farrooj: This is adams analysis
1.e4 c5 2.Nf3 e6 A bit of a surprise, as Joel one of many players who favour 2...Nc6 these days, but my opponent used this variation to make an easy draw against me in Tilburg 1998. 3.d4 cxd4 4.Nxd4 Nc6 5.Nc3 Qc7 6.g3 a6 7.Bg2 d6 8.0-0 Bd7 9.Re1 Be7 10.Nxc6 Bxc6 11.Qg4 h5 12.Qe2 h4 13.a4 hxg3 14.hxg3 Nf6 15.a5 Rc8 An intersting try here is 15...Rh5!?. 16.Be3 Kf8 Joel played this very quickly and was obviously still in his preparation. I believe he was probably influenced by the successful debut of this move in the hard-fought game Svidler-Movsesian, Polanica Zdroj 2000, which contiuned 17.Red1 e5 18.Rd2 draw. The drawback to the king move is that it will take Black some time to complete his development and coordinate his rooks. Good alternatives were 16...Nd7 when I was intending 17.b4 Bf6 18.Ra3 or 16...0-0. The previously mentioned game in Tilburg continued 16...e5!? 17.Bb6 Qb8 Well worth considering was 17...Qd7 18.Rad1 e5.
18.Na4! This seems stronger than 18.Bf3 which was played in Ponomariov-Ivanchuk, Moscow FIDE World Championship final 2002 (the position was reached by a different move order). 18...Bb5 The natural reaction to decentralizing the knight is 18...d5, but it doesn't seem to work here as after 19.exd5 both 19...Bxd5 20.Bxd5 Nxd5 21.Rad1 and 19...Nxd5 20.c4 Nb6 21.Nxb6 Rd8 22.Bxc6 bxc6 23.Qf3 look pleasant for White due to the big lead in development. 19.Qd2 Nd7 20.Be3 Bf6 Better was 20...g6 as after the natural 21.b3 Bf6 22.Bd4 Rxc2 23.Qxc2 Bxd4 24.Rad1 e5 looks like a promising exchange sacrifice. Instead White could play 21.c3 Bg7 but it is clear that Black has a much improved version of the game. 21.c3 Ne5 If instead 21...g6 22.Red1 is awkward to meet. 22.Nb6 Rd8 Perhaps 22...Nc4 was the lesser evil, although White has a pleasant edge. 23.Bd4 Qc7 24.b3 Bc6 25.Rad1 Qe7 26.c4 g5 This move is not without its risks but Black must do something or be gradually rolled off the board. 27.Qc3 Ng4? During the game I was considering 27...g4 28.c5 Rh5 (28...Kg7 29.Nd5! exd5 30.exd5 Rh5 31.dxc6 bxc6 is also good for White) 29.Nc4 Nf3 but in fact 30.Bxf3 Bxd4 31.Rxd4 gxf3 32.Rxd6 is easily good enough. Therefore the only good continuation is 27...Kg7! 28.c5 Bb5! (the key move giving the knight a retreat square) 29.f4 Nc6 30.Bxf6 Qxf6 31.Qxf6 Kxf6 32.cxd6 gxf4 33.gxf4 Na5 when Black is over the worst. Instead White can head for complications with 28.Nd5 but I think the strongest option is 28.f3 preventing the positional threat of g4. After 28...Ng6 White keeps the advantage. 28.Nd5! Also quite good was 28.c5 but the game move is more powerful. 28...Bxd5 29.exd5 e5 30.Bb6 Re8 31.c5 Kg7 32.Qb4 Nh2! Joel finds the only way to keep fighting White should be wary of a counterattack down the h-file. 33.c6 g4 If 33...Rh6 34.Rc1 is very strong. 34.Re4 Rh6
35.Rxg4+! An excellent practical decision. This move serves the handly dual purpose of making my king completely safe whilst exposing my opponent's. This problem combined with my advanced passed c-pawn render Black's position indefensible 35...Nxg4 36.Qxg4+ Rg6 37.Qe4 I didn't find the most accurate ways to finish off the game around this stage. More incisive here was 37.Qf5! e4 (37...Rg5 38.Qh3 Rh8 39.c7!) 38.c7 and the c-pawn can't be stopped. 37...Bg5 38.Qf5 bxc6 The capture is forced as 38...e4 39.c7 wins. 39.dxc6 e4 40.Re1 On the last move of the time control I played it safe but 40.c7! Qe6 (40...Re6 41.Bf1 e3 42.Ba6) 41.Be4 would have won immediately. 40...Re6 41.c7 e3 42.f4 Bf6 43.Bd5 Bc3 44.Re2 Bd2 45.Rh2 Bc3 46.Rh7+ Kg8 47.Rxf7! e2 48.Qh7 Mate,1-0.
sac: 35.Rxg4+