1.Sf3 When I prepared I noticed that Möller wanted to play either Nimzo Indian, Bogo Indian or Hedgehog. Since he is not so experienced I think it's a good idea to avoid his main systems and instead bring him into somewhat deeper water... With 1.Nf3 I think I managed to do this.
1...Sf6 2.c4 c5 [2...e6 3.g3 was my idea. Without the knight on c3 Nimzo Indian isn't possible, and he can actually neither play the Hedgehog: a) 3...b6 4.d4! transposes to Queen's Indian, b) 3...c5 4.Bg2 and it's too late for black to play b6.]
3.Sc3 Here black can try to play Hedgehog in many ways, but he is always getting a worse version, for example 3...b6 4.e4! d6 5.d4 cxd4 6.Nxd4 and black has closed in his dark squared bishop too early. [3.g3 b6 would be a Hedgehog.]
3...e6 4.d4 cxd4 5.Sxd4 Now black can transpose to a slightly unusual version in the Nimzo Indian with 5...Bb4. The most normal move is 5...Nc6 with a symmetrical English.
5...a6 [To try to play Hedgehog with 5...b6? is not working: 6.Sdb5! d6?! 7.Lf4 e5 8.Lg5 White wins material, for example 8...a6 9.Lxf6 gxf6 10.Dd5 and the rook is lost.]
6.g3 Dc7 7.Lg2! [7.Dd3 Sc6 threatens Ne5 and forced white to take on c6.]
7...Lb4?! [Having said A... Black should take the pawn. After 7...Dxc4 8.0-0 white has a lot of development as compensation for the pawn. White scored a convincing win in Agrest-Carlsson 2008, but since Kramnik has played this with black I guess it's playable.]
8.0-0 White does not have to be afraid of getting the pawns doubled, since black will not be able to set up a good blockade with a knight on c5. [8.Db3 Lc5 and white can't threaten the bishop with Nb3.]
8...Lxc3 I think that black should have castled instead, but his play already lacks some consistency.
9.bxc3 0-0 [9...Dxc4 10.La3 would be nice compensation.]
10.La3 d6 [10...Te8?! 11.c5 will cement the holes on b5 and d6. The bishop on c8 and the rook on a8 will never be able to develop!]
11.Dd3?! I thought this was sligtly more exact than 11.Nb3, since it doesn't allow Nc6. Möller told me after the game that he was most afraid of 11.Qa4!, with the threat of Nb5. It seems to be the strongest option.
11...Td8 [A critical try is 11...Sbd7 12.Sb3 Sc5 13.De3 Sfd7 with a blockade on c5, but after 14.Tfd1 with the plan of Rd2 and Rad1, I couldn't see how black should defend the pawn on d6.]
12.Tfd1 d5? [Better was 12...Sbd7 with the idea to play d5 next move, without problems on the d-file.]
13.cxd5 exd5 [13...Sxd5 14.c4 Sf6 15.Sxe6! wins immediately. This theme will also appear in the game.]
14.c4! White opens up before black has fullfilled his development. Möller didn't want to take the c4-pawn in the opening, but now he is forced to.
14...Dxc4! The best try. [14...Le6 15.cxd5 Lxd5 16.e4 Lc4 17.Se6! also wins a rook.; 14...dxc4 15.Se6 cxd3 16.Sxc7 Ta7 17.Lc5 wins the rook.]
15.Dxc4 dxc4 16.Se6?! [Strongest was 16.Sf5 , a move I didn't even think about. The back rank problems will decide the game.]
16...Txd1+ 17.Txd1 Sbd7 The only move. [17...fxe6 18.Td8+ Kf7 19.Tf8+ Kg6 20.Txc8 will win at least a piece.]
18.Sc7 Tb8 19.Ld6 When I played 14.c4 I was that black couldn't avoid losing an exchange here in a meaningful way, but I was not completely sure wheter he could get some compensation.
19...g5 [The critical line in my calculations ran 19...b5 20.Sd5 Sxd5 21.Lxb8 Sc3 22.Td2 Kf8 but I concluded that black's queenside pawn wasn't that dangerous.]
20.Sd5! [Since black's last move doesn't threaten anything, I was considering to improve the position of my king with 20.Kf1 , but I didn't want to give him another chance to play b5.]
20...Ta8 [20...Sxd5 21.Txd5! Ta8 22.Txg5+ Kh8 and white has regained his pawn with a hundred times better pieces.]
21.Se7+?! My greedy plan in the game is to kind. It would have been better to keep the bind.
21...Kg7 22.Sxc8 Txc8 23.Lxb7 Te8 24.Lxa6 c3 [24...Txe2 25.Lxc4 is winning for sure, with the bishop pair and an outside passed pawn.; When I played Nd5 I was mostly afraid of 24...Sb6 , since black threatens both Rxe2 and Ra8xa2. I saw that I could defend against both threats with 25.Tb1 Sfd7! (or 25...Sfd5 26.Lc5 and black loses the c4-pawn.) 26.Tb2 ]
25.Lb4 [Not 25.Tc1?? Te6 and white loses a piece.]
25...Ta8?! Now white can simplify to a won rook ending, but it was lost anyway.
26.Lxc3 Txa6 27.Txd7 Kg6 [27...Txa2 28.Td6 loses the knight.]
28.Lxf6 Kxf6 29.Td2 The rest is very easy and it didn't take us many minutes to play the remaining 30 (!) moves. To resign has never won any points, so it can never be a bad idea to play on.
29...h5 30.Kg2 Ke5 31.e3 g4 32.h3 Exchanging white's only potential weakness.
32...f6 33.hxg4 hxg4 34.Td4 Txa2 35.Txg4 Kf5 36.Tg8 Ta4 37.Kf3 Ke6 38.Tb8 Ta5 39.g4 Tc5 40.Kf4 Tc4+ 41.e4 Ta4 42.Tb5 Ke7 43.g5 fxg5+ 44.Txg5 Kf6 45.Tb5 Ta2 46.Tb6+ Ke7 47.f3 Ta3 48.Kg4 Te3 49.Tb4 Kf6 50.f4 Ta3 51.e5+ Ke6 52.Tb6+ Ke7 53.f5 Te3 54.Tb7+ Ke8 55.e6 Te1 56.Kg5 Th1 57.Kf6 Th6+ 58.Ke5 Th1 [Black could have played for a last trick with 58...Th8 59.Tb8+ Ke7 60.Txh8?? and stalemate.]
59.Tb8+ Ke7 60.f6# 1-0