NM Mark Kernighan-Michael Goeller

Round 9, March 17, 2005
King's Indian Defense, Smyslov Variation [E61]
Kenilworth Chess Club Championship, Kenilworth, NJ USA
Annotated by Michael Goeller

1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.Bg5
This is called The Smyslov Variation or Smyslov System, after the former World Champion who had good results with it. A recent article in New in Chess Yearbook suggests that it should be taken seriously. But I think it is rather tame and when I saw Mark play it against Demetrick earlier in the tournament, I started to think about playing the King's Indian against him. I had put together a good bit of research on the line and was going to prepare well for our game, but I had expected we would not meet until the following week....

I think this is best, though Black can probably play a wide variety of things. But I tend to like an early ...c5 whenever White develops his Bishop so early to g5 -- as in the Trompowski (1.d4 Nf6 2.Bg5 c5!)

a) 4...0–0 5.Nf3 d6 (5...h6! 6.Bf4 g5!? 7.Bg3 d6 8.e3 Nh5 unclear) 6.Qd2! (6.e3 Nbd7 7.Bd3 h6 8.Bh4 c5=) 6...c5 7.dxc5!? dxc5 8.Rd1unclear;

b) 4...d6 is considered the main line: 5.Nf3 h6! 6.Bf4 Nc6! 7.h3 Nd7!? was one thing I considered trying on Mark;

c) 4...h6!? 5.Bf4 c5 is similar to the game.

Position after 4....c5

Mark had played 5.dxc5 Qa5 = in a game with Massey, though I was not aware of that until Mark told me during our post-mortem. I had prepared 5...Na6! =+. Much more common is the pawn push 5.d5, which can lead to positions reminiscent of the Benoni or Benko Gambit.

5...cxd4 6.Nxd4
Mark considered 6.Qxd4 Nc6 7.Qd2 h6= which has been played by Smyslov himself.

6...Nc6 7.Nxc6!?
... and here ends my preparation. I'm glad I did not get to waste another week on preparing, since I probably never would have looked at this specific line. Possible are 7.e3 Qa5 8.Bxf6 Bxf6 9.Qd2= and 7.c5!?

... and I like Black's long-term chances with the preponderance of pawns and open b-file.

8.Qd2 h6!
This move seems necessary at some point in this line, though it can commit Black to a later ... g5. Though it might weaken the kingside, Black does usually get the Bishop pair out of it.

I expected this retreat rather than the more typical Bh4, since Mark had played it before in a game I found online. I also noticed that it was a tendency of his generally. More common is the retreat to h4.

9...Qa5 10.e4 g5!?
Going after the Bishop pair. But perhaps first 10...d6 is safer.

Position after 10...g5

Better 11.Be3 Ng4 12.Bd4 e5! (both Mark and I had examined the line 12...Ne5!? which Mark did not like because Black threatens c5! -- but I did not like it either because of 13.c5! d6 14.cxd6 exd6 15.Be2 +=. I am not sure I would have correctly tried ...e5! instead) 13.Be3 Rb8! 14.Rb1 (14.Be2 Nxe3 15.fxe3 0–0) 14...Nxe3 15.Qxe3 Bf8! 16.Be2 Bc5 17.Qf3 Qd8 18.0–0 d6 19.Na4 Bd4 =+

11...Nh5 12.Be2 Nxg3 13.hxg3 Rb8 14.0–0

Position after 14.0-0

A bit stereotyped. As my analysis of this line suggests, neither player should be doing the traditional Tarraschian stuff. Black should consider a more adventurous strategy involving either the exchange of Bishop for Knight at c3 or the advance of the h-pawn. Both of these ideas run counter to the Tarrasch tradition, but they fit quite well with more contemporary and concrete ideas of playing such a position.

a) Stoyko suggested 14...Bxc3! 15.bxc3 (15.Qxc3?! Qxc3 16.bxc3 d6 17.c5 dxc5 and Black's material advantage gives him some long-term prospects.) 15...d6 and Black has great long-term prospects in the doubled c-pawns and the weak dark squares.

b) I had looked at and rejected 14...h5!? 15.Rab1 h4 with some initiative, which is a line that John Watson might like as an example of "Modern Chess Strategy." Play might go 16.gxh4 (16.g4 d6=+) 16...gxh4 (16...Rxh4 17.g3 Rh8 18.Bg4 d6 19.Bxc8 Rxc8) 17.Bg4! d6 18.Bxc8 Rxc8 19.Qe3 h3!? 20.gxh3 Bd4 (20...Qh5) 21.Qf3 (21.Qxd4? Qg5+!) 21...Qe5 unclear but good for Black.

15.Rab1 Qc5!?
T o stop f4 and b4 thrusts while getting the Queen ready for activity on the kingside. I did not like 15...d6 16.b4! Qe5 17.Rfc1 f5!? unclear

16.Rfd1 d6 17.Na4?!
Better 17.b4! Qe5 (17...Rxb4!? 18.Nd5 cxd5 19.Rxb4 dxe4) 18.Rb3 a5 (18...g4!?) 19.f4 unclear.

I also considered 17...Qb4 18.Qxb4 Rxb4 19.b3 c5 with a lock on dark squares and the two Bishops.

Mark thought the whole game now revolved around my a-pawn, but I wasn't even thinking about it.

18...f5! 19.Qxa7 Rb7 20.Qe3 Qxe4!
Fritz likes this better than taking with the pawn also, but Mark had expected ...fxe4. I still don't know what made the decision, but I wanted to eliminate the blockade at e3. Play may have gone 20...fxe4 21.b4!? Bf5 22.a3 unclear or 22.Nb6!? Qc3 23.c5 Qxe3 24.fxe3 dxc5 25.bxc5 Ra7

21.Qxe4 fxe4 22.Bh5!?
Better 22.c5! d5 23.Ba6 Rb4 24.Bxc8 Rxc8 25.b3 Rf8! (25...Bd4!?) 26.a3 Rb7 27.Kf1 e6 28.Ke2 Rbf7 29.Rf1 g4 (29...Bd4 30.f3 exf3+ 31.gxf3 e5) 30.Rbd1 and though Black has pressure, White should hold.

Position after 22.Bh5

22...Bf5! =+
Mark says he overlooked that this indirectly attacks his Rook at b1.

23.g4 Bh7 24.Rbc1 Bxb2
Better was the immediate 24...e3! which is the sort of thing I had in mind going into Qxe4: 25.fxe3 Bxb2 26.Nxb2 Rxb2 27.Rf1 Rxf1+ 28.Rxf1 Rxa2 29.c5! Rc2!–+

25.Rb1 Rfb8 26.Rxb2 Rxb2 27.Nxb2 Rxb2 28.c5 dxc5 29.Rc1 Rxa2 30.Rxc5 e3!
... and here I offered and he accepted the draw.


Final position after 30...e3

We both had less than 5 minutes, but I feared him in a time-clinch and did not see strong winning chances after 31.fxe3 Be4?! (better 31...Rc2! 32.Re5 Kf8–+) 32.Re5! Bd5?! (better 32...Bxg2! 33.Rxe7 Re2) 33.Rxe7 Rxg2+ 34.Kf1 =+. He thought I should have played on since Black would have all the chances, but I had seen Mark win many a lost game on the clock and did not trust my speed, especially when trying for a win. As Scott Massey told me later, though, "you have to risk losing in order to win."


Updated 04.11.2005 | Contact Michael Goeller