TonyRo,

Forgive my ambiguity, IM Watson advocates the Makogonov line for White in his repertoire book "A Strategic Chess Opening Repertoire for White". I'm a mere 1800 strength player here in the U.S., but I've come to appreciate this line for White.
I recall vaguely that Kasparov referred to it as an "Insideous" line for White. Sounds like a compliment to me. GMs Yermolinsky and Karpov are known to have used this line as White. Karpov has only used it in faster time controls and simuls, fwiw.

The former.

lnn2 wrote on 05/26/14 at 16:46:39:


In some cases White's N on d2 can end up on e4 or c4.

One thing I notice about the Makogonov is that the White knight on f3 doesn't seem to have any good squares to go to in the main line but just goes to d2  and remains there to protect the e4 pawn. If that's the case, then the benefits of having a knight on f3 seem rather overrated ... There is an argument that securing the e4 pawn with 5. f3 gives more flexibility to White's pieces.

The main difference is that 5.f3 supports White's center and 5.h3 does not.  Hence White can more easily "build up" in the Saemisch variation, and also switch play to either wing more easily.  Because e4 is overprotected, plans of putting a knight on c5 and pressuring e4 (as in many other variations of the King's Indian) just don't work for Black.  This is true in lines where White closes the center with d4-d5 and when he allows ...exd4.

In addition, White can often threaten (or at least make Black think about) a h-file attack with the Queen+Bishop battery, which Black sometimes needs (or chooses) to react to.  The popularity of 6...c5 is due in large part to Black not wanting to face the h-file attack, and in the Panno variation with 6...Nc6, 7...a6, and 8...Rb8 Black spends two tempi to immediately prepare queenside play of his own should White choose to castle in the direction.  If you have no inclination of ever launching the old h-file attack, you might want to shy away from the Saemisch.  I personally don't choose the h-file attacking lines all that often, but if Black ever makes a "slow" move (like 8...Re8 in the Panno or 5...0-0, 6...a6, and 7...c6), I'll go for it.

Reed wrote on 05/22/14 at 16:46:55:


Saemisch would play f3 with the idea of a big centre against the Nimzo-Indian as well. The idea that h3 can achieve the same result in the Kings Indian is a more recent development. Back in the days when I first learned the Kings Indian by reference to the Batsford book of that name by Keene, Hartston and Barden, h3 was regarded as a means of improving exchange variations. This was down to  a plan used by Larsen on occasions. 

I'm still struggling a bit to understand what to expect next when White tries these h3 ideas on move 5 or 6. One of them is a Petrosian system style immediate pawn advance meeting e5 with d5. Alternatively when I've tried for an early f5, they've played g4. Then they play gxf5, castle long and generate a strong attack on the g file. 

Lines and themes in the Saemisch can be much better known by the defender. It's also easier particularly against Bg5 lines to switch into Benoni structures and tactics with .. c5 instead of .. e5.

After 1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 --

What's the difference in White's plan after 5.f3 (Saemisch) compared to 5.h3 (Makagonov)? Both variations allow a prophylactic g2-g4 to cramp Black's kingside expansion, except 5.h3 allows the knight to be developed naturally at f3. To my mind that makes it the Makagonov strategically better, but for some reason the Saemisch is way more popular. Why is that? 

I have a couple of possible reasons that I don't think are very good... One is that the e4 pawn is not as well-reinforced in case Black gets in f7-f5 before White can play g2-g4, but in case of ...fxe4 White usually recaptures with Nc3xe4 anyway. 

Another possible reason is that White can advance h2-h4-h5 in the Saemisch. But again, White's pawn advances to g4 and h5 are more prophylactic than aggressive. The intention is not usually to start a kingside attack, so playing g2-g4 seems good enough.