It seems relevant to all of this that 3...Nf6 is seen quite often at high levels, but that 8.Qf3, while it is seen, isn't seen nearly as often.  Personally I think that Coleman's 8...Rb8 is the best way for Black to go, but I'm not about to debate it here.

S*@t, I once again confused the modify button with the quote button. Never mind, this thread is a mess anyway.

sloughter wrote on 02/10/09 at 12:43:27:


I will only believe that these evaluations are Alburt's when you have given a written source, like a magazine or a book. Instead of 18.Bd2 18.Bf4 might be stronger; critical is the idea Rb8 combined with Bf8-c5-d4. In general if Black gets one or even two pawns for the minor exchange he is OK.
But I think Black can improve earlier. If Black's Knight is awkardly placed after 5...Na5, then what must we say about White's Knight after 8.Nh3 ? Even according the way Sloughter understands development this one does nothing but denying square f4.
8.Nh3 Ne6 (Bg4 is ridiculous -10.Bxb5+ Nxb5 11.Qxg4 +-; must we really believe GM Alburt missed this?) 9.Bxb5+ Bd7 10.Qa4 and exactly because of that knight of h3 Black can afford Bxb5 11.Qxb5+ Qd7 12.Qxd7+ Kxd7. This is not an endgame yet, but a Queenless middle game. Black has superior activity, controls the centre and has a nice target in square (pawn) d3. Verdict: sufficient compensation.

Still I would like to know: what was Sloughter's lates (non-)refutation after 8.Ne4 Ne6 again? I kinda lost track.

This is one tricky new variation in the 8.Qf3 line: 

8.Qf3 Qd5?! 9.Be2! (Qxd5? 9.Nxd5 with comp.) 9...e4 10.Qg3 h6 11.Nh3 Nb7 12.Nc3 Qd4 13.d3 Bd6 14.Qe3 Qe5 15.dxe4 Bxh3 16.Qxh3 Bb4 17.Bd2 Rd8 18.O-O-O +-

8.Qf3 h6 9.Ne4 Nd5 10.Be2 Be6 11.d3 Nb4 12.Bd1 c5 13.O-O Nxa2? 14.c4! Nxc1 15.Ba4ch Bd7 16.Bxd7ch Kxd7 17.Qxf7ch Be7 18.Nxc5ch Kc6 19.Rxa5 Ne2ch 20.Kh1 Bxc5 21.Qf3ch Kd6 22.Qd5ch Ke7 23.Qxe5ch Kf6 24.Rxc5 +-

or 10...Be7 11.d3 O-O 12.Qg3 Nf4 13.Nec3 Qd4 14.O-O Nb7 15.Re1 Nd6 16.Bf1 Re8 17.Qf3 Be6 18.Nd2 Nd5 19.Nce4 Nf5 20.c3 Qb6 21.Nc4 +/-sloughter wrote on 02/15/09 at 20:28:47:

Black faces the same problem here as many other variations i.e. Black has some, but not full compensation for the pawn e.g.

4.Ng5 d5 5.exd5 Na5 6.Bb5ch Bd7 7.Qe2 Be7 8.Nc3 O-O 9.O-O & Black has to untrack his Queenside, so he has nothing better than 9...Bxb5 10.Qxb5 c6 11.dxc6 Nxc6 12.Qe2 Nd4 13.Qd1 Nd5 14.Nxd5 Bxg5 15.Ne3 Re8 16.d3 & Black has some but not full compensation for the pawn. Clearly there are a lot of other variations, but the problem facing Black is the traffic jam on the Queenside which more or less compels an exchange on b5, transposing to a variation favorable to White +/=Ametanoitos wrote on 02/15/09 at 19:25:59:

It is highly unlikely that the outcome of these position is dynamic balance. In many critical variations, just one move means the difference between winning and losing and I don't mean crass blunders. In one variation, for example, if Black can capture a piece on d3 doubling the White d-pawns, Black is better. If White can keep his extra pawn without allowing his pawn structure to be compromised, then White is better.

The TKD, is perhaps the last major frontier in chess where modern defensive technique has not caught up in theory or in practice. Black must demonstrate he has enough play for the gambit pawn. Just think about it: In the Ruy Lopez Exchange Variation, White is willing to give up the Bishop pair just to get the advantage of pawn structure. 

In the TKD, Na5 variation, White always gets a Queenside pawn majority and usually an extra pawn in both the 8.Be2 & 8.Qf3 variations. Why is that the proponents of the TKD, just ignore the fact White that routinely gets a Queenside pawn majority in the opening? Somehow, we are led to believe that Black has just enough play to regain the pawn, yet not get the initiative. GM Lev Alburt validated my impression of a critical ending arising from the Na5/Bc5 main line of the 8.Be2 variation. White succeeds in chopping wood but is forced to give up his extra pawn to relieve Black's piece pressure. Just like that, we wind up in an endgame where only White has winning chances based on this superior pawn structure and Queenside pawn majority. I predict that thousands of games will be played in the future where Black's fate depends on his being able to defend a poorer endgame.

Even in one of the better variations for Black, 8.Qf3 Be7, the fundamental problem facing Black long term, is that he cannot afford to enter an endgame without making major progress e.g. suppose in a King and pawn endgame we "give Black" White's c- & d-pawns which Black wins with piece pressure; unfortunately, he loses the initiative in the bargain. Put the White King on d3 and the Black King on d6 with White to move. White is much better because he has the Queenside pawn majority which is desired by most players

What is not fully appreciated is that most Na5-TKD's in the future are going to revolve around the fact that White has the better endgame prospects based on his Queenside pawn majority, and his fewer pawn islands. The point being---Black must make progress in the middlegame; he cannot just afford to swap piece for piece, so White has a crystal clear long term plan---contest the center ASAP with the idea of chopping wood at every opportunity, PROVIDED, that White not lose time to do so. 

For example, if Black opposes Queens in the middlegame as Fritz 8 does routinely against me in the 8.Qf3 variation (Isn't it surprising that Fritz 8 regards my active Queen as enough of a threat that it routinely opposes Queens just to force simplification into an inferior middlegame?), it does so with the idea that if I capture the Queen say on d5, the computer will simply capture with a piece with gain of time. What I do in response is to anchor my Queen so that if Fritz grabs it, Black's Queen can be captured advantageously with either a pawn or a piece.

Once I get done tying up the loose ends in the Wilkes Barre, I will begin to tackle the thicket of variations in the Na5 TKD, and, as a "safety" measure, provide a clear pathway to a White advantage in both the 8.Qf3 & 8.Be2 variations.SWJediknight wrote on 02/14/09 at 21:58:30:

While some might argue that the positions are dynamically balanced, why is it they are not balanced in most of the candidate variations? I don't see anyone claiming that Black can equalize in the Berliner Variation. Then I was told that my failure to consider the sequence 8.Ne4 Ne6 was a fatal flaw. True, Black has good chances to equalize, but when my analysis was capable of being challenged, it was so vigorously. Then when I posted my lastest move order indicating that White could achieve a pull in this variation---silence. 

Then I published the Alburt Variation 8.Nh3 whose analysis and evaluation was done by three time U.S. Open Champion, GM Lev Alburt. Nobody has disputed either the analysis or the evaluations.

Another candidate move, 5...Nxd5 is clearly favorable to White.

The Wilkes Barre/Traxler also appears to favor White and will be the subject of a continuing article in that opening.

Can anyone prove equality Black in the variation 5.exd5 b5 6.Bf1 h6 7.Nxf7?

If the positions are so obviously equal, then why am I able to consistently gain the upper hand against a 2800 level computer in the 8.Qf3 variations?

Black must hang his hat on the 5...Na5 variation. Objectively, this should be the weakest choice. The Knight is on the rim, and it takes it two tempos to get to c6 or c5. Where are those attacking tempos needed to justify a pawn sacrifice? It is generally accepted that in order for the attack to be justified, the attacking side is supposed to have three tempos for the pawn in an open position. Because this is not an open position, Black should need even more time to justify the pawn. That is why White can win in the Steinitz Variation even though down four or five tempos.

For those post members with an open mind, I will post dozens of subvariations in the 5...Na5 lines where White can achieve a plus. It is up to the post members at that point to accept, reject, modify or ignore this analysis. I think the observation that, "Sloughter isn't getting anywhere with his analysis", ignores the obvious: I suppose you mean by not getting anywhere, refuting most of the candidate moves.SWJediknight wrote on 02/14/09 at 14:38:27:

You overlook one minor detail. No one has been able to get beyond the Marshall Gambit i.e. demonstrate a clear advantage White. Grown ups usually engage in civility, something most notable by its absence here. Your's is a typical post i.e. insult rather than edify. Apparently you are unaware that the "children" who have played and studied the Two Knights' Defense from White's standpoint include World Champions Fischer, Karpov, Kasparov, Berliner and Estrin. Apparently to you, these World Champions are naive little children.Uruk wrote on 02/14/09 at 10:56:03:

Black scores well in NM Eric Schiller's data base when White castles Kingside. What is clear is that White must walk a tightrope to an advantage. What is needed is a simple move order that leads to a clear advantage without allowing counterplay.

Some of the complications are evident when Zaitsev maintained in Informant 36 that White could improve on Karpov-Beliavsky with 6.Bd5 d6 7.c3 Qe8 8.d4 exd4 9.Bxc6 Qxc6 10.cxd4. At this point I departed from Zaitsev's analysis in an article I wrote called Early Innovations 1.e4. He gives 10...Bb4ch whereas I gave 10...Bxd4. If now 11.O-O Qb6 12.Nc3 Bxc3 "when Black may survive" 

IM John Donaldson, Inside Chess v.4, Issue 13, p. 19, Death Knell for the Wilkes Barre? extended the analysis and claimed that after 13.bxc3 Bg4 14.Qd3 h6 15.e5! hxg5 16.exf6ch gxf6 17.Re1ch Kd7 "promises the second player plenty of suffering" . But is it so clear after 14.Qd3 Nd7 instead of 14.h6?

William's claim that 6...Rf8 is not good for Black is a good indication that Black is busted. Black must put immediate pressure on f2 to have any hope of an advantage. But we see a good illustration of this concept in action with Losev-Isaev, Trades Union, 1990 6.Bd5 Rf8 7.Bxc6! dxc6 8.Qe2? . Now Black missed the simple drawing variation 8...Ng4 9.f3 Bf2ch! (not Nf2) 10.Kf1 Bb6 when the only way for White to make progress is to play 11.Nh3/Nc3/Nd1/Nhf2. Even though the author got a big advantage with this move order against Fritz 8, it is equal.

This is why I recommended 8.Nf3 instead.

One side variation addressed by Williams is the attempt by Black to avoid 7.Bxc6 with 6.Bd5 Nb4 7.d4! (Estrin) exd4 8.O-O Nbxd5 9.exd5 Qg8 "Estrin claims to refute 6...Nb4...9.Qg8, the refutation appears doubtful. " But what about the simple 10.c4! and White's advantage is obvious?

Interested readers should consult Cramer's Traxler-Gegenangriff Fritz-Variante Ulvestad-Variante, p. 44 for move orders that appear to refute the concept of 6...Nb4. "So theoreticians Estrin, Cramer and Moody all believe that White is clearly better after 6...Nb4 7.d4!"

The move order that Black should avoid is Anand-Beliavsky, Linares, 1991: 6.Bd5 Qe8 7.d3 (Bxc6?) d6 8.Bxc6 bxc6 9.Be3 Qg6 10.Nf3 Bxe3 11.fxe3 Qxg2 12.Rg1 Qh3 13.Rxg7ch +/-.


An idea of Fine according to Estrin's The Two Knights' Defence is to try 6.Bb3 Rf8 7.O-O Qe8 8.c3 Qg6 9.d4. "and the Black King is exposed." This variation is one of the critical ones because it looks like White gets away with castling Kingside without allowing counterplay. About the only try for equality is 6.Bb3 Rf8 7.O-O d6. It gets complicated after 8.d3 Bg4 9.Qd2 h6 10.h3 hxg5 11.hxg4 Qc8 unclear to +/= The next post will deal with the proposed new main line of the Wilkes Barre, Bd5/Bxc6/Nf3.

quote author=richard_moody_jr. link=1230634273/240#250 date=1234566618]Agreed---the point of this post is to test the hypothesis whether White is winning as of 3...Nf6. 

The next attempt is to determine whether White has a clear advantage in the Wilkes Barre. So far, it looks like the Alburt Variation 8.Nh3 gives White good chances to achieve a plus in the Fritz. It should also be pointed out that no one has been able to refute the latest move order I gave in the 8.Ne4 Ne6 variation.

The independent lines in the Ulvestad don't seem to promise equality i.e. 4.Ng5 d5 5.exd5 b5 6.Bf1 h6 7.Nxf7 Kxf7 8.dxc6 Bc5 9.Be2 +/= or 8...Qd5 9.Qf3 +/= The Lolli favors White, so I will attempt to show that the Wilkes Barre/Traxler leads to a simple technical advantage. In the hands of say a Korchnoi easily +/-.

4.Ng5 Bxc5 5.Bxe7ch Ke7 6.Bd5 Rf8 (or d6 or Qe8; Beliavsky was allowed this move order in Karpov-Beliavsky & then Karpov followed with Bxc6 which allowed Qxc6; Karpov gave away three tempos i.e. he made four Bishop moves just to exchange it off and got only one useless move from Black, 5...Ke7. No wonder Black got a strong attack!) 7.Bxc6 dxc6 (bxc6 will be dealt with separately)

7...Bxf2ch? 8.Kxf2 Nxe4ch 9.Ke1+-

7...dxc6 8.Nf3 Nxe4 9.Qe2 Nxf2 10.Rf1 Rf5 11.d3+-. To be continuedmicawber wrote on 02/13/09 at 22:20:02:

[/quote]