Grandmaster Teimour Radjabov won two King's Indian games in Rounds 1 and 3 of the Wijk aan Zee 2007 super tournament. These games were annotated in Chess Today, the first Internet-based daily Chess newspaper.

Van Wely,L (2683) - Radjabov,T (2729) [E97]

Corus A Wijk aan Zee NED (1), 13.01.2007

Notes by GM Mikhail Golubev (Chess Today - 2259, 14th January 2007)

Six days ago, in Odessa, I said to Radjabov that I will be looking for his next King's Indian games. Thanks Teimour. I really did not have to wait long.

1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.Nf3 0-0 6.Be2 e5 7.0-0 Nc6 8.d5 Ne7 9.b4 Nh5 10.Re1 f5 On the same day in Corus-B tournament another famous King's Indian expert, Viorel Bologan, opted for 10...a5 11.bxa5 Rxa5 12.Nd2 Nf4 13.Bf1 Ra8!? (Bologan's speciality) 14.a4 Nh5N (Werle-Bologan) and eventually won the game. 11.Ng5 Nf6 One of the key positions in the modern theory of the King's Indian. In the last 11 years Van Wely has played it in at least 7 games with White, while Radjabov played it at least 13 times with Black (and also twice with White). 12.f3 A principal, safer alternative is Kramnik's trademark 12.Bf3. 12...Kh8 13.Ne6!?

The only previous game where this move occurred was... Van Wely-Radjabov from the 2005 FIDE World Cup. Later Van Wely's novelty 13.Ne6 won the contest "The Most Important Novelty of Chess Informant 96" (see www.chesscafe.com/text/informant51.pdf). The game had huge publicity, and it was annotated, for example, by Grandmaster Kavalek in the Washington Post. In the NiC Magazine 1/2006 Van Wely explained his 13th move in the following way: "After basically trying all White's moves in this position, I came to the conclusion that White hasn't any useful waiting moves until Black plays ...h6 and forces the knight to go to e6." 13...Bxe6 14.dxe6 Nh5 Radjabov follows my suggestion from CT-1853. Well we at CT are not always right but sometimes we have sensible ideas. Alexander Grischuk told me that at the Tal Memorial blitz 2006 Magnus Carlsen defeated him, by using IM Maxim Notkin's suggestion in CT from the notes to one of Grischuk's recent games. In Van Wely-Radjabov, Khanty Mansyisk 2005 White was better after 14...fxe4 15.fxe4 Nc6 (curious but also insufficient for equality is 15...Qc8 16.Nd5 c6, trapping the knight after 17.Nxe7 Qxe6) 16.Nd5! and so on. GM Viktor Mikhalevski at ChessPublishing.com observed that 14...Qc8?! is senseless in view of 15.Nd5. 14...Nc6 also looks promising for White: 15.Nd5! (GM Shipov at ChessPro.ru). In the book "Opening for White according to Kramnik Vol. 1B " alternatives to 14...fxe4 are not examined at all. The editorial team probably did not have enough time to study this very fresh line. 15.g3 Suddenly transposing to an encounter from last year's All China Games. If 15.Bg5, then Black can play 15...Bf6 (CT-1853). 15...Bf6N The second novelty in the game. 15...Nc6 was Ju Wenjun-Wu Xibin, Suzhou 2006. 16.c5 f4 17.g4? This advance seems to be a serious mistake. After preserving his king on the g1-a7 diagonal White will have no time to operate successfully in the centre, because in many lines his king would come under check. 17.Kg2! was much better, with a very complex position. After looking at it for a while, I could not come to a definite conclusion. Possibly, the critical line is 17...Nc6!? (17...a5 18.cxd6 cxd6 19.Nb5 d5 20.exd5 fxg3 looks too risky for Black; after the apparently premature exchange 17...fxg3 18.hxg3 the attempt 18...Nxg3? 19.Kxg3 Bh4+ is refuted not by 20.Kxh4?? Nf5+ where Black wins, but simply by 20.Kg2 Bxe1 21.Qxe1 +/-; finally 17...Ng7 18.Bc4!? looks promising for White at first glance) 18.cxd6 cxd6 19.Nd5, with mutual chances. 17...Ng7 Now the game has transposed to an encounter from the Dubai open 2002! So we will see a third novelty as well. Nothing sensational - in transpositional lines such things occur not too seldom. 18.Bc4 The attempt to attack the d6 pawn by 18.cxd6 cxd6 19.Nb5 fails because Black has the ...Qb6+ resource. 18.Nb5 also favours Black 18...a6!? 19.cxd6 axb5 20.dxe7 Qxe7. 18...Nc6 19.cxd6 cxd6 20.Ne2N Not worse, but perhaps not really better than 20.Nb5 Nxe6!-/+ (Bakre-A.Kuzmin, Dubai open 2002) with the idea of 21.Bxe6 Qb6+. 20...Rc8 -/+. Radjabov has gained an advantage, and played quite convincingly in the remaining part of the game. A dynamic initiative is certainly Teimour's element. 21.Bd5 Nxb4 22.Rb1 Nc2 23.Rf1 b6 24.Rb2 Ne3 25.Bxe3 fxe3 26.Qb3 Bg5 27.Nc3 Rc5 28.Na4 Rc7 29.Nc3 Qc8! 30.Nb5 Rc1! 31.Rb1 31.Nxd6? loses a piece after 31...e2! 32.Rxe2 Qc5+. 31...Rxf1+ 32.Rxf1 Qc5! 33.Kg2 Rc8 34.Re1 The most direct refutation of 34.Nxa7? is 34...Qa5! 35.Nxc8 Qd2+ 36.Kh3 Bf4 37.Rh1 Qf2, winning. 34...a6! 35.Na3 Qd4 Probably, 35...b5! was more precise. 36.Re2? Loek misses a chance to organise some resistance by 36.Rd1! and 36...e2!? (not forced, but probably Teimour planned this, making his previous move) does not win because of 37.Rxd4 e1Q (37...exd4?? 38.Kf2 +-) 38.Qd1! Qh4 39.Rc4! and White is in the game. 36...Rc3 -+ 37.Qb2 h5 The immediate 37...Qd1! was simpler. 38.h3 Here White could have prolonged the game by 38.Nb1, but Black is definitely winning anyway. 38...Qd1! 39.Bb3 Rxb3! 40.axb3 Nxe6 And Loek resigned. Well sometimes, The King's Indian dream comes true. The game is an interesting example of consistency in the opening. Very often in modern chess players are afraid to repeat their lines, with or without direct reasons. But as we see the Kamikazes in the chess elite still exist. 0-1

Shirov,A (2715) - Radjabov,T (2729) [E97]

Corus A Wijk aan Zee NED (3), 15.01.2007

Notes by IM Maxim Notkin (Chess Today - 2261, 16th January 2007)

1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.Be2 0-0 6.Nf3 e5 7.0-0 Nc6 8.d5 Ne7 9.b4 Nh5 10.Re1 f5 11.Ng5 Nf6 12.f3 Kh8 13.Ne6 Bxe6 14.dxe6 Nh5 15.g3 Bf6 16.c5 f4 17.Kg2 Let's start here as the previous moves were explained by Mikhail Golubev in CT-2259 in his annotations to the game of Round 1 Van Wely-Radjabov which saw 17.g4? Ng7 18.Bc4 Nc6 19.cxd6 cxd6 20.Ne2 Rc8 with Black's advantage. Shirov follows the line suggested by Mikhail. 17...Nc6 18.cxd6 cxd6 19.Nd5 Here Mikhail stopped evaluating the position as double-edged. Actually for a commentator there was no point to go deeply into variations as the play is not forced and the possibilities are many. But the players surely studied the position more thoroughly. 19...Nd4 To my mind it makes sense to recapture on e6 with the other knight - 19...Ng7 Perhaps Radjabov disliked 20.e7 but after 20...Bxe7 21.gxf4 Ne6 Black seems to be fine. 20.Bb2 Nxe6 Black spent three moves to gain material advantage and this is not surprising that White managed to activate his pieces. 21.g4 Nhg7 22.Nxf6 22.Bc4 deserved attention with a possible continuation 22...Bh4 23.Re2 h5 {with counterplay} Shirov prefers a more direct way. 22...Rxf6 23.Qd5 Qe7 24.Red1 Rd8 Black has protected his weakness. White needs a square for one of his pieces that are able to exert more pressure on the d6 pawn. 25.Qa5!? At the cost of two tempi Shirov provokes the advance of the b-pawn which provides his rook with the c6 square. 25.Rac1 looks a bit abstract, though, of course it is a possible continuation.; 25.b5 is less convincing as the dark-squared bishop's scope will be restricted by the knight in the line 25...Rff8 26.Ba3 Nc5 followed by Nge6 (in case of 26...Nd4 Black should reckon with the exchange sacrifice). 25...b6 26.Qd5 Rff8 26...Rf7 is a reasonable alternative in order to have an opportunity to protect the d-pawn by Rfd7 after the black queen is sent to the kingside. 27.Rac1 h5 Radjabov chooses a vigorous and spectacular way of launching the kingside attack rejecting the slow 27...Qg5. 28.gxh5 Qh4! In the event of 28...Qg5+ 29.Kh1 Qxh5 30.Rg1 White would be better e.g. 30...Ng5 31.Rc6 and the f3 pawn is untouchable. 29.Rc6?! Bad is 29.hxg6? Rf6. But to switch to the g-file wasn't a bad idea - 29.Rg1! Ng5 30.Kh1 gxh5 31.Rg2 with complex play (premature is 31.Rc6? Nh3). 29...g5! 29...Ng5? runs into the sudden advance of White's h-pawns - 30.h6! Qxh6 (if 30...Qh3+ 31.Kg1 Qxh6 32.Rxd6 and Black collapses) 31.h4! Nf7 {the only move} (31...Qxh4 32.Rh1) 32.Rc7 Kg8 33.Rxf7! Rxf7 34.Bc4 Rdd7 (34...Rdf8 35.Qxd6 Qxh4 36.Qxg6 +-) 35.Qa8+ Kh7 36.Bxf7 Rxf7 37.Qd8 with a winning position for White. 30.Rxd6? Shirov underestimated the power of his opponent's threats. Correct was 30.Kh1 when the position remained unclear for example 30...Rf6 (30...g4 31.Rg1 gxf3 32.Bxf3 Qf2 33.Qb3) 31.Rxd6 Rxd6 32.Qxd6 Rh6 33.Bc4 (or 33.Rd2). 30...g4! Black's idea is very simple - Qh3+ and g4-g3 and it transpires that White has no sufficient defence. 31.Rxe6 If 31.fxg4? f3+ 32.Bxf3 Nf4+ winning the queen; 31.Bxe5? Qh3+ 32.Kg1 g3! 33.Bxg7+ Kxg7 and White has only several harmless checks. In case of 31.Qxe5 Qh3+ 32.Kg1 Black should exchange the rook first by 32...Rxd6! and then g4-g3 decides. While the immediate 32...g3 leads to a fabulous draw after 33.Qxg7+! Nxg7 34.Rh6+ Kg8 35.Bc4+ Ne6 36.Rh8+ Kf7 37.Rh7+ Kg8 38.Rh8+ =, but not 38.Rg7+? Kh8 when the white pieces' harmony is broken and he loses even though looking at the position with the naked eye it's hard to believe that White has no perpetual anymore - 39.Rgd7+ (39.Rxg3+ Kh7; 39.Rf7+ Nd4!) 39...Ng7!. 31...Rxd5 32.Rh6+ Kg8 33.Bc4 gxf3+ 34.Kh1 Nxh5 The computer shows the stunning 34...Qh3! 35.Bxd5+ Rf7 36.Rg6 f2 37.Rg2 Qd3!!. But the tactical operation realized by Radjabov is also neat and well-calculated. 35.Rg1+ Ng3+! 36.Rxg3+ fxg3 37.Rxh4 After 37.Bxd5+ Kg7 38.Rxh4 g2+ 39.Kg1 f2+ White loses control of the f1 square. 37...g2+ 38.Kg1 f2+ 39.Kxg2 f1Q+! 40.Bxf1 Black's rook is no longer pinned and it allows him to obtain a decisive material advantage. The rest is easy. 40...Rd2+ 41.Kg3 Rxb2 42.Bc4+ Kg7 43.Bb3 Rb1 44.Kg2 Rc8 45.Kf3 Rc3+ 46.Kg4 Rf1 47.Kh5 Kf6 0-1
 

-- Chess Today (web site: www.chesstoday.net) is copyright 2000-2007 by GM Alexander Baburin. Reproduction of the material is prohibited without his written permission. My thanks to Alexander Baburin, and also to IM Maxim Notkin for permission to publish here his notes to the second game. --

Links
Chess Today web site: www.chesstoday.net
Corus Chess Tournament: www.coruschess.com
Odessa ACP Rapid World Cup: worldcup.pivdenny.com