The Dragon's Lair (part 2)
We saw in part 1 of the Dragon's Lair what can happen to the would-be DragonSlayer when he takes on the beast without an adequate weapon. In this second instalment we look at a game where the weapon of choice is not the latest razor-sharp novelty at the cutting-edge of theory, but rather a home-made 'blade' designed to wrong-foot the dragon.
-Play through game option is at the end of the article-
1. e4 c5 2. Nf3 d6 3. d4 Nf6 4. Nc3 cxd4 5. Nxd4 g6
Again we see the basic starting position of the Dragon proper.
6. Be3 Bg7 7. f3 Nc6 8. Qd2 O-O 9. Bc4
( 9. O-O-O This move will be looked at in a future episode: at the moment it seems to be the main area of theoretical research amongst those who would wish ill upon the Dragon Variation! )
9... Bd7 10. O-O-O Rc8 11. Bb3 Ne5 12. h4
In a curious case of co-incidence, just this week I had to play the same opponent as in the main game here. This time (almost 8 years on!) I had the white pieces, and knowing my opponent was also a Dragon lover, I had prepared this move (12.Kb1). I figured he might not be au fait with the latest theory, so it seemed like a logical choice. 12... Nc4 13. Bxc4 Rxc4 14. g4 b5 15. b3!
and now my opponent paused for thought. Just as I had been when facing 15.b3 for the first time, he was confused and surprised that this move even existed as a possibility! Some 20 minutes later (a big chunk of evening league time controls) came... 15... Rxc3?! without the disruption of the white pawns as in more normal 'Dragon' exchange sacrifices, this move really should be insufficient. 16. Qxc3 Qa8 17. g5 Rc8 18. Qd3 My initial intention had been 18. Qa5 to prevent any ...a5 ideas from Black, and indeed this would have been preferable. The text move is ok too, but some of the forthcoming calculations will be trickier now. 18... Ne8 19. Nxb5 White's main idea in this line is to grab material and then use the a, b, and c-pawns to provide cover and back-up. 19... a6 20. Nd4 Nc7 21. c4 a5 22. Nb5? One of the problems associated with being an aggressive player is the inability or unwillingness to temper the active approach when necessary. Here I had my first long think, considered many possibilities (almost all of them featuring long variations) and missed the relatively simple solution ( 22. Bc1! with the intention of playing 22... a4 23. Bb2 and asking black to do his worst. ) 22... Nxb5! 23. cxb5 Rc3 and suddenly white is in serious trouble. The Dragon always has teeth, even when its initial fire has been extinguished!
24. Qd5!? This seemed like a good attempt to muddy the waters. The alternatives certainly didn't look too appealing, e.g. ( 24. Qe2 a4 25. Bd4 ( 25. b4!? This might just be ok , but variations like 25... Rb3+ ( 25... Bxb5!? 26. Qxb5 Rxe3 27. Rd3 Rxf3 28. Rxf3 Qxe4+ 29. Qd3 Qe5 ) 26. axb3 axb3 27. Bd4 Bxb5 28. Qxb5?? Qa2+ 29. Kc1 Qc2# are very scary indeed!
When there are a lot of such nasty threats, the prospect of missing one crucial line can be stultifying to the thought process. ) 25... axb3 26. Bxc3 Bxc3 27. axb3 Qa1+ 28. Kc2 Qb2+ 29. Kd3 Bxb5+ is game over. With these variations looking like fun to play, and a nightmare to defend, I sought solace in a queen swap where I have tactical tricks of my own. )
24... Qxd5!? ( Avoiding the queen swap might be even better, as the intended reply 24... Qb8 25. Bd4?? Be6 sees the white queen rather embarrassingly lost in no-man's land! ) 25. Rxd5 ( 25. exd5?? Bf5+ ) 25... Rxe3 26. b6 Bc6 27. Rc1
This was the idea behind 24. Qd5; black can't grab the rook on d5 as the b-pawn would be unstoppable. 27... Rc3?? Looks natural enough, but this is a serious mistake which loses the game for black. Much better was 27... Bc3 when 28. Rxa5 is still possible, but 28... Rxf3 with the threat of taking on e4 with check looks terminal. 29. Rd5 e6 30. Rxc3 Rxc3 31. Rxd6 Bxe4+ 32. Kb2 28. Rxa5! and white is winning again! 28... f5 29. gxf6 Bxf6 30. Rxc3 Bxc3 31. Ra7 Bd4 32. Rc7 and here black resigned 1-0. One of the bishops must fall, after which the a- and b-pawns are too strong to withstand )
Meanwhile, back in the main game, we see again the Soltis Variation. Designed to slow up white's k-side intentions, 12...h5 heralds in one of the main theoretical battlefields of the entire variation.
13. Bg5 Rc5 14. Rdg1!?
White decides to avoid the main continuation of 14. Kb1, opting instead for a 'home-grown' recipe. His intentions are obvious; the rook takes up an aggressive post, hoping after g4 to take an active part in the attack, and black will have to find an answer, probably without the help of the latest theoretical recommendations. Indeed, having never seen this move before in this exact position, I had to find an accurate plan to counter the white strategy.
When facing a new move in the Dragon, this response is always the first one to turn to - 9 times out of 10 it will be at least a very good try! Indeed, in 500-odd games in the chess.com openings explorer which reached 13...Rc5, the reply 14. Rdg1 appears only once, and 14...b5! was the reply - black winning convincingly after white threw almost his entire army away in a fruitless attempt at exposing the black king.
15. g4 b4
Taking on g4 is a serious alternative, but there is a counter-attacking scheme available here which I recalled from some previous games (in slightly different positions.)
16. Nd5 Nxd5 17. Bxd5 Rxd5!?
This is the big idea - the standard Dragon exchange sac on c3 is swapped for a somewhat less frequent one on d5, designed to lessen any problems on the white diagonal and to give the highly-prized c4 square to the black knight. As you may have realised by now, material considerations in this opening are not the main priority. Active pieces, open lines and the initiative are!
18. exd5 Qa5 19. Kb1 Nc4 20. Qd3 Rc8 21. Nc6
21. gxh5? is simply bad, giving up the f5-square, and would be hit by 21... Nxb2 22. Kxb2 Bf5! and the white king and queen are caught in a horrendous cross-fire of pins and mating threats 23. Qd2 Qa3+ 24. Kb1 b3 25. axb3 Rxc2 26. Qxc2 Bxd4 27. Bc1 Qa1# )
It's very hard for a Dragon player NOT to give up the exchange! This second one is not really necessary (in fact it may not even be good! 21...Bxc6 is at least as effective) but sometimes the aesthetic appeal of chess over-rides common sense and practicality. With this move I had seen a beautiful variation which I really hoped would appear on the board, but overall (if I am being sadly honest) reading too many Tal books can seriously damage your rating!
22. dxc6 Nxb2 23. cxd7 Nxd3 24. cxd3?!
( I had been hoping for white to play the obvious rejoinder 24. Bxe7 when I had seen that 24... Qe5! 25. d8=Q+ Kh7 26. c3 Qxc3 27. Rg2 b3!
Leaves white completely defenceless despite his huge material advantage. It is variations such as these that Dragon players live and breathe for! )
Getting carried away with things! It would have been far better to leave the black pawn on b4, round up the white d-pawn imediately, and still have sufficient forces left to mount a second attack later.
25. axb3 Qa1+ 26. Kc2 Qc3+ 27. Kd1 Qxb3+ 28. Ke1 Qe6+ 29. Kd2 Qxd7
So the pawn falls, but black's king becomes a little bit airy.
30. gxh5 Bd4?
Again it would have been better to keep these pieces on the board and use the queen and bishop together to cause the white king problems. Black is better despite this error, but it would have made life easier.
31. Be3 Bxe3+ 32. Kxe3 Qe6+ 33. Kd2 Qf5 34. hxg6 fxg6 35. h5 g5 36. Rg4?!
Probably white should have pushed 36. h6 here, but then one of the rooks would have had to guard the pawn and the other would struggle to overcome the queen on it's own. Once the rooks are separated, as in the game continuation, they are especially vulnerable to the roving black queen.
36...Kh7 37. Ke2 Kh6 38. Rhg1 Kxh5 39. Rh1+ Kg6 40. Rhg1 Kf6 41. Ke3 Qc5+ 42. Ke2 Qe5+ 43. Kd2 a5 Once black gets his pawn moving, the end is nigh. If white had a safe haven for his king then the rooks could mount a defence of sorts but as it stands the queen will combine checks and pawn advances to keep white from co-ordinating his forces.
44. Re1 Qb2+ 45. Ke3 e5 46. Re2 Qb6+ 47. Kd2 d5 48. Reg2 Qb2+ 49. Kd1 Qc3 50. Ke2 a4 51. Rg1 Qc2+ 52. Ke1 a3 53. R4g2 Qb1+ and here white resigned as54 ...a2 next move will win one of the rooks. 0-1
This game is a good practical example of how to face an unusual but dangerous attempt to broach the Dragon's defences.
-Always look for the most active response to distract white from his attacking ideas (move 15.g4 was followed by move 30(!)gxh5 - that's how long it took white to get a threat in!)
-learn and use the typical tactical themes of the Dragon (exchange sacs on c3 and d5, piling Q and B up along the dark diagonal, breakthrough pawn advances ...b5,...b4,...b3).
-remember, never be scared to invest material in order to get the first big blow in!
In the next part of The Dragon's Lair we will look at some of my worst Dragon Disasters (from both sides!). We might not be able to win every time, but we can certainly learn from those games which go badly wrong!
-Coaching in the Dragon (and all other aspects of chess for all levels)