Riding The Winds Of Fashion, Part 3

  • IM Silman
  • | Jul 22, 2014

Warning! This is a fairly advanced article. I try to write for players of all ratings, so sometimes I explore basic stuff, and other times I bow to the 1800-and-above crowd. Though the opening in question is too advanced for beginners, I hope that some of the positions will prove of interest to everyone.

When you take up 1.d4 and decide to enter main lines (and avoid sound but non-critical systems like 1.d4 Nf6 2.Nf3 followed by Bg5 or Bf4), one of the first questions you have to ask yourself is “what do I do after 1.d4 Nf6 2.c4 e6?”

That’s a huge question, since if you play the “righteous” 3.Nc3 then you have to be well prepared for 3...Bb4. And if you play 3.g3, then 3...d5 takes you into a Catalan, while 3…c5 4.d5 leads to a fianchetto Benoni (not the most critical way to play against the Benoni).

Of course, 3.g3 can also be met by 3...Bb4+. The most common reply to 2…e6 is the somewhat cowardly 3.Nf3. (OK, it’s a hugely popular main line, but it is saying, “I’m terrified of the Nimzo-Indian!”) After 3.Nf3, Black can surprise White with the Queen’s Gambit via 3...d5, or (more often) he will toss out the Bogo-Indian (3...Bb4+), or the Queen’s Indian (3...b6).


When I was playing, I experimented with everything. I had some nice wins with the Catalan (3.g3), and I did fine with the Queen’s Indian, but part of me always yearned for the Nimzo-Indian.

Yes, I played it with Black, but why not go for the gusto and play both sides of the Nimzo? In 1980, I bit the bullet and took up the White side of the Nimzo-Indian. In general, I was successful with 4.e3 and won some nice games:

Though this was a nice result, I decided that I should look for something a bit less mainstream. It was at this point that 1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.f3 came to my attention. However, there were a couple problems that needed to be solved. First off, I had to decide if I like the complicated position that resulted after 1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.f3 d5 5.a3 Bxc3+ 6.bxc3 c5 7.cxd5 Nxd5 8.dxc5.

The second problem was the following move sequence: 1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.f3 c5 5.d5 Bxc3+ 6.bxc3 Nh5 7.g3 f5 8.e4 f4, which theory at that time thought was highly promising for Black.

After looking closely at 1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.f3 d5 5.a3 Bxc3+ 6.bxc3 c5 7.cxd5 Nxd5 8.dxc5, I decided that it was, at the very least, highly interesting and offered very good chances to outplay one’s opponent. But 8...f4 was quite another matter.

It was here that the recently deceased U.S. Champion John Grefe came to the rescue. We were good friends at that time and often analyzed together, so I showed him the position after 8...f4 and, after a couple days' work, we cracked it! (Remember that chess engines didn’t exist yet.) 

The Creation of High Fashion

Luck was on my side when just a couple weeks after our discoveries, the very strong master George Kane entered our trap.

Silman – George Kane, [E20] Bagby Memorial 1982

1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.f3 c5 5.d5 Bxc3+ 6.bxc3 Nh5 7.g3 f5 8.e4 f4


The refutation of 8...f4. 


Of course, 9...0-0 loses right away to 10.Qd5 with the double threat of 11.Qxh5 and 11.e7+, while 9...fxg3 gives White a winning advantage after 10.Qd5!

10.Ne2 fxg3 11.Bg2 Qxe6 12.hxg3 Nf6 13.g4 0-0 14.g5 Ne8 15.Nf4 Qe5 16.Nd5

When annotating the Ivanchuk-Csom game (below), Chris Ward (I guess in 1989) mentioned my game up to 15…Qe5 and said, “White went on to win, though not via the fairly straightforward 16.Qd5+.”

Actually, though 16.Qd5+ is very strong, my 16.Nd5 is far stronger.

16...Qg3+ 17.Kf1 Nc6 18.Rh3 Qe5 19.Kg1 g6 20.f4 Qg7 21.e5 d6 22.Nf6+ Nxf6 23.exf6 Qf7 24.Re3 Qc7 25.Bd5+ Kh8 26.Qe2 Bf5 27.Bd2 Rad8 28.Re1 h5 29.Re7 Rd7 30.Re8 Rd8 31.Rxf8+ Rxf8 32.Qe8 Qd8 33.Bxc6 bxc6 34.Qxc6 Qd7 35.Qxd7 Bxd7 36.Re7 Bf5 37.Rxa7 Rb8 38.Ra6 Kg8 39.Rxd6 Ra8 40.Be3, 1-0.

This game appeared in the Player’s Chess News magazine, and after everyone saw that the feared 8...f4 was no longer a problem, 4.f3 quickly became quite popular. Suddenly, instead of my new system being under the radar, it became high fashion!

A funny aside is that this line was mentioned in the very good book, "Play the 4.f3 Nimzo-Indian" by Yuri Yakovich (Gambit, 2004). When mentioning the line 1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.f3 c5 5.d5 Bxc3+ 6.bxc3 Nh5 7.g3 f5 8.e4 f4, Yakovich wrote:

“Up to the end of the 1980s it was generally thought that 7...f4 promised Black a good game. However, Moskalenko’s discovery 8.dxe6! gives White the advantage."

He then gives “analysis” that copied my game up to 15.Nf4. When I saw this in Yakovich’s book I didn’t know whether I should laugh or cry at Grefe’s and my work (in 1982) being attributed to Moskalenko (in 1989).

Such is life in professional chess.

After my game with Kane, I was soon able to try out the position after 1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.f3 d5 5.a3 Bxc3+ 6.bxc3 c5 7.cxd5 Nxd5 8.dxc5 in a few games, and I won all of them. Here’s an example (note White’s use of the two bishops):

Though 4.f3 found itself riding the winds of fashion, eventually people lost interest in it. (I think the 1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.f3 d5 5.a3 Bxc3+ 6.bxc3 c5 7.cxd5 Nxd5 8.dxc5 line was thought to be fine for Black, and everyone moved on to the next best flavor of the day.)


A Good Opening Never Dies

However, good openings never die, and now 4.f3 is red hot again! I suspect that the thing that turned off 4.f3 fans (the 1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.f3 d5 5.a3 Bxc3+ 6.bxc3 c5 7.cxd5 Nxd5 8.dxc5 line) was rejuvenated with new ideas, thus allowing a whole new generation to enjoy a love affair with this system. 

When a game like this is played, everyone pays attention. When a World Champion plays it, everyone immediately copies it! So it was that 4.f3 was once again on the menu!

From Popular to Out-of-Control Wildfire!

Can anyone forget the following game? Anand played 4.f3 against Magnus Carlsen in their world championship match and seemingly built up a winning position. Many players watched this game live, and the commentators were sure that Magnus was doomed. But, for poor Anand, it was not to be.

White lost, but EVERYONE wanted the kind of position Anand had. Suddenly 4.f3 was on fire! Incredibly, shortly after the match, Nakamura (who wanted some of that mate-magic which Anand seemed to have discovered) used 4.f3 against Magnus, too. Sure enough, Carlsen was completely crushed, but (as if history has a grudge against Hikaru) the American player missed win after win and ended up losing.

Unfortunately, Nakamura’s 4.f3 curse has continued (he had a pleasant position in both games):

4.f3 is still on top of the world, living large in the ultra-fashion lane. But, like all things do, it will eventually fall out of favor, though it wouldn’t surprise me if it rose to the top of the heap again after other lines get their day in the sun.

For those that want more 4.f3, here are some recent games:

We will finish with another Nakamura game, though this time he has the Black pieces!



  • 2 years ago

    IM pfren

    Actually the plan Carlsen chose against Anand is the computer's top suggestion. I am very sceptical about its real value, IMO Black is risking too much.

    Trusting the computer's suggestions in such complex positions, where the engine's "thinking" horizon is nowhere close to the actual truth about the position, is a very unwise choice.

    Anand's nice 16.Nd4! against Wang Hao is actually not a new idea, and it's possible only due to Black's bad fifteenth move (15...Be6?!).

    It's well established that Black has a fine game after the correct reply 15...Na6! which should be attributed to GM Alex Delchev, and has successfully passed the tests at all levels of play.

    If Black wants to retain winning chances against the 4.f3 Nimzo, he should probably investigate this sharp line (already employed as Black by several GM's):

    IMO Black has a good game, since white cannot maintain his strong center, and can play for a win with little risk.

  • 2 years ago


    Mr. Silman,


    I have enjoyed all of the past articles that you wrote and I find them great, it is as if a friend tells me this stories about chess. 

    I thank you for the very nice ideas involved and the juvenile spirit.

    Also since chess is an intellectual pure game it is very likely someone will catch up your ideas and turn them into his own, happens in every other sport.


  • 2 years ago


    very nicely demonstrated and deeply exhibited article. thanks

  • 2 years ago


    I count 6 white wins, 7 draws, and 7 black wins (one of which was nakamura v carlsen where white had a totally WON position)

    so a fighting opening with chances for both sides, interesting fertile ground for experimentation

    only the brave need enter! material provides tons of canon-fodder to stock your arsenals

  • 2 years ago


    Did any of you examples with 4. f3 win? I saw a bunch of black wins and draws.

  • 2 years ago


    Magnus Carlsen should learn that the Nimzo-Indian just isn't for him. I mean he wins the games, but really he should lose them!

  • 2 years ago


    What about, if I'm black, after the cowardly (and common) 3. Nf3 I tried ...c5, the Blumenfeld Countergambit? Is it a sound response? Statistically it is supposed to offer better chances for a win for black than ...d5, ...b6 or ...Bb4+, and be less drawish.


  • 2 years ago


  • 2 years ago


    Good article on Nimzo, 4.f3 var.  Your examples reveal who studies and who does not. (Anand renown for homework, Carlsen not so much)  Studying also helps to maintain tactical and positional awareness.

  • 2 years ago


    Thanks IM Silman, I enjoyed very much the chess content of the article and the stories :)

  • 2 years ago


    Anand lost because of the tremendous amount of pressure of playing in his hometown and defending the title..This time things will be different..

  • 2 years ago


    The amusing thing is that in the 2013 and 2014 games, there are 6 draws, 7 wins for black, and no wins for white. I'll take a win from an inferior position, any day, over a loss from a winning position, which is the recent history shown in this article.

  • 2 years ago


    well, i find this article highly useful.

    lately, i have been trying to find interesting replies to nimzo-indian.

    here in chess.com , a user named achja( pretty stronger than me ) adviced me to take up f3 and told me to be careful about pawn pushes in this line.

    i went through each and every game,these games somehow reminds me of the KID: Smaeich variation games ....

    also, these games are showcase a very awesome mix of tactics and positional chess...

    i think that after reading this article , i will take it up..


    thank you silman!


    regarding nakamure, i feel sad about him....that even after getting winning position,he is not able to convert it...i don't how nakamura deals with this .

  • 2 years ago


    I've been playing Bf4 and I usually get a good game.  Winning lately.  I may try this out thought, perhaps it suits my style.  Thanks for the article.

  • 2 years ago


    Thanks beautiful and very useful.Laughing

  • 2 years ago


    Thanks Silman! Awesome article!

  • 2 years ago


    Based on this article, do you get the impression that Nakamura should ditch the Nimzo from his repertoire?

  • 2 years ago


    Why does every Silman article always have some disparaging undertones about the chess level of the reader?

  • 2 years ago

    IM DanielRensch

    Great read actually. I played the Nimzo, and have played just about every system / position referenced in this article. Jeremy, as usual, does a great job providing advice and summaries of the lines!

    Thumbs up!

  • 2 years ago


    I never got what was so scary about the Nimzo-Indian...

Back to Top

Post your reply: