Thanks very much for the comprehensive overviews, TN. I've been mostly following Dismantling the Sicilian (which also is English Attack-based) but I may explore a few of these sidelines for when I don't feel like going down the main line of theory.

Here's my view on each of the main Anti-Sicilian setups:

2.a3 - completely harmless. 2...g6 is the simplest equaliser.

2.d3 - This is harmless, but only if Black are prepared. Shaw's antidote in 'Experts vs. the Sicilian' is pretty convincing.

2.Ne2 - I think this move is somewhat underestimated. White can follow with g3, Bg2, 0-0, c3 and d4, reaching original positions and even maintaining slight chances of being better.

2.c4 - If you are going to play like this, you should play c4 a move earlier. 2...Nc6 3.Nf3/Ne2 e5 is completely fine for Black.

2.b4 - Objectively equal, but Black has to know his theory. A decent choice for a gambit fan willing to study some theory.

2.b3 - Creative, but not very challenging. Either ...Nf6/...g6 or ...d6/...e5 setups are pretty balanced. 

2.g3 - Similar to 2.d3. 2...d5 is probably the most annoying line. 

2.f4 - This version of the Grand Prix is an interesting surprise weapon in the hands of a prepared player. If Black avoids 2...d5 then White's chances of an edge are not bad, but after 2...d5 3.ed5 Nf6 4.Bb5 both ways of blocking the check are quite strong. I'd even say that Black's position is easier to play provided he knows the theory.

2.d4 cd4 3.c3 - Another line that is a decent surprise weapon, but it can only be recommended for gambit fans who are willing to put in a lot of work. 3...Nf6 is quite annoying for a SMG player too.

2.c3 - In my view the Alapin is as good as the Rossolimo/Moscow. I like White's chances of achieving an edge in the 2...d6 3.d4 Nf6 4.Bd3 and 2...e6 3.d4 d5 4.ed5 lines. 

2.c3 d5 - After 3.ed5 Qd5 4.d4 White has good chances of emerging with a small advantage in the 4...Nc6 5.Nf3 Bg4 or ...e5 lines, but 4...Nf6 5.Nf3 Bg4 and 5...e6 are objectively equal. 

2.c3 Nf6 3.e5 Nd5 - Delaying d4 for a long time (4.Nf3 and 5.Bc4) is harmless. After 4.d4 cd4, 5.Qd4 is a tricky sideline but with best play Black's position is easier to play. After 5.Nf3, both 5...e6 6.cd4 d6 and 5...Nc6 6.cd4 d6 7.Bc4 Nb6 are equal, but against other lines White is a bit better. After 6...d6 White should play 7.Bc4 Nc6 8.0-0 Be7 9.Qe2 0-0 10.Rd1 or 10.Qe4, and after 7...Nb6 I think 8.Bb5 is objectively better, but at lower levels 8.Bb3 is probably the best practical choice.

2.Nc3 a6 - White has some chances of an edge if he plays a GPA setup. Other development schemes are harmless.

2.Nc3 d6 - Here 3.f4 gives White the easier to play position, though objectively Black cannot be worse.

2.Nc3 e6 - This is annoying for Closed Sicilian and Grand Prix Attack players since 3.f4 d5 and 3.g3 d5 are both equal for Black. 

2.Nc3 Nc6 3.Bb5 - I really like this line for White. In my view only 3...Nd4 4.Bc4 e6 5.Nge2 (more testing than 5.Nf3) 5...Nf6 ensures equality for Black, and even then Black has to be well prepared to guarantee that.

2.Nc3 Nc6 3.Nge2 - This is not a bad line, and will definitely catch a number of players out. 3...e5 is close to equal, and after 3...Nf6 or 3...g6, 4.g3 is a reasonable way to keep play in Anti-Sicilian waters. Closed Sicilian players will probably like this line.

2.Nc3 Nc6 3.f4 e6 4.Nf3 - I think 3...e6 is Black's best reply to the Grand Prix Attack. Avrukh's line in 'Experts vs. the Anti-Sicilians' is a convincing antidote theoretically, but in practice most Black players play one of the lines that give White an edge after 4...d5 5.Bb5. 4...a6 and 4...Nge7 are both a bit annoying for White but he should have some chances of an advantage after these moves.

2.Nc3 Nc6 3.f4 g6 4.Nf3 Bg7 - 5.Bb5 Nd4 6.0-0 is perhaps a bit easier for White to play, but about equal. 5.Bc4 e6 is definitely not worse for Black, but he has to know his theory well after 6.f5 and hence this is a good practical choice for White. 

2.Nc3 Nc6 3.g3 - This is not theoretically testing at all, but can be recommended for fans of system openings and fairly strategic positions. After 3...g6 4.Bg2 Bg7 5.d3 d6 I think 6.Be3 is White's best try for an edge, when White has chances to be better after 6...Rb8 and 6...e5 but not after 6...Nd4 or 6...e6. After 6.f4, the critical line is 6...e6 7.Nf3 Nge7 8.0-0 0-0 9.Be3 b6.

2.Nf3 g6 3.c3 - I sort of like this line, since after 3...d5 4.ed5 Qd5 5.d4 White is slightly better, but after 3...Bg7 4.d4 cd4 5.cd4 d5 6.e5 Nc6 or 6.ed5 Nf6 Black's position is not worse. Nonetheless this can be recommended to anyone with experience in the Alapin.

2.Nf3 g6 3.d4 cd4 4.Qd4 - I used to think this line was better for White, but 4...Nf6 5.e5 Nc6 6.Qa4 Nd5 7.Qe4 Nc7 is approximately equal. Alternatively, 5.Bb5 Nc6 6.Bc6 dc6 is comfortable for Black.

2.Nf3 a6 - 3.c3 is often recommended but it leads to fairly typical Alapin positions after 3...d5. 3.c4 is probably best, followed by Be2, 0-0, Nc3 and only then d4. 

2.Nf3 Nf6 - I like the 3.e5 Nd5 4.d4 line best, but 4.g3 may be more fun for Anti-Sicilian exponents. I like Black's chances of equalising after 4.Nc3 from a theoretical perspective, but in practice White is clearly for preference.

2.Nf3 Nc6 3.Bc4 - This is rubbish and not worth your time.

2.Nf3 Nc6 3.d3/g3 - The KIA is a good system opening for lazy players, but it does not pose any theoretical problems for Black. A kingside fianchetto is the simplest equaliser.

2.Nf3 Nc6 3.Nc3 - I really like this move order trick. I think White is very slightly better after 3...e5 4.Bc4 Be7 5.d3 followed eventually by Ng5, and 3...Nf6 and 3...g6 4.Bb5 are standard lines of the Rossolimo. 

2.Nf3 Nc6 3.Bb5 - Black's alternatives to 3...g6, 3...e6, 3...d6 and 3...Nf6 all give White a slight edge.

2.Nf3 Nc6 3.Bb5 Nf6 - I quite like this line for Black. 4.Bc6 is harmless, and after the critical 4.Nc3, I think 4...g6 gives Black better chances of full equality than 4...Qc7, in light of the line 5.0-0 Nd4 6.Re1 a6 7.Bc4.

2.Nf3 Nc6 3.Bb5 e6 - This is a tough line to crack but still offers both sides plenty of dynamic possibilities. I haven't found a clear route to equality for Black in the 4.c3 sideline, but I don't think 4.Bc6 bc6 5.0-0 Ne7 poses Black any theoretical problems. After 4.0-0 Nge7, White can reach Open Sicilian type positions with 5.Nc3 or 5.Re1 followed by meeting ...a6 with Bxc6 and d4, but I prefer 5.c3 a6 6.Ba4 when I think White's chances of an edge aren't much worse than in the Open Sicilian.

2.Nf3 Nc6 3.Bb5 g6 - Of the main answers to the Rossolimo, this is the one I most like facing. 4.Bc6 is objectively equal, but after 4...dc6 5.d3 Bg7 6.h3 Nf6 7.Nc3 I think 7...Nd7 8.Be3 e5 9.Qd2 h6 is the only line which doesn't give White a nagging edge. I prefer 4.0-0 Bg7 when 5.Re1 is best met by 5...e5 when Black is not worse, but 5.c3 is better, when after 5...Nf6 White's best tries for an edge are 6.e5 Nd5 7.d4 cd4 8.cd4 0-0 9.Nc3 and 6.d4. Instead 6.Re1 is not challenging for Black, though 6.Qa4 is only equal if Black knows his theory.

2.Nf3 e6 3.g3 - This is probably the most accurate way of reaching a KIA. After 3...d5 I think 4.ed5 ed5 5.d4 offers decent chances of an edge but 3...Nf6 is of course equal.

2.Nf3 e6 3.d3 - similar to 3.g3. Almost any reasonable system of development gives Black equality.

2.Nf3 e6 3.c4 - I am not a fan of this line at all. 3...Nc6 4.Nc3 e5 is very stodgy, and 4...Nge7 is also completely equal. However one line that favours White slightly is 4...Nf6 5.Be2 d5 6.ed5 ed5 7.d4 and White's extra tempo is not insignificant.

2.Nf3 e6 3.c3 - This is quite similar to 2.c3. After 3...d5 I think White should play 4.ed5 with some chances of an edge, otherwise 4.e5 d4 is completely equal, as demonstrated in 'Experts vs. the Anti-Sicilians'.

2.Nf3 e6 3.b3 - I can recommend this line to players who like to go their own way in the opening. 3...Nf6 is probably the simplest equaliser but an unprepared opponent could easily face some difficulties.

2.Nf3 e6 3.d3 - Similar to 3.g3. You can't fight for an opening advantage with the KIA, but it is very easy to learn and play.

2.Nf3 d6 3.Bc4 - This is a good practical choice. Objectively it is just dead equal, but White's opening system is quite sensible and an unprepared opponent could definitely slip up.

2.Nf3 d6 3.b3 - This is similar to 2...e6 3.b3, only that Black could probably equalise even without any theoretical knowledge.

2.Nf3 d6 3.Nc3 Nf6 4.e5 - This line is not dangerous for Black, but will appeal to fans of a slow positional game. 

2.Nf3 d6 3.c3 Nf6 - Of White's options, 4.Bd3 Bg4 is comfortable for Black, 4.Be2 is equally harmless, but 4.Bd3 is quite tricky and can be recommended as a surprise weapon. If White has studied the line properly, Black equalises only if he has some sort of answer prepared.

2.Nf3 d6 3.Bb5 Nd7 - I have been a fan of this line ever since its advocation by Hillarp Persson in 'Experts vs. the Anti-Sicilians'. White's position seems a bit easier to play but objectively the position is dynamically balanced. 4.d4 is White's best try for an edge, but tends to like to Open Sicilian type positions.

2.Nf3 d6 3.Bb5 Nc6 - White has a large choice against this line but I don't think any of his options give him an advantage. 4.Bc6 bc6 5.0-0 e5 or 5...Bg4 are very balanced, but 4.0-0 Bd7 5.Re1 is more critical. After 5...Nf6 6.c3 a6, 7.Bc6 Bc6 8.d4 Be4 9.Bg5 only requires Black to be a bit careful to fully equalise, and 7.Bf1 Bg4 8.d4 is just balanced. White's best try is 8.d3 followed by 9.Nbd2 when White has a slight, slight edge. 

2.Nf3 d6 3.Bb5 Bd7 4.Bd7 Nd7 - In my view this line is a bit underestimated, but nonetheless after 5.c4 Nf6 6.Nbc3 I prefer White's game slightly.

2.Nf3 d6 3.Bb5 Bd7 4.Bd7 Qd7 5.0-0 - I don't think the c3/d4 lines are a theoretical test for Black at all, but they certainly require Black to be well prepared if he is to count on equalising out of the opening. 

2.Nf3 d6 3.Bb5 Bd7 4.Bd7 Qd7 5.c4 - Based on 'Experts vs. the Anti-Sicilians', White's position is slightly more pleasant but Black achieves decent counterplay. 

Personally I think the Alapin Sicilian and Rossolimo/Moscow are White's best tries for an advantage other than the Open Sicilian. The next-best Anti-Sicilians are probably 2.Ne2, 2.Nc3 Nc6 3.Bb5 and 2.Nf3 d6 3.c3.

When restarting playing chess otb in some tournaments and cc three years ago I started building up an anti-sicilian repertoire. I'm happy to see it's very similar to the Gawain Jones recommendations in his new book.

What to choose depends on your goals. If you want to have something similar against the french and sicilian e6 lines have a look at the kings indian attack - Emms Starting out is a great introduction to this. If you want to learn a lot about middlegame patterns - Sveshnikov's c3 - book is for the average club player a 12 on a scale from 1 to 10. For the Bb5 systems take the new Bologan book.

Atm there's really a great time for anti-sicilians on the book market! It gives me stuff to read and work for years.

Btw - the material in the chesspublishing.com anti-sicilian section is a great deal too in my opinion, if you haven't subscribed yet. The pdf-files give a good introduction and going through the commented games helps a lot too.

Stigma's recommendation and Jones book give a good start.

Edit:

Helpful was for me to see the choice of systems against the sicilian principally. 1. - c5 is a weak move. It works against the center but does nothing for the development. If you play later d4 in the open sicilian manner as white you answer with puting one and a half imp on one imp. Take the time black has given with preparing this move adequaltely. (Cum grano salis.)

Quote:


Well, I was writing from my perspective. True, that OP situtation might be different but the principle remains. If you go for open sicilians you need to put 10x more time to study theory than black player. If you are not a pro, you will be worse prepared in every game you play because in every game you will be walking straight into opponents pet line.
This is argument for playing less known moves if they are still decent of course. Going 2.c3 is probably not "decent" in a sense that black has many systems leading to interesting/equal play. This is not the case against Rossolimo or against 2..e6 3.c3 though.

If you are not super GM you may as well try to look for advantage in slightly offbeat line than in main line Loventhal or w/e black player throws at you because even if (and that's big if) white is objectively a bit better there than in Rossolimo for example it doesn't matter anyway as preparation and experience difference will trump it. The weaker the level, the stronger the effect.

Quote:


Why ?
Assuming you are amateur player you have limited resources to learn your theory. If you use it to prepare something for every main line sicilian and every main line spanish for example (Berlin, Open, Breyer, Chigorin, Marshall, Zaitsev, Janisch, Steinitz, Archangielsk, 3..Bc5 stuff and w/e else is left) you will be worse prepared than your opponents who decided to put the same limited resources into learning some pet lines of theirs.
As long as those lines are playable, it's just better practical strategy.

I really line 2.Nf3 and then decide idea.
I think that vs 2..Nc6 Rossolimo is better choice than d4. Black has so many options after this, all requiring tons of theory to even try for pleasant position.
For example vs Loventhal it's not clear if white has anything (and it's easy to lose if you don't know your stuff). Maroczy might be good for white but you have to like this kind of positions. In Sveshnikov again, you have to study huge system to hope for anything. Kan - too many better players lost points against guys who only new to go a6-b5, bb7, h5 and put the knight on g4. If you don't know theory there, you have no chance.
3.Bb5 is basically main line and annoys the hell out of black players Highly reccomended.

Against 2..e6 again I think that going for antisicilians is good. 2..e6 usually indicated you will face either Kan or Taimanov, both requiring heavy preparation and experience to play for anything.

Against 2..d6 I for one like open sicilian. All the trickery is left out and it's white who chooses a system against both Dragon and Najdorf. Also I don't find any antisicilian compelling vs 2...d6. Probably 3.c3 and 4.Be2 is the way to go if you really don't want enter open sic.

I think that the mantra "play open sicilian, it's good for you chess" is bad. These days people are prepared too well and it's really no fun to prepare vs 10 or so major systems (Najdorf e6, Najdorf e5, Najdorf Ng4, Kan, Taimanov, Dragon, acc Dragon, Sveshnikov, Qb6, Rauzer, classical) just to face Loventhal player, who plays only that, knows his theory up to move 20 and has ton of experience in this very unusual but sound (as computer proved) opening.

I agree with MNb that certain Anti-Sicilians are theory heavy. The Smith-Morra Gambit is one of them. If you're giving up material, you'd better know what you're doing, so I'd say playing the SMG requires just as much prep as the Open.

But I doubt if most other anti-Sicilian lines require that much theory. I can't say for sure, as the SMG and Open are the only ones I've ever looked at in detail. 

But in wanting to get away from the SMG, I bought Jesus de la Villa's book "Dismantling the Sicilian", which is an Open Sicilian repertoire for white based on an English Attack setup in most lines. But there are so many variations, each one requiring different setups, that I quickly lost track. Given that by definition, I'd be walking into my opponents' pet lines in every game, I would have to put in a ton of study to feel ready to begin playing the Open at my level. 

At lower levels, it doesn't matter as much if you know the theory, so it's probably best to begin playing the Open right from beginner level, and learn a little as you go. But as a 1700 trying to score some upsets against 1900-2000+ opponents, it requires more study than any other opening I can think of. 

That's why a quieter anti-Sicilian might be easier for me. In something like the Grand Prix or Alapin, both of which I've considered taking up, I'm avoiding my opponents' pet lines, not giving up any material as in a gambit, and these openings aren't as sharp as the SMG or Open, so one minor inaccuracy is less likely to be lethal (for either side). That makes for easier openings to begin playing. 

I have had fantastic results with the Wing Gambit but I put in as much time researching it and learning it as a main line would take.

My recommendation is that if you intend to head for the open sicilian later regardless... then it is better to start there now.

Pingudon wrote on 07/20/11 at 16:05:30:


First of all, what level of player are you? I think the answer for a 1400 level player would be very different than the answer for a 2000 level player.

Second, is there a reason you're only considering those 3 options? Personally, I've done well with the Smith-Morra Gambit, but I'm only 1700 (USCF). I know a couple of guys rated 1800-2000 who used to play the SMG, but gave it up because they claim it just isn't strong enough against guys above 1800. So I'm considering switching to the Grand Prix, until I can find some time to really focus on learning the Open. Or I might just stick to the Morra until I actually start encountering these great Sicilian players who it won't work against, instead of taking other people's word for it. I haven't decided yet.

But my point is that you might want to consider the Grand Prix. If you're below 1600, you might want to consider the Smith-Morra Gambit, but given its reputation, I don't know if I'd recommend anyone above that level first start out with it. Like I said, is there a reason you're only considering those 3 variations?

Third, how do you define the word "best"? Absolute theoretical strength? Easiest to get at least a slight advantage with? Easiest to learn? Most fun to play? I'd think that below FM level (at least), any opening that isn't regarded as complete rubbish should be good enough to get a reasonable game. Again, that might (arguably) exclude the Smith-Morra Gambit, but all 3 of your initial choices would qualify, along with the Grand Prix and maybe other anti-Sicilians.

--Fromper