After:  6.Nge2 c5 7.d5 e6 8.Ng3 exd5 9.cxd5 a6 10.a4 h5 11.Be2 Nh7 12.Be3 h4 13.Nf1, and now instead of 13...Nd7, John Doknjas in his Opening Repertoire: Modern Benoni book recommends the immediate 13...f5 and if 14.exf5 he gives the novelty 14...Bxf5 with very complicated play but perhaps satisfactory for Black.

Usually, when picking up a complex opening like the Saemisch KID, I take a look at recent trends / correspondence games and do some of my own opening analysis to figure out the state of theory.

1) I first became interested in 6. Bg5 after reading Eric's book. Unfortunately, I think Black's equalizing task is easiest here:

6...c5 7. d5 e6 8. Qd2 h6 9. Be3 exd5 10. cxd5 a6 11. a4 Re8 12. Nge2 Nbd7 13. Nd1 (13. Nc1 Nxe4!!) Ne5 14. Nec3 g5!? 15. Be2 Nh5 16. O-O. So far, we've been following Dreev-Kasparov 1996. Here, probably the simplest continuation for Black is 16...b6 =, preventing the fixing of the queenside.

2) 6. Be3 c5 is the traditional mainline. 7. dxc5 or 7. d5 won't yield an advantage against prepared opponents, and are simply easier to play with Black. Instead 7. Nge2 is the most challenging continuation, aiming for an improved Benoni.

The main theoretical discussion is correspondence play centers around 7... Nc6 8. d5 Ne5 9. Ng3 h5 10. Be2 h4 11. Nf1 e6 12. Bg5!? Qb6 13. Qd2 exd5 14. Bxf6 Bxf6 15. Nxd5 Qd8 16. Nxf6+ Qxf6 17. Ne3 Nc6 18. O-O-O Nd4 19. Rhf1. Stockfish is quite happy with White here, but Black has good control over the dark squares, and has held the draw consistently.

Black has various other options, such as 7...Qa5!? 8. Nc1 cxd4 9. Nb3 Qb6!?, which is a tough nut to crack. Overall, though, I don't think I would mind being White in any of these lines.

3) 6. Nge2 seems to be the trendy line in super-GM praxis. One point is that the bishop placement is delayed, but another point is more concrete, and has to do with the Benoni:

6...c5 7. d5 (7. Be3, with a transposition above, is also possible) e6 8. Ng3 exd5 9. cxd5 a6 10. a4 h5 11. Be2 Nh7 12. Be3 h4 13. Nf1 Nd7 14. Nd2 f5 15. exf5 gxf5 16. O-O Ne5 17. Kh1 +=

By speeding up kingside development as much as possible, White "funnels" Black into a line which is strategically dangerous. From a modern preparation standpoint, this is especially poisonous, since Stockfish initially likes 17...f4 18. Bf2 for Black, before realizing how large White's advantage really is.

Considering that the Panno approach has not been doing very well, attention has shifted to 6...a6 7. Be3 Nbd7, intending 8. Qd2 b5!, which has scored heavily for Black.

White should not accept this gambit (which is an equalizer), but should instead aim for a quick kingside assault with something along the lines of 8. g4 or 9. h4, when Black's task is more challenging.

If I had to play the Saemisch KID (or the f3 anti-Grunfeld), I think I would choose the 6. Nge2 move-order, like a few recent authors have done.

Paddy wrote on 08/01/20 at 11:08:00:


We're talking about the same book, and I do think that it's excellent.  I refer to it often in my own book, and already said that it inspired me to play the Samisch.  Still by today's standards (and the standards of Gallagher's later repertoire books from the Black side), there isn't much explanation.  As a beginning player I learned a ton from it, but I had to truly puzzle out a lot of things on my own--which ultimately made me understand the Samisch much better.  Still, much of what I puzzled out on my own with Gallagher's text back in 1995-1998 or so has been explained in more recent books, some on the opening but some on positional play, etc. 

For many years I would have recommended Gallagher's book above any other single source on the Samisch.  It's harder to do that twenty-five years after it was published.  Theory moves on, but so do reader expectations.  If someone purchases it expecting explanations like those found in Gallagher's later Play the King's Indian, they'll be disappointed.  I just want a prospective buyer to be aware of that, though as I've stated it's still an excellent book. 

I completely agree about your later points regarding studying classic games, etc.  I would point out, however, that Gallagher's The Samisch King's Indian was full of the latest theory at the time of writing, and doesn't have many of those old classics that you refer to. 

As an aside, I originally wanted to have a "Strategic Themes" type of chapter in my book, and had a lot of examples of games in KID structures picked out, many that didn't arise from the Samisch.  In the end, almost all of those examples found their way into the book, but interspersed throughout in the relevant sections, and not as a separate chapter.

Anyhow I'm obviously a bit biased regarding the Samisch, so I think I'll step back from the thread and not dominate the discussion.

HAJS wrote on 07/31/20 at 18:57:28:


This book was excellent when it was published twenty-five years ago and inspired me to play the Samisch.  However a lot has changed in two and a half decades, and there isn't very much explanation in his old text. 

Being completely honest, I don't see a lot of material devoted to the Samisch that I would give a strong recommendation for.  I don't like disparaging other works, but there are some that I wouldn't bother buying even if they were $1.99 in a used book store! 

You don't need a lot of material to get started, really.  One or two good opening books, maybe a collection of games that has some relevant material (Mastering the Endgame Volume 2, for instance), a good database, and get going.

Paddy wrote on 07/30/20 at 22:38:54:


Another great summary. A lot of information packed into one reply. Thank you Paddy.

With regards to Joe Gallagher's book on the Saemisch, I will have to get that one. I briefly read parts of his repertoire from 2004 when preparing for an opponent and I found his explanations and proposed lines great. Unfortunately I did not get as far as reading the whole book, only a specific chapter that I was intending on playing from the White side. In the games I played against this player I would describe all I got was dynamic equality. Have you read his book and do you happen to know of any weaknesses in his recommendations? 

HAJS wrote on 07/30/20 at 17:07:33:

Barrish just writes about 6.Nge2:
"By not committing our dark bishop yet, we have the extra option of playing Bg5 sometimes."

Paddy wrote on 07/30/20 at 22:38:54:


My understanding of the 6th move issue is each of the three moves has advantages and disadvantages:

6.Be3 was the undisputed main line until around 1990, since everyone assumed it stopped 6...c5. But then several games by Shirov, Gelfand and others showed Black can gambit the pawn and get enough compensation, even with the Queens exchanged. White could play 6.Be3 c5 7.d5 instead and aim for a Benoni transposition, but the usual view is the bishop would have been slightly better on g5 in that case.

Both 6.Bg5, 6.Nge2 and 6.Be3 c5 7.Nge2 are attempts to solve this dilemma - the latter move order allows Black to transpose to a Maroczy Bind or continue to offer a Benoni structure, but with the knight forced to choose a square possibly a bit earlier than he would ideally have liked with 7...Nc6. Then 8.d5 Ne5 is standard and has developed a lot of sometimes razor-sharp theory, while 8.d5 Na5, which was little-known before 2010, is more experimental but has done well in practice. But even after 7...Nc6 8.d5 Ne5 there are lines where the bishop turns out to belong on g5, so White takes a tempo loss with Be3-g5.

I believe if Black promised not to play 6...c5 there would be no reason for White to avoid 6.Be3, as it overprotects the centre and is a great bishop placement in all lines except the Benoni lines. But the other moves also have their downsides:

6.Nge2 keeps the option of Bg5 in one move, but loses the option of playing Nh3, which is normally the ideal response to an early ...Nbd7 by Black, when ...Bxh3 is no longer possible - on h3 the knight avoids blocking White's other pieces and usually goes on to find a convenient home on f2. From e2 the the knight will have to move again to allow Bf1 to delevop, but each of the possible squares g3, c1 and f4 are also a bit awkward.
So if Black wants to try to exploit the downsides of 6.Nge2, it's logical to either go for a line with an early ...Nbd7 (i.e. the Byrne system or the Benko attempt with Nbd7/a6/c5/b5) or allow a Benoni but try to exploit White's usual Ne2-g3 by answering it with ...h5-h4 or the Fischeresque ...Nh5!?.

6.Bg5 gets the bishop to its best "Benoni square" immediately and as a bonus prevents 6...e5, a traditional main line against 6.Be3 (I still get 6.Bg5 e5?? 7.dxe5 dxe5 8.Qxd8 Rxd8 9.Nd5 +- sometimes in bullet or blitz). White isn't worried about 6...h6, since he will win that tempo back with 7.Be3 and a later Qd2. The downside is the centre and specifically d4 is a bit less secure than with Be3. I think if Black wants to exploit this he should try some of the many lines based on quick Queenside play with ...b5, since that may be more uncomfortable for White if his centre doesn't feel quite secure. There are many setups where Black plays or at least threatens ...b5: The Panno with Nc6, a6 and Rb8, the Byrne system with a6, c6 and Nbd7, the Benko attempt with Nbd7/a6/c5/b5, a recently popular move order covered by Vigorito on ChessPublishing with ...a6, Nbd7 and ...b5 as a pawn sac, etc. Well, this is my impression after dabbling in 6.Bg5 without ever learning the theory very well - I'm sure ErictheRed will correct my misconceptions!

So to turn the board around, Black can try to be crafty and play the main lines of 6.Be3 c5 but against 6.Nge2 and 6.Bg5 do something different to try to exploit their specific downsides. But in practice most Black players, especially on amateur level, have one favorite system and try to make it work against all the three 6th moves.

There's an irony with 6.Be3 c5 though: Even though it's objectively strong, with Black I would be tempted to avoid it because if White accepts the gambit and knows what he's doing, the queens come off early and things can end up quite drawish. So actually the traditional 6.Be3 with acceptance of the gambit could be a shrewd psychological weapon against King's Indian players out for blood. But that ploy requires a readiness to play in radically different styles from White.

Quote:

Very interesting, thanks -- I hadn't considered 5 Nge2. After 5 ...g6, can White thereby do any better than a regular Fianchetto-with-Nbd7 position?  I guess he can try a Saemisch (but advantageously? ...), thus bringing us nicely back on-topic! But Black can surely also keep flexi with 5 ...c6.

Michael Ayton wrote on 07/30/20 at 17:03:12:


I think White may well go 4 e4 e5 5 Nge2 or something, in that case.