I am convinced with your analysis that White gains nothing with 8.Ne4 because the Two Knights' Tango all are developing moves as I define development. No wonder 8...Ne6 works so well! However, I am going to check with GM Alburt about 7.cxd4 bxc4 8.dxe5 & get his take on the positions.

Amazingly enough, in the Steinitz variation, 4.Ng5 d5 5.exd5 Na5 6.Bb5ch c6 7.dxc6 bxc6 8.Be2 h6 9.Nh3?! Bd6 10.d3 Qc7 11.Ng1! O-O, Black has nothing unless someone can find a way for Black to get in e4 with advantage. 

White has a Universal Position, the most difficult kind of position to attack. According to Kasparov (If you want a textbook case of what a Universal position is, just check out the World Championship games between Kasparov and Short with Garry on the Black side of the Sicilian.) it is necessary to advance your own pawns to break up a pristine pawn position.

The King's Indian Saemisch is probably favorable to White, because it is, to the best of my knowledge, the only major opening by White where White stays in Universal mode in the opening. Has anyone tried the following strategy c4/d4/Nc3/e4/f3/g3/Bg2/Nh3/Nf2 & then just push the Queenside pawns, rearrange your pieces behind the pawn wall and then break through? 

In the Steinitz variation, a huge lead in development dissipates by pushing the Kingside pawns and permits White to catch up in development and push the Queenside pawns. Qb8 is an attempt to stop that strategy. The main problem is that White can play Nbd2 & Nc4 or Ne4 & the Black Bishop must move off the d6 square (Bxh2 is never a threat as long as White hasn't castled; all it does is "develop" White's King Rook by giving it an open file, and, even worse, stops the Kingside pawn roller). This then permits the King to step out of the center with Kf1 or O-O-O)

Correct against the Steinitz is 9...Bd6 10.d3 Bd7! (To keep the e6 square for the Knight on a5) 11.Ng1 Qb8!! so that a Knight hit on d6 can be met with Bc7. More importantly, Black forces White to deal with the hit on the b-pawn. If Black plays b3, then he has to be concerned about ideas like a5/a4. However, Nbd2/Nb3 is bad because it slows down the pawn roller on the Queenside, so Black wins on the Kingside.
The Rook isn't trapped because it has a Rook lift on the seventh or sixth.

In the Steininz, White only wins when Black rushes in; the correct way to beat White is to stay in Universal mode on the Kingside (i.e. all of your pieces are behind your pawn wall, push them forward in a coordinated manner until you win something big) and Classical mode on the Queenside (active piece play), deny White space, and then use Karpov's boa constrictor style.



MNb wrote on 01/10/09 at 19:50:58:

"Logically" your variation makes sense, but it is antipositional to me. I can't calculate very well in antipositional positions since I rely solely on intuition. When I play computers, I don't calculate; I only place the pieces and pawns where they belong. It would be impossible for me to analyze this variation. The silicone monster in antipositional variations can calculate 7-8 moves ahead. With luck, I can only look two or three moves ahead. When I played the computer in a variation that IM Silman analyzed as favorable to White, I lost 80 games in a row to Fritz just because it was antipositional. I'd trust the 'beast's' wisdom on this one. OTB it's a nightmare. In postal, with no takebacks of moves, it's a minefield. As a dry theoretical exercise, have Fritz 11 or another strong computer look at it. 

I'll give it a try, and I do thank you for asking. If I can convert the position into something I recognize as positional, I'll be a lot happier than just crunching "moves".luskifuski wrote on 01/15/09 at 17:32:46:

what was your conclusion on

1. e4 e5 2. Nf3 Nc6 3. Bc4 Nf6 4. Ng5 d5 5. exd5 b5 (
5... Nd4 6. c3 b5) 6. Bf1 6... Nd4 7. c3 Nxd5 8. cxd4 Qxg5 9. Bxb5+ Kd8 10. O-O Bb7 11. Qf3