1985-08-01 - Elliott, Sean 1-0 Ward, Chris
Sean Elliott does a bit of Dragon slaying and the victim is a future GM and British Champion. It's all book up to move 19, then Sean comes up with a move that seems to improve on known theory. He picks up two Rooks for Queen and two Pawns. The Rooks go active whilst the Queen goes passive. The key piece is Sean's own dark-squared Bishop which helps out in a tidy finish.
[Event "BRITISH U18 CHAMPS"] [Site "BRIGHTON"] [Date "1985.08.01"] [Round "?"] [White "Elliott, Sean"] [Black "Ward, Chris"] [Result "1-0"] [ECO "B76"] [Annotator "Mordue,Tyson"] [PlyCount "65"] {One of Sean's best games from his title-winning British U18 Championship campaign. His opponent, Chris Ward, went on to become a GM and British Champion in 1996 and is the author of several well-known chess books. Original notes by a combination of Sean Elliott and Graham Burgess (SEGB).} 1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 g6 {The Dragon variation of the Sicilian Defence. There's a joke that it's so-called because the theoretical arguments about this opening drag-on and on, but we get to move 20 in this game before we diverge from theory so it's a very apt joke!} 6. Be3 Bg7 7. f3 Nc6 8. Qd2 O-O 9. g4 Be6 10. O-O-O Nxd4 11. Bxd4 Qa5 12. a3 Rab8 13. h4 Rfc8 {"Although this is the 'book' move 13...b5 is possibly better, when White cannot play 14 h5 as this is a tempo down on the game. Instead, better [is] 14 Nd5 Qxd2+ 15 Rxd2 Bxd5 16 exd5 Rfc8 equal ... although beware 17 Bxa7 Ra8 18 Ba7 moves b4!" SEGB. 13...b5 has now succeeded 13...Rfc8 as the most popular move in this position by a large factor, 80 to 15 in the database. I recall seeing the relevant page in Chris Ward's book 'Winning with the Sicilian Dragon' and he regarded 13...b5 as an improvement. He also gave the 14 Nd5 line remarking that it was probably equal, but added that if White was too quick to swap dark-squared Bishops then he might end up with a bad Bishop versus good Knight ending. Chris also quoted the line with 17 Bxa7 yet the only game on the database featuring this ended in a win for White! The game was Thomas Eccleston v Gavyn Cooper, Millfield Young Masters 2004, which continued 17... Ra8 18 Bd4 b4 19 Rd3 bxa3 20 bxa3 Ra4?! (20...Ra5 seems to be more accurate) 21 Bb2 Rc5 22 Rb3 Nxd5 23 Rb8+ Bf8 24 c4 Nf4 25 Kc2 and White had a small plus - 1-0 in 54 moves.} 14. h5 b5 15. h6 {Probably the height of fashion in 1985 but nowadays 15 hxg6 is more common and probably more natural too. Why voluntarily block the wing on which you are attacking?} b4 {Apparently the natural 15...Bh8 is well met by 16 Nd5 and White has the advantage in the ending because of his advanced h-Pawn. The stats are clearly in White's favour so the sharp text is favoured.} 16. hxg7 bxa3 $1 {Poor is the natural 16... bxc3 17 Bxc3 Rxc3 18 Qxc3 Qxc3 19 bxc3 Kxg7 20 g5 with a significant plus of 1. 37 to White. No surprise as Black has lost his Queen, dark-squared Bishop and the Exchange. The text is more dangerous.} 17. Nb5 {"Tal gives here 17 Qh6 axb2+ 18 Kd2 'winning' but Mestel's 18...Bxg4!! wins for Black! 17 Nb5 is Nunn's suggestion in his book 'Beating The Sicilian'." SEGB. Mestel's suggested sac of 18...Bxg4 won for Black in Plaskett- William Watson, Brighton 1983 which went 19.Bxf6 Bh5 20.Bd4 e5 21.Rxh5 gxh5 22.Qg5 Qb4 23.Bd3 Qxd4 24. Nd5 Qf2+ 25.Be2 Rxc2+ 26.Kxc2 Qxe2+ 27.Kc3 Qxf3+ 28.Kc4 Qb3 mate. A later game Germer-Mergard, Waldshut 2002 went 19 Bxf6 Bh5 20 Rxh5 gxh5 21 Bh3 exf6 22 Bxc8 Rxc8 23.Qxf6 Qb4 24.Rb1 a5 25.Kd3 Qc4+ 26.Kd2 Qb4 27.Kd3 Qc4+ 28.Kd2 Qb4 1/2-1/2 which followed Deep Rybka's analysis to the letter! Nunn's note runs "17 Nxb5 Rxb5 18 Bxb5 Qxb5 19 bxa3 [as per this game - TM] is the natural continuation. White's Bishop is a useful defensive piece ... but even so it is impossible to give a definite verdict on the line." The database has a clear verdict - 4/4 for Black in this line! The first edition of 'Beating The Sicilian' was published in 1984 so the players of this game were right up to date with the theory. It should be said that Sean proves his Bishop to be a useful attacking piece as well as a defensive one!} Rxb5 18. Bxb5 Qxb5 19. bxa3 Qa4 {"19...Bb3 looks stronger." SEGB. Poulsen- P Jakobsen, Copenhagen Obro Open 1989 continued 19...Bb3 20 Bc3 Ba2!? 21 Qe3 - 21 Qd4 looks better - with a small advantage to White. Poulsen pressed too hard in a level ending and lost in 64 moves. After 20 Bc3 (not 20 c3? Bxd1 21 Rxd1 e5 and Black is winning.) 20...Ba4 Deep Rybka says that it's about level. The text is soon going to win the White Queen in a very complicated position so that move or 19. ..Bb3 is a matter of taste. Three other players chose the text move as well.} 20. Bb2 {"An interesting idea here is 20 Qh2." SEGB. This is our departure point from theory. Huenerkopf-Hodgson, Copenhagen 1985 - they do like this line in Copenhagen! - featured 20 Qd3 and Black won an entertaining game in 46 moves. Two other games went 20 Rh2 Qxa3+ 21 Bb2 Qxf3 and Black was never troubled in either one. After 20 Qh2 Deep Rybka looked at two lines: A) 20. ..Qxa3+ 21.Bb2 Qe3+ 22.Rd2 h5 23.Qf2 Qxf2 24.Rxf2 hxg4 25.Bxf6 exf6 26.Rh8+ Kxg7 27.Rxc8 Bxc8 28.fxg4 Bxg4 29.Rd2 g5 30.Rxd6 with possible winning chances for White, or: B) 20...Bb3 21.Bxf6 Rxc2+ 22.Qxc2 Bxc2 23.Rxh7 exf6 24.Rdh1 f5 25.g5 Qxa3+ 26.Kxc2 Qa2+ with a perpetual check. The conclusion is that 20 Qh2 forces Black to clarify the position immediately although some of his moves in A are not forced. Personally I think 20 Bb2 is more interesting than 20 Qh2.} Bb3 21. Rh2 {Deep Rybka says 21 Bc3 is a better move. It analyses 21.. .Qa3+ 22 Bb2 Qa2 23 Bc3 which of course gives Black the chance to immediately force a draw, if he wants it, with 23...Qa3+. The text is simply more combative and I suspect both players were going for a win. Nothing like a good unbalanced position to try that in and that's what we get!} Bxc2 22. Qxc2 Rxc2+ 23. Rxc2 g5 $6 {"This enables Black to play ...Kxg7 without allowing g4-g5. The Black Queen looks better than White's Rooks but White surprisingly has excellent practical chances." SEGB. Our long opening line and a short tactical resolution has brought us to this position. The material imbalance is two Rooks for a Queen and two Pawns but I regard White's Bishop as a much stronger piece than the Black Knight here so add that to the pot. You can have various evaluations here but much of it is down to how a player feels about his position. Remember that Chris Ward is very fond of having an active Bishop on the a1-h8 diagonal and here it is his opponent who has that! If White wanted to draw this position he could probably do so quite readily with 24 Bxf6 and 25 Rd3. I would have been reluctant to make this suggestion before Black's last move but with the g6 Pawn on g5 Black no longer has ...h7-h5 producing an outside passed Pawn. 23...g5 - the ?! is mine - has denied Black certain options and I think 23...h6 was a better move. Whatever, it is certainly in order for White to play actively.} 24. Rc7 Qb3 {By the same token expressed in the previous note Black does well to go active rather than play the passive 24...Qe8 suggested by Deep Rybka.} 25. Rxe7 Qxf3 26. e5 {This is given an ! by SEGB but I don't see why. Better is simply 26 Rxd6 because 26... Qf4+ 27 Rd2 Nxe4?? loses instantly to 28 Re8 mate, a motif that we'll see a lot of. Instead 27...Kxg7 looks suspicious after 28 Re5 threatening 29 Rf5 while after 28...Qe3 - which has the merit of defending a7 - 29 Bxf6 Qxa3+ Deep Rybka claims Black has a perpetual check, although 29 Bxf6 is not forced.} dxe5 27. Rxe5 Qxg4 $6 {This looks dubious and I prefer 27...h6.} 28. Ree1 { Given an ! without any explanation and I wonder if it's a typo! After 28.Rd8+ Kxg7 29.Red5! h5 30.R5d6 White is winning the Knight but Deep Rybka claims that Black has a perpetual check mainly by checking on c4, f1 and f4 as appropriate. It could well be right but unless Black is confident about this I don't think he should be playing 27...Qxg4. It might be that the point of the text is that Black has no moves available for his under-fire Knight because they allow back rank mates, but Black has both the defence he plays in the game and also the simple 28...h5 because if 29 Bxf6 Qf4+ recovers the piece and ensures the elimination of the g7 Pawn.} Qc8+ 29. Kb1 Ne8 {Deep Rybka points out that 29...g4 is playable because 30 Bxf6 Qf5+ is okay. The best reply is probably 31 Re7 and that's also possible after the text.} 30. Re5 h6 {"If 30...f6 31 Re7 is good for White. If 30...g4 31 Rde1 Nxg7 32 Rg5 wins or 31...Nf6 32 Rf1 Ne8 (if 32...Qd8? 33 Rxf6!) 33 Re7 wins." SEGB. After 30...f6 31 Re7 Nxg7 32 Rdd7 Nf5 33 Rxh7 is given as equal by Deep Rybka but I'd certainly rather be White! 33...g4?? loses instantly to 34 Bxf6 so the engine suggests 33...Qe8 or 33...Qf8. As the saying goes "To allow one Rook on to your second rank may be regarded as unfortunate, but allowing two is careless." That's the effect of 30...f6. Instead after 30...g4 31 Rde1 Nxg7?? 32 Rg5 does indeed win but Black does better to play 31...Nd6. However, after 31...Nf6 32 Rf1 Ne8 33 Re7 Black continues with 33...Qc6 answering 34 R either xf7 with 34...Qg6+, so it suggests 34 Ka1 Nxg7 35 R either xf7 Ne6 and says it's still equal. Intricate stuff. 30...h6 which seems the natural move is, compared to earlier, actually the right move!} 31. Rh1 Nxg7 $2 {"In desperate time-trouble Ward rejected 31...Qc6! because of 32.Rxh6!? [if 32... Qxh6?? 33 Rxe8+ Kh7 34 g8=Q mate or 32...Nf6? 33 Rxf6! Qxf6 33 Re8+ wins the Queen. TM] but 32...f6 appears to draw e.g. 33.Rxe8+ [33 Rh8+ Kxg7 34 Rexe8 is a transposition - TM] 33...Kxg7 34.Rhh8 Qd5 35.Reg8+ Qxg8 36.Rxg8 Kxg8 37.Bxf6 draws [because White has the wrong Bishop to go with his a-Pawn. TM]." SEGB. There's not much arguing with this line. Deep Rybka doesn't appreciate the wrong Bishop and Rook's Pawn scenario for a while so it's evaluation is skewed. On the other hand White doesn't have to play 35 R (either) g8+, but he does have to avoid perpetual checks and that's not easy. It appears that 31... Nxg7 is a mistake - after all it puts a Knight on Knight two - and that 31... Qc6 is a better move. However, the time-control is move 40 so Black was always going to have a job to get there safely now that Sean has maximised his Rook's activity.} 32. Rxh6 Qg4 $4 {"This is a blunder in a difficult position, especially in time-trouble. 32...Ne6 fails to 33.Rxg5+!" SEGB. Indeed White is doing well. Deep Rybka's suggestions are passive. First 32...Qd7 which does draw by perpetual check after 33.Rxg5? Qd3+, but if White plays 33 Ka1 first it's difficult. Secondly there's 32...Qd8 to which the engine suggests 33.Rc5 Qd7 34 Ka2 and Black is having difficulty finding moves. The main point is that any Queen move off the back rank that doesn't stop a check by the centralised Rook fails to the very tactic that ended the game.} 33. Re8+ $1 { After 33..Nxe8 34 Rh8 mate it's yet another a triumph for a fianchettoed dark-squared Bishop in a Sicilian Dragon. However, for a change it's a White one!} 1-0
The games viewer used here was developed by
http://chesstempo.com
and is licensed under
CC BY 2.0.
