Essentially, the reason boils down to the fact that Nc6
develops a piece, while a6
does not. First, we'll play through variations in the main line showing how black should and should not fight for the center. Then, let's look at a few concrete lines showing why a6
first is inaccurate.
0. Main lines and fight for the d4
square.
[FEN ""]
1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.f3 O-O 6.Be3 Nc6 {Black wants to push `e5` to control the center.} 7.Nge2 {This prevents `e5`, so instead black looks for counterplay on the queenside by the means of `b5`.} (7... e5?! 8.d5 {Forcing black back, since otherwise it loses a pawn.} Ncd4 9.Nxd4 exd4 10.Bxd4 ) a6 8.Qd2 (8... e5?! {Is still not playable.} 9.d5 {Forcing black back, since otherwise it loses a pawn.} Ncd4 10.Nxd4 exd4 11.Bxd4 ) Rb8 {Playing for counterplay on the queenside.} 9. Nc1 {Now that black has committed two moves to the queenside, white regroups, allowing `e5`.} e5 10.d5 Nd4 {The key idea, white can't take the pawn now.} 11.Nb3 (11.Bxd4 exd4 12.Qxd4 Nxe4 {Where black has a deadly attack.}) Nxb3
1. Attack on the black king.
If black is not careful, sometimes a direct attack can work. With 6... Nc6
white does not obtain much by playing 7.h4
, but against 6... a6
then 7.h4
hurts since the queenside knight is not able to fight for the d4
square.
[FEN ""]
1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.f3 O-O 6.Be3 Nc6 (6...a6?! 7.h4! {Immediately attacking the castled black king. Now the thematic `Nh5` does not work.} Nc6 (7... Nbd7 8.Qd2 c5 {Is a whole other can of worms, but I also do not think that black does better since it completely concedes the `d4` square.}) (7... Nh5 8.Nge2 f5 9.exf5 gxf5 10.Qd2 Nc6 11.O-O-O e5 12.Bg5 Qe8 13.Rg1 {Where the black king is definitely weaker.}) 8.Nge2 {Stopping `d5`.} b5 {Looking for counterplay on the queenside} (8... Rb8 {This is extremely dangerous because of `h5`, and white is faster.} 9.h5! Nxh5 10.g4 Nf6 11.Qd2 {Where it is hard to argue that black has a better version of a Sicilian Dragon.}) 9.cxb5 axb5 10.Nxb5 {Where black does not have as much counterplay while white has a passed `a` pawn.}) 7.h4 Nh5 {Thematic, preparing the break `f5` while still keeping `e5` available.} (7... e5 {This is also an option.} 8.d5 Nd4 9.Nge2 c5 10.dxc6 bxc6 11.Nxd4 exd4 12.Bxd4 c5 {With typical compensation for black; the white king is weak, `f5` is an option, and there is long term pressure on the queenside.}) 8.Nge2 f5! 9.Qd2 e5! 10.d5 Nd4 {Although the position is incredibly complicated, black has definitely made progress.}
2. Fight for the d4
square.
The move 6... a6
does not threaten e5
, giving white just enough time to get its light colored bishop into the action.
[FEN ""]
1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.f3 O-O 6.Be3 Nc6 (6...a6?! 7.Bd3! Nc6 (7...c5 {Continuing the play on the queenside.} 8.dxc5 dxc5 9.e5 {Followed by `f4`, and white is more developed, has more space, and should be better.} (8...Nc6 9.cxd6 Qxd6 {Simply loses a pawn.})) (7...Nfd7 {Although I cannot assess this position objectively, white is soon to be fully developed while black's queenside is completely stuck, and I do not think that black has accomplished anything.}) 8.Nge2 {Is a much better version of the main line.} ) 7.Bd3 e5! 8.d5 Nd4 9.Nge2 Nh5 {Threatening `f5` and `Qh4`.} 10.Qd2 c5 11.dxc6 bxc6 {Where black is threatening `d5`, crashing through.}
3. Transpositions favorable for white.
If white is not extremely accurate, it is very likely that they will stumble into the main line anyways. The pitfalls of the move order are only working against black, rarely against white.
[FEN ""]
1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.f3 O-O 6.Be3 Nc6 (6...a6?! 7.Nge2 (7.Qd2 {Is even more flexible for white.} Nc6 8.Rd1 (8.Nge2 {Is the main line.}) e5 9.d5 Ne7 (9...Nd4 10.Nge2 Nxe2 11.Bxe2 {Where white is faster on the queenside and black has no control of the center because of the battery on the `d` file.}) 10.g4 {Where black is incredibly cramped, and white has all the attacking chances.} ) Nc6 {Is still the best move, transposing into the main line.} (7... Nbd7 {Surrenders the `d4` square and abandons the queenside, allowing white to continue its expansion.} 8.a4 {Where white has everything they desire; initiative on the queenside, control of the center (with `d5` if necessary), attacking chances on the kingside with `h4` or `g4`, tons of space...} (8.h4 {Is also an interesting move since the knight on `f6`, but it will very likely transpose.} h5 9.a4) ) (7... c6 {Surrenders the `d4` square.}) (7... b5 {Continuing the play on the queenside.} 8.cxb5 axb5 9.Nxb5 {Loses a pawn.}) 8.Qd2 ) 7.Nge2 a6 8.Qd2 {Is the main line.}
Summary.
The opening principles are there for a reason, even in highly theoretical openings. The developing move 6...Nc6
plays a key role in the main ideas of the King's Indian, namely it fights for the d4
square. The move 6...a6
is a slow attempt to play on the queenside, and offers nothing towards the control of the center nor the defense of black's king.
6...Nc6
if white plays7.Qd2
then black can go for7...e5 8.d5 Nd4