trw wrote on 05/14/18 at 10:41:22:


Earlier I mentioned about that transposing to an old major line with colors reversed if Black plays 9...0-0.  Certainly that's not the only move, but in any case I'm not sure how it is straightforward for Black and tough for White etc. ...

Thanks for this. I often play 6 b3 iso 6 a4, so it's natural for me to try 6 a4 a5 7 b3. In the 6 b3 line I often end up feeling unsure where to develop my Nb1, but I notice that in the 6 a4 line White often goes Nb1-a3, then c2-c3, with the potential for a nice squeeze ... Meanwhile, I notice that after 6 a4 a5, Epishin plays 7 Bf4 (which without a4/..a5 in seems to score terribly) and always wins!

But after 6 a4 a5 7 Re1 d5, how does White avoid (8)Nc3, transposing to the stodgy line? [Edit. Ah, I see perhaps? -- 8 Nbd2, I imagine ...]

Quote:

In all my databases this scores very well for White! Can anyone summarise the theory on it (such as there might be in such an 'anti-theory' line), or indicate where there's any writing on it?

I've played the KIA on and off since I was 15 so over 50 years now. Also used it in correspondence from time to time as computers don't really understand it. As white I always found the Kings Indian Defence set-up most annoying. White really has to transpose into a mainline of the fianchetto variations to get an advantage which is often not where KIA players want to go.

an ordinary chessplayer wrote on 03/31/18 at 17:58:34:


You could also play the move order 1. e4 e5 2. Nf3 Nc6 3. g3, although again, how to meet 3. .. f5 is a question.

Alternatively there's 1.e4 e5 2. Nc3 Nf6 when g3 is respected. Also there's 1. e4 e5 2. Nf3 Nc6 3. Nc3 Nf6 4. g3 . In both cases though the Knight's on c3 and Black can play .. d5, so it's not really a KIA type of position. 

Many years ago if I intended to play the KIA from a 1. e4 move order against both the Sicilian and French, I concluded that a "normal" approach was necessary against everything else.

One thing about e.g. 1. e4 e5 2. Nf3 Nc6 3. g3 Nf6 4. d3 is that it could lead to Black playing a respected KIA-vs.-French line (1. e4 e6 2. d3 d5 3. Nd2 Nf6 4. Ngf3 Nc6 5. g3 de with ...Bc5 and ...e5) with an extra tempo.

fjd wrote on 03/30/18 at 18:55:45:


I'm not really blaming McDonald. As far as I am aware, no other book on the KIA covers g3 lines in the double-king pawn openings. And it is true that some the variations in the Glek and Smyslov do not feel like the KIA. However some do, at least to me. From ChessPublishing:
1. e4 e5 2. Nc3 Nc6 3. g3 g6 4. Bg2 Bg7 5. d3 d6 6. Be3
and
1. e4 e5 2. Nf3 Nc6 3. Nc3 Nf6 4. g3 Bc5 5. Bg2 d6 6. d3 O-O 7. Na4 Bb6 8. O-O h6 9. Nxb6 axb6 10. Re1 Bg4 11. h3 Be6 12. b3 Qd7 13. Kh2
and
1. e4 e5 2. Nf3 Nc6 3. Nc3 Nf6 4. g3 d5 5. exd5 Nxd5 6. 

Many of the lines McDonald presents have a KIA feel, but some do not.

Against the Dutch he presents 1Nf3 f5 2d3 Nf6 3e4 fxe 4dxe or 3...d6 4exf Bxf5 5d4

In the chapter on the French we have games that start:
1. e4 e6 2. d3 d5 3. Qe2 Nc6 4. Nf3 e5 5. g3 dxe4 6. dxe4 Nf6 7. Bg2 Bc5 8. O-O O-O 9. c3 a5 
and
1. e4 e6 2. d3 d5 3. Qe2 dxe4 4. dxe4 b6 5. Nd2 Ba6 6. Nc4 Nf6 7. Nf3 Nc6 8. e5   

White can fianchetto against the French, Caro-Kan, and Sicilian, and in double-king pawn openings. I think some readers would want coverage of all the preceding. 

It may be that full coverage of the Smyslov and Glek lines would be too much of a distraction. It's a judgment call that perhaps can only be made by the author.

To give credit where it is due, McDonald provides full coverage of all responses to 1Nf3 (where white intends the KIA). Truly, the scope of the KIA is vast. That is some of it's appeal. From 1Nf3, it may be the only opening you need for white. From 1e4, it is almost the only opening you need.

I bought GM McDonald's book (Kindle) on the KIA published in 2014. GM analysis of games is bound to be highly valuable. That is especially the case here, as the KIA is more about plans than about specific moves. Still, there are some specific lines that I would have liked to have seen. For example after:
1. e4 e6 2. d3 d5 3. Nd2 c5 4. Ngf3 Nc6 5. g3 Bd6 6. Bg2 Nge7 7. O-O O-O 8.Re1 
I would have liked to seen coverage of 8...Qc7, which has been recommended by John Watson.

The book has eight chapters which address both the 1e4 and 1Nf3 move orders. It could have contained an additional chapter addressing the KIA in double king pawn openings. There is the Smyslov system 1e4 e5 2Nc3 Nf6 3g3, and the and Gleck system 1e4 e5 2Nf3 Nc6 3Nc3 Nf6 4g3

Although white can hope for an advantage from the KIA, I think white should be psychologically prepared for an equal middle game, while bypassing tons of theory.

I formerly thought there were various "complete answers" to the KIA. By that I mean that particular moves completely neutralized the KIA when played against the French, or the Caro-Kan, or the Sicilian, or even from the 1Nf3 move order. Perhaps, but perhaps not! On closer inspection, it seems that repertoire books for black promise "equal chances" after a select mover order. That may be OK for white, who may sensibly favor his chances in a familiar position.

Laramonet wrote on 03/12/18 at 20:38:22:


Thanks. I haven't bought the Ntirlis book on the Queen's Gambit, or the Emms or McDonald books on the KIA. It's as much about disk space as about money with my older ASUS Ultrabook.

The idea you describe, with Rfc8 and Qd8 sounds like a good resource for black. At some point the black king will probably need some defensive help.

I have a 2010 version of Stockfish. It tends to overestimate black's chance, having a "show me" attitude towards white's impending king side attack. By the time it sees it, it's too late!

The positions are clearly rich. I imagine many players choose the KIA hoping to get this line. Yet I think black players can get chances too.

A completely different approach has been recommended for black in some repertoire books: d5 c6 Bf5 or Bg4. It's kind of a Slav/Caro, rather than a French/QGD, way of playing things. It's an answer, but not THE answer, IMO, - a matter of taste.

ErictheRed wrote on 03/09/18 at 17:00:18:


As this variation leads to a Chigorin Defence (by transposition), it is analysed by authors presenting this opening like Morozevich, Williams or Burgess (for White). E.g. Morozevich likes here 3...Bf5 4.Bg2 e6, with Nb4 to follow (considering immediate 4...Nb4 premature).

When my black opening was the QGD Tarrasch, I had a lot of success with a reversed Saemisch.

1.Nf3 d5 2.g3 c5 3.Bg2 Nc6

• 4.d4 e6 5.c4 Nf6 6.cxd5, transposing to the main line Tarrasch, was the invariable choice of strong players.
• 4.O-O e5 5.d3 f6 6.Nbd2, with a rather negative position for white, was the invariable choice of "other" players. Nbd2 is a compliant move, but it is also the only one I have ever faced! Almost anything else would be better, e.g. 6.e4 d4 7.Nh4, 6.c4 d4 7.e3, 6.Re1 Be6 7.Nc3, 6.a3, 6.c3, etc.

These days my repertoire is completely different, but one blitz variation I get a lot is the Elephant Gambit declined(!), e.g. 1. e4 e5 2. Nf3 Nf6 3. Nc3! d5!! 4. d3!!! d4 5. Ne2 Bd6 6. g3 c5 7. Bg2 Nc6 8. O-O h6!? So far nobody has played what must be the critical move 9. Nh4. I guess they don't know too much about the King's Indian. Anyway 9...g5 10. Nf5 Bxf5 11. exf5 Qc7 12. f4 gxf4 13. gxf4 O-O-O is quite sharp, not sure who is better without a few tests. That wBg2 is not exactly sleeping! Stockfish prefers 11...Qd7 which in fact is the first idea I had, but later I concluded the wPf5 is a red herring. Playing against the wBc1 seems stronger.

I have been thinking about RdC's reversed Pirc suggestion. After 1.Nf3 d5 2.g3 Nc6 3.Bg2 e5 4.O-O I don't think just any old reversed white system works equally well. In particular, black has to be alert for a sequence c2-c4, ...d5-d4, b2-b4!?. And white has another idea in d2-d4, ...e5-e4, Nf3-e5!?. 

I decided 4...f6 might be best. We should compare with variations after 1.Nf3 d5 2.c4 d4. A very recent blitz game went 4...f6 5.d4 e4 6.Ne1 h5 7.h4 g5 8.hxg5 h4 9.gxh4 Rxh4 10.Nf3?? exf3 11. exf3 Qd6 12. f4 fxg5 13. Re1+ Kf7 14. fxg5!! Qh2+ 15. Kf1 Bh3 16. Qf3+ Kg7 17. g6 Qh1+ 18. Ke2 Re8+ 19. Kd3 Nb4+ 20. Kc3 Qxe1+ 21. Bd2 Qe6 and 0-1 shortly. Black's early attack with 7...g5 was hardly convincing, but it did induce a panic from white. In retrospect, simply 6...f5 was good, when white has castled too soon in this Gurgenidze structure.

RdC also suggested the reversed 150 Attack. Both 4...Be6 5.d4 e4 6.Ng5 and 4...Nf6 5.d4 e4 6.Ne5 are interesting for white.

Maybe black could try a reversed Yugoslav idea from the 1970s: 4...Be7 5.c4 dxc4 (5...d4 6.b4!) 6.Qa4 Kf8!? 7.Qxc4 h5. Fischer gave this Kf1 idea for white in My 60 Memorable Games, but I believe it was just for blitz.

I always was reluctant to play/enter in KIA positions with Black (I play the French), due to some quick kills (complete annihilation would be more close to what happened) in my first days as a Frenchie.

But recently I entered "by error" in two KIA via Nf3-g3 move order (first by error, second more or less voluntarily) and got very decent positions with Black. I White wants blood in this kind of positions... well... Be7 plus g5 plus 0-0-0 and all the danger goes to White safety. Usual set-up with e6-d5-Nf6-c5.

Sources I use in this lines:

- Watson in his videos for ICC (Sharpen Your Chess sense series) on the French
- Niclas Huschenbeth videos for chess24 (his series against 1.Nf3)

Another set-up I like a lot is 1.e4 e6 2.d3 d5 3.Nd2 Nf6 4.Ngf3 b6 5.g3 dxe4 (via French move order).

Salut,

PD: As someone said before, one of the key points of 1.Nf3 is to be flexible. If someone only uses this to get a KIA against 1.Nf3 d5, then he/she is very predictable.

TN wrote on 08/01/14 at 17:55:38:




Really, you want me to give away the biggest secret about the KIA from my experience of having played it for over 30 years 



Well it's the secret to dealing with a lot of these reversed openings, it's to play the set up White would play against, in this case, the KID, i.e. d5/c5/Nc6/d5

White shouldn't let you get this set up, but KIA player purists will be complicit in allowing it, but the extra tempo does not help White , took me many many years to realise that. The KID with a tempo up is a marketing ploy for KIA books. The position is equal, but it's certainly winnable for Black.

FreeRepublic wrote on 03/09/18 at 17:31:14:

I lost, but not due to the opening.

TD wrote on 03/09/18 at 08:33:40:


Thanks for the excellent reply. For me, the f6 lines are new and seem murky. I guess I can't complain. Nobody said chess was easy. How did you do in your game?

The main lines, 10...a5 etc., are definitely interesting. However from the black perspective, I'm put off by white's favorable results in general, and by the (recent?) idea of racing the h pawn. So I'm not sure my study time will pay dividends.

Thanks for mentioning the Emms book from 2005. He is such a good author that I'm tempted to by the kindle edition. However Neil McDonald wrote a KIA book in 2014, also published by Everyman. It has all the same chapters, except the last chapter has been split into two. Both books seem to be from the white perspective, based upon the chapter selection. Have you seen the McDonald book?

FreeRepublic wrote on 03/08/18 at 23:24:25:

I got the position with 10.Nf1 once and what did my opponent play? Well, he played 10...f6!? and I was on unknown territory... Emms considers this move in "Starting Out: King's Indian Attack", with 11.exf6 Nxf6 12.Bf4! Stockfish approves of 11...Nxf6, but Emms thinks maybe 11...Bxf6!? is better. He offers hardly any analysis. Dvoretsky in "Opening Preparation" gives 11...Bxf6 12.h4! Nb6 13.Ng5!. 11...Bxf6 has been played most, with better results for Black and by better players.

RdC wrote on 08/05/14 at 10:58:09:


I was recently thinking along those lines. 3Bg2 is natural. A friend of mine once said that when you play g3 (or g6), you don't have to think about your next move. But here white can stop 3...e5 by playing 3d4. This can lead to a somewhat novel Catalan, where black has played an early Nc6. For example:  1. Nf3 d5 2. g3 Nc6!? 3. d4 e6 4. Bg2 Nf6 5. O-O Bd6!? Black contemplates ...e5.

kylemeister wrote on 08/05/14 at 16:05:53:


I was rather more thinking of a Reversed 150. If White uses the extra tempo to do something useless, castling long and hacking down the h file could have its points. Otherwise just play Be6/g4, Nf6, Bd6, Qd7 and 0-0-0 or 0-0 depending on where White pieces go. 

Even in a reverse Austrian, if they play c4 as white, you just take it and compare to the corresponding position from an English against e5, f5 systems.

There aren't that many examples, Hector of Sweden gives it a try from time to time. 

TN wrote on 08/01/14 at 17:55:38:


You can go for broke with 1. Nf3 d5 2. g3 Nc6 . The idea is to follow up the natural 3. Bg2 with 3. .. e5 . In a manner of speaking you now have the white pieces and can play whatever is your favourite anti-Pirc. On 3. d4 you have a Catalan v Chigorin type of position.

The most flexible players will not always play 1. Nf3 always intending a KIA. Some at least will be looking for transpositions to mainstream QP openings. This has to influence Black's choice of defence. 

I thought the system with ...c5, ...e6, ...d5, ...Nf6, and ...g6/...Bg7 was supposed to provide good chances for Black.

White has to be careful because he can hang a pawn with e4-e5 at the wrong time, and Black just moves pawns on the Queenside until White is actually threatening something (if ever), by which time he probably has the queenside busted open anyway.

I think ideas with ...d6 for Black in general is less promising for White in KIA set-ups, than the lines with an early ...d5 (and ...e6). Objectively, Black is fine in those too, but they are what White wants to play against, and what e.g. Fischer liked playing against.

I usually meet the KIA set-ups when playing French and Sicilian. I that case I play ...c5 (French) or have already played it. With your particular move order, Black can consider ...c6 and maybe play for ...e5 instead. Or after 3. d4, have a regular position that shouldn't be better for White.