1.e4 e6 2.c4 d5 3.cxd5 exd5 4.Qb3 and now, instead of 4...dxe4, let's see what happens if Black tries 4...Nf6. My responses to this move changed over time, and we'll take a look at them in more or less the order in which I tried them.

(A) 5.e5
This isn't a bad idea; it's just that White's (now) poor queen placement and lack of development render this a blank shot. After 5...Ne4 6.Nf3 (6.d4 Nc6 [6...c5 is also good for Black] 7.Nf3 Bb4+ and Black is already at least equal) 6...c6 7.d4 Qb6 the position is balanced and, at least equally important for the fan of this line, relatively dull.
(B) 5.d3
This option has two things going for it. First, it keeps White's pawn structure more or less intact; second, it comes with a cheapo: if 5...dxe4 6.dxe4 Nxe4??, White wins the knight with 7.Qa4+ and 8.Qxe4+. Every once in a while the cheapo works, but if Black avoids it and focuses instead on rapid development and control of the d4 square, then White's position is inferior. Thus 5...dxe4 6.dxe4 Bc5! 7.Nf3 O-O 8.Bd3 Be6! (8...Nc6 is good too) 9.Qc3 (9.Qxb7? Qxd3 10.Qxa8 Nxe4 wins due to the double threat of 11...Bxf2# and the queen-trapping 11...Bd5) Nc6! gives Black the advantage.
(C) 5.Nc3 d4 6.e5
First of all, note that 5...dxe4 6.Bc4! justifies the delay by transposing back into the sort of position examined in the initial post. Black can also play to hold the d5 point with 5...c6; that will probably be the subject of yet another post.
Sticking to 5...d4 lines then, you might remember that I presented a game in this line in the original post, which continued 6.e5 dxc3 7.exf6 cxd2+ 8.Bxd2 gxf6?!, when White won quickly. After the obvious and natural 8...Qxf6, however, Black is in good shape.
(D) 5.Nc3 d4 6.Bc4!!

This very nice idea is one of my best finds in this variation, and I'm pleased to say I found it in a blitz game rather than by asking my software for its opinion. (In case you're wondering, Don Fritz and the rest of the family have since given the move their blessing.) Of course, had Black played 5...dxe4, 6.Bc4 would be the obvious rejoinder - but even here, at the potential cost of a piece, it still works out well! After 5...d4 6.Bc4!! dxc3 7.Bxf7+ Ke7, 8.e5 leaves White with more than enough compensation.
Black should probably return the knight with 8...Nc6 9.bxc3 Qd3 10.Ba3+ Kd8 11.exf6 gxf6 12.Bxf8 Rxf8 13.Qd5+ Qxd5 14.Bxd5, when White is essentially a clean pawn ahead but Black can still resist. (Another possibility is 8...cxd2+ 9.Bd2 Ne4, but White has a winning attack with 10.O-O-O.)
It's more likely that your opponents will try to hang on to the piece with 8...Ng4, and then things get fun - at least if you have the White pieces. Best now is 9.d4! (threatening 10.Bg5+), when Black has two options:
(a) 9...cxb2 10.Bg5+ Nf6 11.Rb1 with a significant advantage, thanks to his attack and huge leads in space and development.
(b) 9...Qxd4 10.Bg5+ Nf6 11.Nf3 cxb2 (everything loses, but this is the most entertaining option) 12.O-O! bxa1Q 13.Nxd4 Qxd4 14.exf6+ gxf6 15.Re1+ Kd6 (15...Kd8 16.Re8+ followed by 17.Qe6#) and now, as Mike Tyson might say, 16.Bxf6! sends Black to "bolivian", as 16...Qxf6 17.Qd5 is mate.

In sum, while there was a time when I didn't enjoy facing the 4...Nf6 line, those days are long gone, thanks to the 6.Bc4! idea. Further, the results back this up: in the 5.Nc3 line (playing 6.Bc4 whenever facing 5...d4), my score in ICC 3 0 blitz against players rated from 2486 to 2712 is a terrific 14-1 (and I was better in the one loss, too, against one of the lower-rated players in the bunch). Better still, many of my wins were extremely quick - in the 5...d4 6.Bc4 line, for example, my score is a clean 5-0 with the wins coming in just 21, 11, 17, 13 and 16 moves!
I have been honest and warned the reader: there are problems with this whole 4.Qb3 line for White. Speaking from a purely practical perspective, however, there are many more ways for Black to go wrong than White - and when Black takes a wrong step in this variation, it's often straight into the abyss. So give it a shot, French Defense foes!
(The analysis can be replayed here.)
[Reposted from my previous blog.]
Related Posts (on one page):
- Something Different vs. the French: Part 4
- Something Different vs. the French: Part 3
- Something Different vs. the French: Part 2
- Something Different vs. the French Defense: Part 1
I showed this analysis to a student who was frustrated with the French defense. He really liked it until we discovered a troublesome line. In the first part first, the line after 5...Qe7 6.Nc3 Nf6 7.d3 exd3+ 8.Kf1 ...Nbd7!(as played in one of your blitz games listed) 9.Bg5 Nc5 10.Qa3 Be6(...Qe5 played in the blitz game is weak) leaves almost no compensation (as far as I can tell)for the missing material. Any thoughts? Thanks in advance,
Roger
Roger: Good to hear from you. You're welcome - I'm glad you're enjoying the site.
That is a problem line, but there is an improvement ready to hand: 8.Be3. Still, this line is ultimately a bag of tricks - stay tuned for subsequent posts. (Albeit a bag of very dangerous tricks.)
I don't have a lot of time to study openings, and so I like to be consistent and comprehensive - no matter what White plays, I can fit it to one of my two main defenses, get 5-8 moves in, and know I can start the middlegame from a comfortable position. But an important point, which I tend to ignore even as the books mention it, is that you have to respond to the moves on the board, and not try to fit the board to match your preparation.
So if it means that 2.c4 makes it too scary or messy to stay in the French waters, so be it. Thanks for scaring me - and reminding me that opening prep should not override your understanding of the game at hand.