11...Ne5 is the book move. I would expect the line 11...Ne5 12. Kb1! to happen, given the immense popularity of this line lately at the top levels, and it certainly poses Black the most difficult problems. Just to give people an idea of what's happening, here's White's idea - White takes one move out to simply improve his position a little bit. In general, Black reacts on the 12th move to whatever White does, for instance 12. h4 h5! (In my opinion best) or 12. Bh6 Bxh6 13. Qxh6 Rxc3! 14. bxc3 a5! with counterplay. That said, Black runs into some slight problems if he's forced to move here. One try is just the simply 12...Nc4, and after 13. Bxc4 Rxc4 14. g4! , white is thought to have a really solid advantage here. For instance, 14...b5!? 15. b3! Rc8 16. Ndxb5 is good for White, 14...Qa5 15. Nb3 Qa6 16. e5! is good for White, and 14...Qc7 15. g5 Nh5 16. Nd5 Qd8 17. Ne2! is good for White. I'm leaving one variation out, and I'll return to it a bit later.
Another try for Black after 12. Kb1 is 12...Re8!? Black's main points to playing this move are:
1. White played a waiting move that improved his position, why can't I? The points are:
2. Black might now be able to play ...Bh8 in response to Bh6 at some point.
3. Occasionally Black runs into problems with a weak e-pawn, so ...Re8 fixes that as well.
4. ...Qa5 might now be possible, since after the old Nd5 Qxd2 Nxe7+ followed by Rxd2 trick doesn't work.
5. Occasionally White tries a plan based on f4 and e5, when he may take on f6 or d6. This opens the e-file where a rook is already placed.
That said, the position after 12...Re8 13. h4 h5 14. Bh6 Nc4 15. Bxc4 Rxc4 16. Bxg7 Kxg7 17. Nd5! is pretty prospectless for Black. For instance, the main line generally continues 17...e5 18. Nxf6 Qxf6 19. Nb3 Re6 20. Qa5 Ra5 21. Qc7! , and Black has to suffer a little bit. In my opinion, 19. Ne2!? is quite good, planning to reroute the second knight to d5. It's quite difficult to find earlier deviations for Black as well- Some people occasionally suggest 17...Nxd5?! 18. exd5 Qb6 as an improvement for Black, but he's actually losing after 19. g4! This isn't particularly surprising either, since he completely abandoned his kingside with his 17th move.
The last option is the most recent one as well, 12...a6!? This is in the same "prophylactic" waiting boat as 12...Re8, only now Black doesn't have a problem with the same Nd5 line, for example 12...a6!? 13. h4 h5 14. Bh6 Nc4 15. Bxc4 Rxc4 16. Bxg7 Kxg7 17. Nd5 e5 18. Nxf6? is met by 18...Rxd4, since there is now no 19. Nxe8+ This line was discussed quite a bit by a few Dragoneers (myself included) on the ChessPub Dragon Forum. Here is the link:
http://www.chesspub.com/cgi-bin/yabb2/YaBB.pl?num=1204238047/0
Here are some of the top level games:
http://www.chessgames.com/perl/chessgame?gid=1501664
http://www.chessgames.com/perl/chessgame?gid=1499642
http://www.chessgames.com/perl/chessgame?gid=1504596
http://www.chessgames.com/perl/chessgame?gid=1499385
http://www.chessgames.com/perl/chessgame?gid=1504215
http://www.chessgames.com/perl/chessgame?gid=1499553
There was also discussion of improvements for Black in the 12...a6 games that were lost (out of the above ones). In general, I think Black is in good shape after the popular 13. h4 h5 14. g4 hxg4! (this overzealous plan for White almost never works in the Dragon, and I don't think it does here either.)
Okay, now for me to finish the teaser I mentioned earlier:
One line I really like that hasn't been mentioned yet against 12. Kb1 is 12...Nc4 13. Bxc4 Rxc4 14. g4 b5 15. b3! b4!? , when 16. bxc4 bxc3 17. Qxc3 Qc7! is critical. It's a very complicated game, but obviously Black has compensation for the exchange. White's kingside is permanently messed up, and it's not particularly easy for him to find a plan. I've got some homebrewed analysis in this position, but we can see what everything thinks about this option if it happens. Here are two games by one of the most consistent GM Dragoneers around, Pontus Carlsson:
http://www.chessgames.com/perl/chessgame?gid=1482081
[Event "Swedish Championship 2008"]
[Site "Växjö"]
[Date "2008.07.04"]
[Round "?"]
[White "FM Victor Nithander"]
[Black "GM Pontus Carlsson"]
[Result "0-1"]
[ECO "B78"]
[PlyCount "98"]
[TimeControl "600"]
1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 g6 6. Be3 Bg7 7. f3 O-O 8. Qd2 Nc6 9. Bc4 Bd7 10. O-O-O Rc8 11. Bb3 Ne5 12. Kb1 Nc4 13. Bxc4 Rxc4 14. g4 b5 15. b3 b4 16. bxc4 bxc3 17. Qxc3 Qc7 18. Ka1
Rc8 19. Rb1 Be8 20. h4 Nd7 21. h5 Nb6 22. hxg6 fxg6 23. Rh2 Nxc4 24. Bf2 Qc5 25. Rb3 Na5 26. Qb2 Nxb3+ 27. axb3 a5 28. Kb1 Qb4 29. Rh1 Bf7 30. Rd1 a4 31. Be1 Qc5 32. bxa4 Qa7 33. a5 Rc4 34. Qb6 Rxd4 35. Qd8+ Bf8 36. Rxd4 Qxd4 37. Bh4 Qb4+ 38. Kc1 Kg7 39. Bxe7 Qe1+ 40. Kb2 Qb4+ 41. Kc1 Bxe7 42. Qxe7 Qe1+ 43. Kb2 Qb4+ 44. Kc1 Qa3+ 45. Kb1 Qa2+ 46. Kc1 Qa1+ 47. Kd2 Qxa5+ 48. c3 Qe5 49. Qc7 Qc5 0-1
The plan with ...Be8 followed by ...Nd7-b6/e5 - a4/c4 is quite strong in my opinion, and was common in my analysis as well. If Pontus Carlsson thinks it's good enough, I'd definitely play it too.
Hope you guys enjoy, sorry for the sheer amount of information. That said, for me, the decisions come later. 11...Ne5 is definitely best here, there are barely any games in my database with anything else.