Markovich wrote on 03/18/10 at 19:48:19:

After 1.e4 c6 2.Nc3 d5 3.Qf3 dxe4 4.Nxe4 Nd7, White wants to play 5.d4 but here 5...Ndf6 is a double attack on d4 & e4, e.g. 6.c3 Nxe4 7.Qxe4 Nf6 with fine development for Black. This could be the reason?

I wonder if anyone thinks that the gambit form of the Two Knights promising: 1.e4 c6 2.Nc3 d5 3.Nf3 Bg4 4.h3 Bxf3 5.Qxf3 e6 6.d4 dxe4 7.Nxd4 Qxd4 8.Bd3 and now for example, 8...Nf6 9.Be3!? Qxb7 10.0-0.  Two pawns is perhaps a lot to give up, but White has what appears to have a lot of comp.

A related offbeat line is 3.Qf3, also with plans of sacrificing the d-pawn.  I have less confidence in it, though.  As I recall, the antidote is supposed to be ...dxe4 followed by ...Nd7, ...Ndf6.  Why, I forget.

Conquistador wrote on 02/23/10 at 05:31:48:


The Rasa-Studier Gambit arises after 1...e6 iso 1...c6. This Anti-Caro Kann gambit is invented and played by Von Hennig in the 20's of the last century, long before Rasa and Studier showed up. It is one of the oldest members of the BDG-family. But after 4...exf3 5.Nxf3 Nf6 (an important position of the BDG) I doubt if White has any better than 6.Bc4, after which you must rely on Bücker (see his Chesscafe column) if you want to prove compensation.
Still 4.Bc4 might be more precise than 4.f3 - and 3.f3 might even be a better version.

TN wrote on 02/23/10 at 06:56:28:


You might want to take a look at 5.Bg5 in this line. The rather outrageous idea is to meet 5...Qb6 with 6.Bd3 Bxd3 7.Qxd3 Qxb2 8.e6! 

It looks very unlikely but I have a big plus score with this in online blitz (about +200 rating performance). The main point is the queen often gets trapped on a1 and ends up being exchanged for the second white rook on f1 after white castles. The resulting material balance might even favor black were it not for the fact that his k-side is going nowhere anytime soon giving white a massive temporary material advantage. Add to that the black queenside is shaky (b7 is hard to defend without letting the Queen in) and the black king is potentially vulnerable and you have a very effective practical system.

I don't doubt that with perfect defence, black may be OK but he has significant short-term problems.  In addition, there is the useful psychological factor that this isn't the kind of position black was expecting when he played 1...c6.

There is an article by the indefatigable Michael Goeller (urusov) at http://www.kenilworthchessclub.org/games/java/2007/caro-adv-h4.htm

Thanks for the link Lou_Cyber!

The handle of Michael Goeller is urusov. I used to play this line and gave it up a couple of years ago, but I just might give it a whirl again...

I had a quick look through the PGN file, and interestingly Goeller recommends the 4...Nf6 5.d4 g6 line but claims that White can secure an advantage with precise play, whereas after 4...Nf6 5.d4 Nc6 6.Bb5 Bd7 he feels that Black should equalise with 7.Nd7 Qd7 8.c3 a6 9.Bd3 e5!, which is one reason I quit playing this variation.

Conquistador wrote on 02/23/10 at 05:31:48:

4.Bc4 Nf6 5.Qe2 Bg4 6.Qe3 Qb6 and White must accept a queen trade or lose time with 7.Qc3 Nd5. That said, 7.Nc3 Qxe3+ 8.fxe3 e6 9.0-0 doesn't look too bad for White, as Black's extra pawn is doubled and White has a nice half-open f-file plus lead in development.

After 5.Nc3 Qd4 6.Qe2 Bg4, White's best is probably 7.f3 (I suspect that this would be a much stronger gambit if the knight was still at g1!) 7...exf3 8.gxf3 Bxf3 9.Bxf7+ Kxf7 10.Qxf3, and White may have enough compensation.