New Subvariation Two Knights' Defense

The following idea has not been given on the 8.Qf3 variation of the Two Knights' Defense until today:

4.Ng5 d5 5.exd5 Na5 6.Bb5ch c6 7.dxc6 bxc6 8.Qf3 Be7 9.b4! Bxb4 10.Bxc6ch Nxc6 11.Qxc6ch Bd7 12.Qc4 Qe7 13.c3 Bd6 14.Ne4 Nxe4 15.Qxe4  White has a protected passed pawn; Black the Bishops

I don't think your novelty can be played more than once.

Interesting idea and a bit tricky in some lines but I dont trust the position

Bg4 isnt a total refutation but there seems to be a better move with h6 (i included the games) so not sure if your idea will pan out really but interesting

Good analysis (which engine?) but in your main line, I'd prefer 14...Bxg5 15.fxg4 Nf4 16.O-O Qd4+ 17.Kh1 h5!, and Critter agrees.

8.Qf3 Be7 9.b4?! Bg4! =/+ to -/+ But here is a new try in the main variation: 8.Qf3 Be7 9.Bxc6ch Nxc6 10.Qxc6ch Bd7 11.Qc4 O-O 12.b4! Now the White Queen can get in back of the pawns. My chess computer program is down so I can't crunch the variations, but it looks like 12...Rc8 13.Qb3 h6 14.Nh3 Be6 15.Qb2 is playable. White meet most of the obvious threats like Qc7 16.Nc3  or Qb6 16.a3

Minor correction of the above post. White should meet 16...Qb6 with 17.c3

Rb8 seems to be popular off and on but h6 seems to shut everything down. the two GM vs GM games I found both end up in fairly lopsided wins some of the Rb8 lines were not so clear. but this is b ased on my database not deep research

12...Rc8 13.Qb3 Nh5 14.h4 looks perfectly playable.

I used Spassky world champion version 1.0

Its an older engine but  a pretty good one even after all these years

8.Qf3 Be7 9.Bxc6ch Nxc6 10.Qxc6ch Bd7 11.Qc4 O-O 12.b4 Rc8 13.Qb3 Nh5 14.h4 Nf4 15.g3 Ng2ch 16.Kf1 Bc6 17.Rh2 +/-

8.Qf3 Be7 9.Bxc6ch Nxc6 10.Qxc6ch Bd7 11.Qc4 O-O 12.b4 Rc8 13.Qb3 h6 14.Nf3 Be6 15.Qb2 e4 16.Nd4 +/-

As for the tries 8...h6 and 8...Rb8, these are covered in another thread; if you like I will post the games where at tournament level against Fritz 12, I got an advantage. If Fritz 12 can't find any tactical shots that work, clearly White has good prospects for a slight plus against either 8...Rb8 or 8...h6.

White made a huge blunder in the h6 variation above. This may come from an old Estrin game where the correspondence World Champion gave up the Bishop for no reason.

8.Qf3 h6 9.Ne4 Nd5 10.Ba4 (To hold c2 & close the b-file with Bb3 at some point) Ba6 (there are other choices) 11.Nec3 +/- or

10.Ba4 Rb8 11.Nec3 Rb4 12.O-O Bg4 13.Qg3 (White threatens f3/Qe1) +/-

13...Ng4 is simply winning so 12.b4? is correct.

12.b4 was a "desperate" attempt to find anything to give White some hope of achieving a plus. In the age of computers 99% of "revolutionary" ideas are likely to be wrong. This is one of the 99%. This just reinforces the general consensus that 8.Qf3 Be7 is equal. Against 8...h6 and 8...Rb8, I've been able to get a plus against Fritz 12 at tournament level, but could only equalize with White against 8...Be7 with 9.Bd3. While no proof of the soundness of the games, clearly if Fritz 12 at several minutes/move can't find equality, the average player should do well.

8.Qf3 Be7 9.Bxc6ch gives Black better practical chances, but White may be able to equalize.

f you would like to see the thoughts of a correspondence master on this subject, see http://www.ebay.com/itm/Two-Knights-A-Chess-Killing-Machine-with-an-Emphasis-on-4-Ng5-/271319251507?ssPageName=ADME:L:LCA:US:1123