So I've been looking into the 3. Bb5+ Sicilian, and MCO is not very helpful. It barely mentions a Maroczy bind type of setup, but that seems to be what all the GMs are playing. The most logical plan for White (in the 2. ...d6 Sicilian) meets the following equalizing maneuver:

1. e4 c5 2. Nf3 d6 3. Bb5+ Bd7 4. Bxd7+ Qxd7 5. O-O Nc6 6. c3 Nf6 7. Re1 e6 8. d4 cxd4 9. cxd4 d5! 10. e5 Ne4 =

What about 8. c4!? The point is that if White is going to play c3 and d4, Black wants his bishop on g7, not e7. But once Black plays ...e6, then after White plays d4 Black can't just play ...cxd4 without having to worry about his weakness on d6 and hemmed-in bishop. But if he plays ...g6, then his kingside dark squares will be weak, and he still doesn't particularly want to play ...cxd4. And due to the position of the rook on e1, White can prevent ...d5 just long enough to play d4-d5, when Black doesn't want to take (because it would open the d-file leading to d6), but if he plays ...e5, then White will have the better game due to the superiority of his bishop over Black's. He will have a hole on d4, but with the better minor pieces and more space, he should have a lasting advantage.

If White doesn't want to incur a hole on d4, he can simply decline to play d5 and hold the tension in the center, with an edge.

Thoughts?

I'm not an expert on this kind of position by any stretch, but I don't see why ... Be7 d4 cxd4 Nxd4 0-0 is bad for black at all. d6 is backwards, but it's really well defended by pieces. Black just seems tempi up over a normal bind structure. If we play fantasy chess and start with no light-squared bishops, we get

1. e4 c5
2. Nf3 d6
3. 0-0 Nf6
4. Re1 e6
5. c4

and before black moves here, he gets Qd7 Nc6 in for free, which are useful moves. I'd be pretty shocked if chopping light-squared bishops here is worth more than 2 full tempi.

I'm not so sure that the queen does anything on d7, since it would probably prefer to be on a dark square. Also, in a Maroczy bind structure the LSB is White's worst piece but is usually a good piece for Black.

there is an alternative plan to the equal line you give: 7. d4! is a fun gambit that is normally far too tempting for Black to decline (it might only lead to equality if accepted theoretically though), if he does decline it with cd cd d5 e5 Ne4 we end up in the same position you give above except that White hasn't played Re1 and Black hasn't played e6 so White has 10. Ne1! with the idea of f3 trapping the knight or at least running it out of that good square.

I don't think your idea is totally without merit though, I can never figure out what to do in these Maroczy bind positions, so spending a tempo doing nothing might be better than my normal play, and spending two tempi getting rid of the light squared bishops must be good.

Interesting and creative idea but I don't really see how white can really lay claim to any advantage. The weakness of d6 in these positions is almost always just illusory. I see a simple plan for black to improve his position since white basically has to play d4 at some point to get anything at all. Black can then simply play: Be7/o-o/a6/Rc8/b5 and get good queenside play. I'm not sure what white should do. But I think I'm also in the same shoes as RoundTower here. I just don't see what white does in Maroczy Bind positions. They look so good - tons of space, easy development and black is cramped and has a weak d pawn. Yet somehow it seems black often is the one with the best of these positions.

1. e4 c5 2. Nf3 d6 3. Bb5+ Bd7 4. Bxd7+ Qxd7 5. O-O Nc6 6. c3 Nf6 7. Re1 e6 8.c4 Ne5!? is a common idea, and i don't see an advantageous answer for white.

Very good thread. I have plenty to add:

10...Ng8 is even better as it allows Black to play for a win.

Well the thing is that White can choose to play c4 in one go at an earlier point. For example 5.c4 - here White is offering a gambit because Black could grab a pawn with 5...Qg4 but this is considered to be very dangerous for Black after 6.0-0 Qxe4 7.d4.

So the point of delaying c4 could only be that White waits for Black to commit his dark-squared bishop, i.e. setting up a Maroczy once Black has played e6 and playing c3 and d4 after Black has fianchettoed.

However, I don't believe that White has to sacrifice a tempo (c2-c3-c4) to do this, like he does in your OP line.

That's what I think as well. It's fine to give up the tempo compared to your fantasy line, because the light-squared bishop really sucks in a Maroczy structure but I really don't see why we would spend another tempo for c2-c3-c4. I haven't dug deeper yet but I can well imagine that Black will be able to use those tempi for getting in the d6-d5 break.

The line is brutal if Black is hit unprepared. I once handed a mighty blow to a booked up Dragon player with it (-:

Unfortunately, Palliser has a good a line against it in his 3.Bb5 Sicilian book, based on a game of Ivanchuk...

There's a lot of such nasty sh*t that Black can fall victim to in the Rossolimo. I once analysed the following idea with a colleague of mine:

1.e4 c5 2.Nf3 Nc6 3.Bb5 g6 4.0-0 Bg7 5.Re1 Nf6 6.c3 0-0 and now not 7.d4 (the main line) but 7.h3 - Black now can't really play d5 because the knight cannot go to e4. If he plays 7...d6 White will play 8.d4 and will have an extra tempo (h3) if Black chose to play d6-d5 now. Back then we tried to find out how this extra move could be utilized best.

I used to play the 7.d4 gambit line, but a big issue that eventually drove me off the whole line was that 5.-Nf6(!) prevents that and 6.e5 lacks bite in the established variations starting with 6.-dxe5 7.Nxe5 Qc8.

That's quite true. I have also noticed this and I'm still to find the reason why every world class player still plays 5...Nc6 first. It must either have to do with 6.e5 (maybe 6.d4?) or the world class players just don't like 6.e5 because the position becomes too drawish...

Does anyone have Rybka? I'm fairly certain that if Black can use his two extra tempi to force ...d5 through, Rybka will find it and play it. Unfortunately I only have Crafty (via FICS), which is a very weak engine, only 2500 ELO or so.

Rybka wants to play Be7, 0-0, a6 and b5 which I guess is the drawback of not having the light squared bishop.

Okay, I've been thinking about this position and I truly believe that this can work well for White. The idea is that after d4 cxd4 Nxd4 (or else White will play d5), White is going to willingly create a hole on b4 by playing a4 and b3. Black will put a rook on c8 to try to get in ...b5 based on the undefended c3 knight, but White will just defend the knight by playing Qd2.

At this point Black will be unable to play ...b5 or ...d5, so he will naturally seek to exploit the b4 hole. But White will just retreat his d4 knight to c2 and play Ba3, covering the hole, clearing the d file, and aiming the bishop at d6 (at which point one has to wonder how "illusory" the d6 weakness is). Black can't get another piece to the b4 point fast enough, and White will switch his queen to d3 once the b4 point is secure and then triple on the d file.

I don't see how this isn't just great for White. Can Black somehow force through ...d5 or ...b5, or can he get an attack on the kingside? If Black tries to arrange ...d5, White will just play Nc2 before other moves, and then Qd2 and Rd1, and Ne3 if necessary. The Black knight will do absolutely nothing on b4 even if it can get there and stay there, because Black will never get ...d5 anyway, and if he plays ...a5 to support the knight then after White takes the b4 point away Black is just really bad because he'll never get either of his pawn breaks in, and White can challenge the b4 knight at his leisure, clearing the a3-f8 diagonal for his bishop. It looks like Black has to play ...Ne8 and ...f5, but I don't have any faith in this idea because it just weakens e6.

Okay, here is a game that I played against Crafty with its analysis running, so it was basically Crafty vs. itself with me forcing it to play what I considered to be strategically superior moves for White on occasion. Keep in mind that Crafty is only ~2500 ELO. Still, I believe that this game is illustrative. The engine decides early on that it's not going to get to play ...b5 or ...d5, so it posts its knight on c5 in an attempt to tie down one of my rooks while it looks to gain kingside counterplay by lifting both rooks to the g file and pushing its f pawn to f4. Note that if the LSBs were on the board Black would probably win by first forcing White's kingside pawns onto g2, f3, and h3, and then saccing the bishop to let the rooks, queen, and f pawn in on the dark squares, a la KID. However, the black knight is not up to the task as it must keep an eye on the d6 weakness...

1. e4 c5 2. Nf3 d6 3. Bb5+ Bd7 4. Bxd7+ Qxd7 5. O-O Nc6 6. c3 Nf6 7. Re1 e6 8. c4 Be7 9. Nc3 O-O 10. d4 cxd4 11. Nxd4 a6 12. a4 Nxd4 13. Qxd4 Qc7 14. b3 Nd7 15. Ba3 Nc5 16. Rab1 Rad8 17. Qe3 Bf6 18. f3 b6 19. Ne2 Bh4 20. Red1 f5 21. Nd4 f4 22. Qd2 Bf6 23. Bb4 Be5 24. Qe1 Rb8 25. Bc3 Bf6 26. Ne2 Bxc3 27. Qxc3 Rbd8 28. Rd4 Rf6 29. Rbd1 Nb7 30. R4d2 Qc5+ 31. Qd4 Qc7 32. Qf2 Rdf8 33. Nc3 Rg6 34. Na2 Rg5 35. Nb4 a5 36. Nc2 Rf6 37. Nd4 Rfg6 38. Qf1 Qc5 39. Kh1 Rh5 40. h3 Rhg5 41. Nb5 e5 42. Qf2 Qc6 43. Qh4 h6 44. Nc3 Nc5 45. Nd5 Kf8 46. Rb2 Ne6 47. Qe1 Rg3 48. b4 Qxa4 49. Ra1 (Black ought to resign) 1-0

The absence of the LSBs allowed White to place all of his pawns on light squares without worrying about a sacrificial breakthrough or winding up, e.g., with a bad bishop vs. a centralized knight in the endgame.

Hopefully this will spur some discussion. Interestingly, Crafty considered White to be up about a tenth of a pawn for most of the game, until suddenly it realized that Black's position was garbage, at which point the evaluation quickly surged to .6 and then, finally, to 1.4.

It is true that this was a bit like building a strawman and then knocking it down, since Black had only the engine while White had the engine and a human's positional sense. However, it can't be denied that Black played reasonably if somewhat shortsightedly. The move ...b6 was a key error, allowing the knight to attack the a6 pawn, forcing it to advance, creating a hole on b5 from which the White knight was able to attack the d6 weakness. I also believe that this position is closed enough, and White's plans involve enough long-term maneuvering, that I could beat Crafty as White from the starting position (after White's c4!?) even if I didn't have the analysis running.

I've found one single game in my database (CB Mega 2009) so at least it's not a novelty, albeit rare. In my opinion it's not surprising that it's rare because White could reach the position after 8.c4 with a tempo up if he's willing to allow the fianchetto.

Ok, so what does the game tell us? Black pushed through d6-d5 at first opportunity and according to an engine check, it leads to complete equality after 13.exd5.

The game continuation 13.e5 is just crap, leading to a huge advantage for Black.

We also see that it is 12.b3?! that allowed Black to play 12...d5. Alternatives that prevent 12...d5 are 12.Be3 and 12.Nc2. Other than that White has no sensible alternative moves.

After 12.Be3 I considered 12...Ne5. It is totally clear that Black doesn't want to play a passive default Maroczy, because as Discipline already said, the absence of the light-squared bishop favours White. After 12...Ne5 13.b3 a6 14.f4 Black can keep that knight at e5 with 14...Qc7, preparing to retreat it to d7 after White did something against the d-file pin. He'll then bring the knight to c5, putting pressure on e4. I really don't see how Black is any worse here. After 13.Qe2 instead if 13.b3 Black can e.g. stir up a fuss with 14...Nfg4 15.Bf4 Bh4 but I haven't had time to look at the consequences yet.

The other alternative 12.Nc2 looks more sensible. Play might continue 12...a6 13.b3 Rac8 14.Bb2 Qc7 and at least White has reached some kind of default Maroczy now. Black must remember to bring his Nc6 to c5 instead of the Nf6 by means of Nc6-e5-d7-c5. This is for example what Crafty messed up, but Craftly is known to be useless at positional chess.

In the game Viika just referenced white has an interesting continuation that even rybka initially missed:

12. .. d5 13. Nxc6 bxc6 14. exd5 cxd5 15. cxd5 Bb4 16. Qd4 Nxd5 17. Nxd5! with the idea of 17. .. Bxe1 18. Bb2 f5 19. Nf6+ etc

White has some threats although it will probably settle down into a draw pretty quickly. But I think the position is quite drawy anyhow, so may as well give your opponent a few chances to go wrong before shaking hands.

Compared to the line 1. e4 c5 2. Nf3 d6 3. Bb5+ Bd7 4. Bxd7+ Qxd7 5. O-O Nc6 6. Re1 Nf6 7. c4 e6 (or 7...g6 or 7...Ne5 or 7...e5) which occurred in some games without yielding White anything, you're a tempo down, but still OK, of course. So you can't claim any opening advantage but it's more important to like the position and play it well.

As for your game the plan 19...Bh4, 20...f5 is unusual and slightly suspicious. Obviously Crafty was trying to stirr things up. These programs think that they are the superior player and tend to do that. If 19...d5 is not possible the typical hedgehog player would probably play something like 19...Rfe8.

Am I missing something or isn't 15...Nxd5 instead of 15...Bb4 simply equal? I have neither a board nor an engine at hand but as far as I can see I'll just keep the pawn at e6, recapturing with pieces on d5.


Crafty just doesn't have much positional knowlegde, so it's clueless about this type of position. When I was coding my chess engine, I've had a couple of looks into Crafty's source code and there was mostly tactical stuff in its evaluation function.

15. .. Nxd5 seems kind of bad to me. It definitely kills the game, but why would black want to do this? He finally makes his thematic break, his pieces are coming to life, white's development is very awkward and black plays to quickly draw the game?

15...Nxd5 is the emergency exit in case Rybka's line proves to be dangerous. It shows that Black is able to equalise.

It's not Rybka's line. It's my line. Rybka gives it as somewhat better for black and needs to be walked through to realize it's equal at worst. I like it since it gives black practical losing chances.

Obviously I would play Nc2 rather than Be3, and I haven't even looked at that game or position--I just know that I'm not going to allow ...d5, or else the whole idea is crap and White has just lost time.

As for playing 7. c4 to save a tempo, the whole idea is that if Black fianchettos I'd rather go c3-d4, but if he plays e6 then I want a Maroczy Bind without LSBs.

It is tough to explain, but the Maroczy-Setup without light squared bishops is lacking bite. You cannot prevent any of the black pawn breaks. Also the pawn d6 will never become weak. You may play d4 and attack it later, but black always has enough pieces to defend it.

Last but not least, the idea was played (twice) already. The oldest game is Nezhmetdinov-Shamkovich, UdSSR 1956 and the result was a draw (of course).

I can and will prevent both ...d5 and ...b5, and I will attack on the weakened light squares if ...f5. If Black can't achieve anything, I will lift my rooks and play for a direct attack on the king.

I know that even after the c4 d4 lines, black has gotten away with a quick e6, d5. I think Christiansen had black against someone in the past, maybe Wolff vs Christiansen or vice versa from like 15 years ago. White will be down a tempo so probably should work out for black. Haven't looked at board or position at all though.

Technically, the idea was only played once. In the game you referenced, Nezhmetdinov played c3-c4 and then followed up with d2-d3.