ErictheRed wrote on 01/18/14 at 15:24:05:
I agree that 10...Nc6 looks better. But I don't think I would rush with the d5-d6 push as White, anyway: maybe just 10.Be2 to complete development and trusting in the compensation. If 10...Nd7 11.d6 should gain in strength, and if some other move I'll probably go 11.Ne5 intending Bf3 and Rhe1, because you no longer have the annoying ...Nh5 response (for instance, 10.Ne5 Nh5!).
I mean we could go back and forth like this all day, but White has compensation here. Is he better? Probably not. Is the extra tempo useful? Very much so.
Well, yes of course we should go back and forth like this. As we should be able to define the compensation somehow beyond mystic words, Or are we just romatically diemeresque?
Please remember that I referred to Scheerer's book, which is the most up-to-date source by now, and there, besides the bright words about the possible future of this variation, the standard of research to the topic is quite low, in my eyes.
As said he gave 10.d6"!"
To quote myself: "I don't think that white should push d5-d6 too early."
To quote you: "I don't think I would rush with the d5-d6 push as White."
So we both agree on this.
And my finding on 10... Nc6 etc as given is in clear contradiction to the book, where white's play rises to unjustified power as 13... Ne4! is not topic there.
So there is a must to go back and forth to find the thin line to best white play.
Btw 1:
Stefan Bücker gave "10.Bh6!" saying that the tempo plus should result in some benefit in those sharp attacking schemes. But one move is no analysis.
The question remains how white should get things going on the king's side as long as the Bc8 ranges down to g4. And I'm not sure that black is willing to obstruct it deliberately with Nbd7, as he has b7-b5 asf at hand - and that is a result of white's early castling long: giving targets to go for.
Exactly there begins the discussion of the virtue of being a tempo up. Just counting tempi does not meet the point. No side is forced to just follow common ways (well trodden paths) but can adept the play to the exact situation and information.
There is the german off beat-theoretican/nerd Rainer Schlenker who invented the term "reflex theorem" (see Keilhack "Die Tarrasch-Verteidigung"). He says that chess is a game of information and having recived a certain information will influence your play, your reflex on the very situation. In the very case: Having white's king castled long and seeing an attack coming on the king's side, black will surley refuse to play sth like Nbd7 but head for aggressive measures on his own against the given target.
I confer to another topic to make my argument clear: Just take the classical king's indian (E97etc.) and you know about the typical attacks of black's on the king's side.
And now take the exact variation in the king's indian attack. There the classical approach by black (c5/d5/e5, Nc6, Nf6, Be7) will naturally result in positions where black is a tempo down to the normal KID. And, given that black plays it best way, we can define this missing tempo on the black side: He misses 0-0.
And that completely changes the whole picture as now any avalanche of white's on the king's side (as is usual in the highly theoretical colours reversed version) will have no target there.
In other words: just counting tempi and forgeting about given informations will fail.
On your 10.Be2 (I had that too and I'm not sure about it), black can go for 10...b5 with b5-b4 in mind as the B occupies a possible retreat for the Nc3 (but there is a4 available, but then black may think about Ne4... I havn't done much on it).
Btw 2:
The idea of snatching the c5 pawn in the variation 10... exd6 11.Bxd6 Re8 was not mine. It was given in a somehow obscure way by Scheerer, and there was Bücker in Kaissiber who pointed out the fact, that white may catch the pawn c5 as there is no protecting N on d7 yet.
But that is of lesser importance. After all as black I wouldn't go for the cooperative 10... exd6 as 10...Nc6 is more to the point.