Update on anti-hypermodern systems

[Rudel]

Hi all,
I just wanted to give a quick update [not very useful in terms of applicability] to my efforts at forming a cohesive system against the hypermodern K-side fianchetto defenses.

The basic idea [as seen in this thread] is to play e3 to give d4 more stability, which is very uncommon in many K-side fianchetto defenses that simply assume White will play e4.

The most interesting question right now is what to do if Black plays completely non-commitally by simply setting up his K-side fortress immediately:

1.d4 Nf6 2.Nf3 g6 3.e3 Bg7 4.c4 0-0

From here, White would generally like to play Nc3, Be2, and 0-0 in some order... Black has several options:
...d5/Gruenfeld play
...d6+...c5, transposing to a type of Benoni structure
...d6+...e5, KID play
...d6+ ...Bg5, a type of pircish setup.

Unfortunately neither Be2 nor Nc3 is without its drawbacks. Nc3 gives Black something to exchange his Knight for after 5.Nc3 d5 6.cxd5 Nxd5 7.e4. Be2 may well end up being a wasted move if played immediately. In some lines Bd3 may actually be better, and in others White may wish to retain the option of Bxc4 should White capture on that square.

What I'm looking at right now:

5.h3!?
This is a decent waiting move that removes ...Bg5 but mostly is meant to force Black to show his hand. Unfortunately, now that Black has castled, some of the power of my Sneaky Gruenfeld line has been lost. Still I think (provisionally) that if you don't know what your opponent is playing, it is my current choice for best.

5...d5 6.cxd5 Nxd5 7.e4 Nb6 8.Nc3 c5, and now both 9.d5!? f5!? and 9.Be3 look interesting because it looks like Black will be moved to exchange his B/g7, and that has to be at least some concession.

5.Be2

This looks good if you know your opponent likes the Gruenfeld or KID because I think White gets something pretty easily unless Black plays 5...c5 immediately. After that, White will gain a nice center but Black will get pressure down the e-file...still I think White may get something there as well:

5.Be2 c5 6.d5 e6 7.0-0 exd5 8.cxd5 d6 9.Nc3 Re8 10.Nd2 Qe7 11.e4

Looks pretty good...but I'm not sure about the implications of Black slipping ...Na6 in somewhere...

5.Nc3
This simply ignores the problem I mentioned earlier, expecting that White gets something else out of the bargain. For example, after 5.Nc3 d5 6.cxd5 Nxd5 7.Bc4, White's Bishops look much better than Blacks if Black continues 7...Nxc3 8.bxc3 c5 9.0-0 Nc6 10.Ba3 b6

But if Black is more restrained and retreats with 7...Nb6, I don't know that White can find the materials for a compelling plan after 8.Be2 (8.Bb3 c5! 9.0-0 cxd4 gets murky but might have some gems in it) 8...c5 9.0-0 cxd4

I welcome any thoughts on these lines.

[Rudel]

One thing I didn't mention about 5.Be2...

5...c5 gains extra strength after 5.Be2 because, had White known Black was going in for a Benoni, he might have wanted to put his Bishop on d3 to help push through e4 without needing Nd2.

Note that 5...d6 (after any of the options) is generally somewhat weak for the simple fact that following up with ...c5 has much less punch than had Black played ...c5 first. Part of the strength of 5...c5 is the threat of 6...d5, creating a formation White really would prefer to avoid in general. This is also part of the punch behind 5...d5, threatening to set up the same formation with 6...c5.

[MaxwellsDemon]

Have you read Yusupov's third book yet? I know you said that you had ordered it. His recommendation against the KID/Grunfeld is the Smyslov line with e3. It seems to bear a lot of resemblance to the line you are working on. Perhaps some of his ideas can be applied to your line. Just a thought.

-Max

[Rudel]

Right! I had forgotten about that. I think he advices b3 a bit before I had in mind, but I will definitely give it a try.

Thanks for reminding me!

[Rudel]

Looking into his repertoire, I found odd that after 1.d4 Nf6 2.Nf3 g6 3.e3 Bg7 4.c4 0-0 5.Nc3 d5 6.Qb3 dxc4 7.Bxc4, Yusupov only considers lines with ...Nbd7 when the immediately 7...Nc6 is (at least in my database) more common than all others combined.

However, I noticed that no one above 2600 has played 7...Nc6, so perhaps as I look further into that line I'll find out what happens after 8.Ne5 Nxe5 9.dxe5 Nd7 10.e6...looks juicy!

[Rudel]

It looks like 5.Be2 is going to be the official recommendation here, but I"ve found interesting tries in other lines.

One thing I love about 5.Be2 is that class players who are comfortable with the Benoni are extremely likely to have already played ...c5 earlier [1.d4 Nf6 2.Nf3 c5], so I'm hopeful that often those games that even get to the position I mention [1.d4 Nf6 2.Nf3 g6 3.e3 Bg7 4.c4 0-0] will be ones where Black is not a Benoni player. This is relevant because I think the best response to 5.Be2 is 5...c5, but my guess is that White will actually encounter other moves far more often in terms of practice.

[Rudel]

I found today that Joe Gallagher's "Play the King's Indian" (which claims to be a complete repertoire) does not even mention this setup where White keeps his pawn on e3.