there is no check on f5.
23.Rxb8! 23.Qe5 Bh3+ 24.Nxh3 Rxe5 should have been embarassing.
23...Rxb8
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
24.Qe5 Rb1+ 24...Bh3+ 25.Nxh3 Qxh3+ 26.Kf2 Qh4+ 27.Kg2 is also just a few checks.
25.Kg2 1–0 The fifth game during the last two months that I won in exactly 25 moves. Black resigned due to 25...Rxg1+ 26.Kxg1 Qe1+ 27.Kg2 Bh3+ 28.Kxh3 Qf1+ 29.Bg2 when he is mated in three moves: 29...f6 30.Qe8+ Kh7 31.Qf7+ Kxh6 32.Qg6#
Comment by RiV: Wonderful game, Axel. Congratulations...