21.Nxe6 Qxe6 22.gxf5 Nf4+ 23.Kd2! does probably not save White, for example:

25. ... Rf2+ 26 KxN g5! 27 hg Bg6+ 28. Kc3 Rc2+ 29. Kb3 h5 as Martin pointed out and ... now here are a few sample lines from Fritz9


1. -+ (-1.94): 30.a4 Rac8 31.a5 R2c3+ 32.Kb2 Rxe3 33.Rg2 h4 34.e7 h3 35.Rh2 Rg3 36.Ka2 a6 
2. -+ (-1.94): 30.e4 Bxe4 31.e7 Rcc8 32.Rd7 Rab8+ 33.Ka2 h4 34.Rc1 Kf7 35.g6+ Ke8 36.Rxc8+ Rxc8 37.Rxa7 Bxg6 38.Ra6 
3. -+ (-2.31): 30.Rf1 a5 31.e4 Bxe4 32.Rff7 a4+ 33.Kb4 h4 34.Rg7+ Kh8 35.Rgd7 Bg6 36.e7 Rc4+ 37.Kb5 

Instead of 28. Kc3, White will  - by comparison - be better off by playing 28.e4 Be4+ 29. Ke3 Rf3+ 30. Kd2/Ke2, but even than Black is better

My personal preference is to follow Martin's original suggestion, which is to avoid 0-0 by White, by playing 12. ... cd4 13. cd4 Qc3+ 14. Ke2 Nf5, because I am ok with transposing back into Anand - Lputian with 15. g4, based on the improvement 20. ... f5. So I  guess the way to go for white is probably 15. Qg4 (instead of 15. g4) avoiding the transposition.

If this doesn't work out too well for White either, than I think White should look into alternatives for 12. h4, in particular 12. g4

This is a response to Craig's question why it isn't possible to play 12. ... f5 instead of Bd7 or Nf5.

I think in NIC 71  (or was it 70?) there has been a letter in the forum section dealing with 12. ...f5 . I am at work right now and can't check the volumne # and page #, but I will once I get home.   

Anyways, the analysis was interesting and I think it is worth looking into it.

Good point to avoid the oo-line, but I don't see how to avoid going back to Anand-Lputian game

12...cxd4!? 13.cxd4 Qc3+ 14.Ke2 Nf5 

and now if 15.Qg4 then Black can play 15..f5 or 15...Ne7 with an easy (?) game

and if 15.g4 Nxe3 16.fxe3 we are now in the Anand-Lputian game

Ok HgMan and all the rest of us "french fanatics", after extensive home analysis I think I have found a decent solution to blacks problems in the hertneck variation:

1. e4 e6
2. d4 d5
3. Nc3 Bb4
4. e5 c5
5. a3 Bc3
6. bc Ne7
7. Qg4 0-0
8. Bd3 Nbc6
9. Qh5 Ng6
10. Nf3 Qc7
11. Be3 c4
12. Bg6 fg
13. Qg4 Bd7
14. h4 Rf5
15. h5 gh
16. Rh5 Raf8! Watson says that this is 'dicey' but i disagree

This seems solid enough and Nh4 ideas stopping Be8-g6 ideas seem to be met by Rf2. Does anyone have any white responses that lead to any advantage?

It appears that 14 f4 loses to NF5!! and Fritz gives -2 for white, so the question remains

1. e4 e6 2. d4 d5 3. Nc3 Bb4 4. e5 c5 5. a3 Bxc3+ 6.bxc3 Ne7 7. Qg4 O-O 8. Bd3 Nbc6 9. Qh5 Ng6 10. Nf3 Qc7 { Hertneck's line } 11. Be3 Nce7 { last idea to improve the line } 12. h4 {12.g4!? a flexible move if White wants to delay oo and h4 }

Can someone tell me why F5! is not played here?????? Thank you

I hope forum members won't mind my reviving a series of older postings from the message board.  This game strikes me as very interesting and I should note that another game played 12 ... Nf5 to a draw in August.

Your analysis is fascinating.  20 ... f5! looks like a strong improvement on the game.  Just a quick question (if you can recall your good analysis here from several months ago): You suggest 21 exf6 Rxf6 and Black is better.  I agree, but is there even a trick here: 21 exf6? Nf4+ 22 exf4 Rxf6 (not 22 ... Bxh5?? 23 Rxg7+)?  The queen is trapped, and I don't see any mating threats with Qxh6 or Bh7+ and Nxe6 can be met by Qxe6+ (the check is the killer), though I'm sure I may be missing something.  White can salvage the queen with Bh7+, but I don't see the mating attack.  Your line may indeed be simpler.

Of course, 21 Nxe6 is the critical line and I wonder if Black can play for more?  21 ... Qxe6 22 gxf5 Nf4+, for example.  If 23 exf4 Bxh5+.  If 23 Kd2 Nxd3 24 fxe6 Bxh5 25 Kxd3 Bg6+ (which also has the advantage of protecting g7 from Rg1) with (probably) Rfe8 to follow--I'm not sure I like the idea of the roaming Rf2.  White has four pawns for the piece, but given that three of them are on the e-file, I'm not sure White's necessarily better.  Just some general musing, though I'm not sure it's an improvement over your own suggested line.  It just seems as though there must be some way for Black to exploit the White king on e2.  If nothing else, thank you for a pleasant evening of lazy analysis!!

I've been working on my French and trying to "grow up."  I've often resorted to 4 ... b6 lines in the Winawer, chiefly as a means of avoiding theory, but would like strengthen my repertoire here.  Can any forum members recommend a favored line--either with 7 Qg4 0-0 or 7 Qg4 Qc7?