Hi Chess Poster,

Does the Fritz Variation of the Two Knights’ Defense favor White? In a previous
post I was able to present four cooks of the Berliner Gambit. Here I expand on
my earlier analysis with the goal of demonstrating advantage White. Do you see
any flaws in my analysis?

1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 d5 5.exd5 Nd4 (The Fritz is a relative new
comer and seems odd since White can drive the Knight back with 6.c3 [Nc3=] (An
opening trap here is 6.d6? Qxd6 7.Nxf7 Qc6-+), when Black responds with the
shocking 6…b5! If Black plays 6…Nf5 then 7.O-O h6 8.Nf3 e4 9.Re1 Be7 10.Ne5 O-O
11.d4 exd3 12.Qxd3 +/=) 7.Bf1! This surprising retreat is best. The alternative:
4.Ng5 d5 5.exd5 Nd4 6.c3 b5 7.cxd4 bxc4 8.dxe5 Qxd5 9.O-O! [exf6 Qxg5 9.Qf3 Rb8
10.Qe3ch Qxe3 11.dxe3 gxf6! & Harding thinks the open lines give Black enough
for his busted pawn structure] Bb7 10.Nf3 Nd7 11.Nc3 Qd3 12.Re1 O-O-O with a
complicated position where Black has a decent game.

7.Bf1 Nxd5 (At this point, GM Lev Alburt and I analyzed the idea 8.Nh3 in order
to contest the f4 square, and, indeed, White gets a slight plus after 8.Nh3
Bxh3!? 9.cxd4! [gxh3?! Qh4! ICM van der Tak] Bd7 10.dxe5 & Black has some but
not full compensation for the pawn)

We now address the main alternative to the Berliner Gambit: 4.Ng5 d5 5.exd5 Nd4
6.c3 b5 7.Bf1 Nxd5 8.Ne4 Ne6!

It is time to clear up a misconception about development. Many players regard
development as moving pieces away from their original square. The Knights on e6
and d5 are worth far more than Knights on f6 and c6. In other words for the
Knights to occupy their suburb squares requires multiple moves. Each one is a
“developing” move. Even though White picks up a pawn and can avoid
simplification with 9.Bxb5ch Bd7 10.Bc4!?, Black’s great Knights’ yield
excellent compensation for the pawn.

10.Bc4 is the best way to steer for complications. For those who choose the
better known 10.Bxd7ch Qxd7, I’ve looked at these variations with GM Lev Alburt
for about an hour and the best that White might get is a miniscule plus with
minimal chances to convert the point. In the critical variation White has to
play d4 and wind up with an isolated d-pawn. Even though White may wind up with
an extra pawn on d4, Black just follows the Nimzowitch strategy: restrain,
blockade, destroy and the d-pawn falls followed by simplification and an equal
position e.g. 10.Bxd7ch Qxd7 11.d4 exd4 12.O-O Be7 13.cxd4 O-O 14.Nbc3 Rfd8
15.Be3 Nxc3 16.Nxc3 Nxd4 17.Re1 c5! =.

Here are four move orders that demonstrate the difficulties for both sides:
10.Bc4 Nb6 (Bc6?! 11.O-O Nb6 12.Bxe6 Bxe4 13.Bg4 +/=) 11.Bb3 Bc6 12.d3 Qd7
13.O-O O-O-O 14.Nbd2! Black has two basic choices: 1) to immediately regain the
pawn i.e. 14…Qxd3 15.Bxe6ch fxe6 +/= or 2) 14…Nf4 15.Nc4 Nxd3 16.Ng5 Nxc1
17.Qxd7ch Rxd7 18.Raxc1 Nxc4 19.Bxc4 +/-

The second variation features an early f5 by Black: 10.Bc4 Nb6 11.Bb3 f5 12.Ng3
Qf6 13.O-O O-O-O 14.d4 exd4 15.Nh5 Qh4 16.Re1 dxc3 17.Nxc3 Nd4 18.Be3 Bc6 19.g3
Qg4 20.Qxg4 fxg4 21.Bxd4 Rxd4 22.Rad1 +/=

Black can try for equality earlier with 8.Ne4 Ne6 9.Bxb5ch Bd7 10.Bc4 Bc6 11.O-O
Nf6?! 12.d3 Nxe4 13.dxe4 Qxd1 14.Rxd1 Bxe4 15.Be3 Rb8 16.Nd2 GM Alburt +/= These
are just three of the many variations in this line.

Here is an amusing miniature: 10.Bc4 Ndf4? 11.O-O Bc6 12.Re1 Nd3 13.Re3 Nef4?
14.Qf3! Nxc1? 15.Qxf4! exf4 16.Nf6 #

Normally I don’t like anti-positional concepts but a reader on line pointed out
that White might improve on the main line. In other words, if Black can equalize
in the main line with 8…Ne6, then players must try to find equality for Black
here also.

4.Ng5 d5 5.exd5 Nd4 6.c3 b5 7.Bf1 Nxd5 8.cxd4!? Qxg5 9.Bxb5ch Kd8 10.O-O Bb7
11.Qf3 exd4!? 12.d3 Qe5 13.Nd2 Bd6 14.g3 Rb8 15.Nc4 Nb4 16.Nxe5 Bxf3 17.Nxf7ch
Ke7 18.Bc4 Rhf8 19.Ng5 White retains his extra pawn & Black faces a difficult
endgame.

White seems to have a slight plus after 11.Qf3 Rb8 12.dxe5 Ne3 13.Qh3 Qxg2ch
14.Qxg2 Nxg2 (Bxg2 15.dxe3 +-) 15.d4 Be7 16.Be2 +/=

In a long mostly forced sequence (Long analysis, wrong analysis!?), Black can
regain the pawn but stands worse: 11.Qf3 Rb8 12.dxe5 Ne3 13.Qh3 Qxg2ch 14.Qxg2
Nxg2 15.d4 Nh4?! 16.Bg5ch Be7 17.Bxh4 Bxh4 18.Nc3 Bf3 19.b3 Rb6 20.Bd3 Rc6
21.Rac1 Bg5 22.Rc2 Bf4 23.Re1 Rh6 24.Be4 Bxh2ch 25.Kf1 Bxe4 26.Rxe4 f5 27.Re1 c6
28.Kg2 +/-

Although a final verdict of equal in the 8…Ne6 variation seems likely the
variations are complex and require considerable accuracy by both players.
However, Black still must demonstrate equality after 8.cxd4.

One last bit of analysis---the Ulvestad. While having minimal independent
significance, the Ulvestad may be a better way to reach the main line i.e. 4.Ng5
d5 5.exd5 b5 6.Bf1 Nd4 7.c3 is the main line of the Fritz, but Black can vary
with 6…h6?! 7.Nxf7 Kxf7 8.dxc6 Qd5 9.Nc3 Qxc6 10.Qf3! This takes all the fun out
of the opening. Black has nothing better than 10…Qxf3 (10…e4? 11.Bxb5! Qxb5
12.Qxf6ch gxf6 13.Nxb5 +/-) 11.gxf3 b4 12.Bc4ch Be6 13.Bxe6ch Kxe6 14.Ne4 +/=