Leon_Trotsky wrote on 01/16/20 at 00:45:46:

The problem with 2... a6 is that it virtually loses a tempo in the event that white plays a Closed Sicilian, since black is very likely to play ...a5 in the future. When combined with an early ...d6 (which is necessary in order to avoid being move ordered into a Kan or Taimanov) this is enough to give white legitimate chances to fight for an advantage, as for example in the line 1. e4 c5 2. Nc3 a6 3. Nge2 Nf6 4. g3 d6 5. Bg2. It's not a catastrpohe for black of course, but current consensus seems to be that it offers more of an advantage than 2. Nc3 d6 3. d4 and perhaps also more than 3. Nf3 e5 or 3. Nge2 e5/Nd4.

I would think that a Najdorf player would choose 2...a6 over both options.

TopNotch wrote on 01/15/20 at 22:32:55:

Fair enough. If one interprets the question as an empirical one about why Najdorf players would choose 2... d6 then it absolutely might be the case that laziness is a major reason. I don't really share your high opinion of the 2... d6 Grand Prix, and sincerely believe that knowledge of only concrete lines will be enough to give black the better practical chances at any level of play where people are studying openings. Of course if you do forget your lines you can be in trouble, but frankly I would struggle to contain my laughter if I heard a Najdorf player making such an argument! Especially at club level where people can expect to face the Grand Prix perhaps even more often than the sharpest Najdorf mainlines I really don't see why one would want to avoid playing and learning the 2... d6 Grand Prix lines.

TopNotch wrote on 01/15/20 at 14:30:40:

The thing is that I don't even agree with the bolded statement.  I sincerely believe that 2. Nc3 Nc6 3. Bb5 is at least as promising for white as 2. Nc3 d6 3. f4 and this so-called Tiviakov Grand Prix is increasingly taking over from 3. f4 as a reply to 2... Nc6 amongst Grand Prix players: for examples of this see ChessMood's Anti-Sicilian course and Eric Rosen's YouTube video for Saint Louis Chess Club on the 2. Nc3 Sicilian. Of course black is fine there as well, but black objectively stands at least as well after 2... d6 3. f4, so even from an objective point of view I see no reason to prefer 2... Nc6 over 2... d6 against GPA players. Furthermore, and as I indicated in my previous post, the 2... d6 move order has the practical advantage of baiting white into dubious attacking lines with an early Bc4 - the line I gave there was literally produced by meeting white's most common moves on Lichess with the theoretically approved black replies, and white was already seriously worse by move 9!
Theoretically speaking 3. g3, 3. f4 and perhaps also 3. Bb5 are so non-threatening as to be a complete non-issue in choosing between 2... d6 and 2... Nc6. The real debate is whether 2... d6 3. d4 on the one hand or 2... Nc6 3. Nf3 e5 and 2... Nc6 3. Nge2 e5/Nd4 is more promising for white, and the choices of the world's top players indicate quite strongly that it is the latter. Perhaps stylistic preference is playing a role here, as a Najdorf player is much more likely to feel comfortable in the position arising after 1. e4 c5 2. Nc3 d6 3. d4 cxd4 4. Qxd4 Nc6 5. Qd2 Nf6 6. b3 e6 7. Bb2 Be7 than the static positions arising after 3. Nf3 e5.
Of course there is a debate to be had about what counts as 'laziness' in opening preparation. Is 'laziness' the right word for a Marshall player who prefers the move order 1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 Nf6 5. 0-0 Be7 6. Re1 b5 over 5... b5 6. Bb3 Be7 (allowing 7. d4 with no benefit for black) or a Classical French player who prefers 1. e4 e6 2. d4 d5 3. Nc3 Nf6 4. e5 Nfd7 5. f4 c5 6. Nf3 Nc6 7. Be3 cxd4 over 6... cxd4 (allowing 7. Nb5, as played in an AlphaZero-Stockfish game, with no benefit for black)? The term wouldn't be entirely inappropriate, because the sole benefits of the established move orders is to cut down on white's options and therefore on black's study time. Personally I would prefer to use the term only when one is sacrificing something objectively in order to cut down on their theoretical workload. I would, for example, say that it is laziness which motivates me to allow the Nimzo-Indian instead of learning all the lines arising after 3. Nf3 d5 4. Nc3, even though I think the latter leads to a slightly larger advantage to white. In the case of our topic about meeting 2. Nc3 as a Najdorf player I really think that 2... d6 is just a better move.

Jack Hughes wrote on 01/15/20 at 11:06:31:


I stand by my statement that in otb chess Laziness is a big determining factor. In Correspondence Chess there is probably nothing to choose between the two move-orders [1.e4 c5 2.Nc3 d6 or 1.e4 c5 2.Nc3 d6] since memory and extra workload is not a factor, in otb chess however if a Najdorf player can make 2...d6 work satisfactorily he will. It is well known though that the 2...d6 move-order offers Grand-Prix Attack players much more dangerous attacking options than after 2...Nc6, so why do it.... Laziness. I doubt you will find many Dragon players using the 2...d6 move-order, because why allow additional attacking options if you don't have to, I guess you could say laziness again or use a euphemism like 'pragmatism' if laziness offends your sensibilities.   

Come on, think about, you know what I'm saying is true. 

In response to the idea that Najdorf players are simply choosing an inferior line in order to save their work load I would point out that it is not just club players doing this but also the very elite: looking at TWIC games from 2019 and 2020 where both players were rated above 2700 black has played 2... d6 in 16 of the 20 games to reach that position, and of the three games with 2... Nc6 one was with Grischuk (a Sveshnikov player) as black and the other two were blitz games. These statistics are of course greatly skewed by the fact that white players are unlikely to go for 2. Nc3 unless their opponent is a Najdorf player, but it is noteworthy that the Najdorf specialists (in particular MVL, Nepomniachtchi and Giri) all prefer 2... d6. Even in the 2019 ICCF archives, where strong players are likely to be more flexible in their openings than MVL or Nepo and where the practicalities of studying lines are a complete non-issue: if one includes only games where both players were above 2400 then 24 players have chosen 2... d6 with a score of 52.1% (-36 elo) compared to 52 who have chosen 2... Nc6 with a score of 50% (-36 elo).
Taking the analysis away from the abstract and into the concrete I'm also really not sure what basis there is for thinking that 2... Nc6 is the objectively superior move. True, it allows white a slightly improved version of the Closed and the Grand Prix but neither of those is improved enough to let white fight for an advantage. From a practical perspective I would even prefer to play 2... d6 against a known Grand Prix player, since this move is much more likely to provoke them into self-destructing with a line like 1. e4 c5 2. Nc3 d6 3. f4 g6 4. Nf3 Bg7 5. Bc4 Nc6 6. 0-0 e6 7. d3?! (Grischuk's 7. d4 is in my view the only playable option for white) Nge7 8. Qe1 0-0 9. f5  gxf5 when black is already seriously better - if you instead play 2... Nc6 and delay ...d6 then white is statistically more likely to play a much sounder approach with Bb5 instead of Bc4. All this is to say nothing of the line with 2. Nc3 Nc6 3. Bb5, which is a spiritual cousin of the Grand Prix that is at least as promising for white as anything after 2... d6 3. f4, even if not promising enough to really fight for an advantage. Theoretically speaking the biggest downside of 2... d6 has to be 3. d4, but the idea that this is obviously more promising for white than 2. Nc3 Nc6 3. Nf3 e5 or 3. Nge2 e5/Nd4 is somewhat mystifying to me. The idea certainly is not supported by any of my engines or by the opening choices of the world's elite in either OTB or correspondence chess.

LordChaos21 wrote on 01/11/20 at 14:54:14:


Sorry, my fault, I had misread the post and thought you were talking about the position after 2. nf3 not Nc3.

I don't see why it should be laziness; i agree if you intend to, after say 1.e4-c5 2.Nc3-Nc6 3. Nf3 (or Nge2), to transpose to the Sveshnikov, then it is indeed a lot more work. But if you are willing to play the position after 3...e5, it's just one line, which is also easy to learn in my opinion.

As for the Closed Sicilian move order, the problem is when White is good with his move-orders. He can play 2.Nc3 against a Najdorf player. Say we play 2...d6, he goes for 3.d4 and I think White is supposed to have an edge here. A Najdorf player usually to my knowledge prefers 2...a6 (MVL for example has played this), but then White goes for 3.g3, and we are into a bad version of the Closed Sicilian (again, in my opinion).
So the only solution to this conundrum for a Najdorf player seems to be 2...Nc6.