Another Claim to Fame

I have already demonstrated my longest chessgame. I also have a game in 'The Quickest Chess Victories of All Time' (Graham Burgess, Cadogan 1998). A very nice book in my opinion but I may be biased as I luckily am on the winning side:

Sv.Johnsen - G.Nesheim
Gausdal Open Ch NOR, 1985

1.e4 c5 2.c3 Nf6
Recently Tarrasch' old 2...Qa5 has had a small boost in popularity. After 3.Nf3 Nc6 some notable continuations are:
a) 4.Na3 e6 5.Nc4 Qc7 6.d4 cxd4 7.Nxd4 a6 8.Bd3 b5 9.Ne3 Nf6 = Alapin-Tarrasch, Vienna 1898.
b) 4.Bc4 d6 5.Qe2 Nf6 6.h3 e5 7.0–0 Be7 = Rainfray-Movsesian, France 2003.
c) 4.g3 Nf6 5.Qe2 d5 6.exd5 Nxd5 7.Bg2 Bg4 8.h3 Bh5 9.0–0 e6 = Rozentalis-Movsesian, Hastings 1996.

3.e5 Nd5 4.d4 cxd4 5.Nf3 Qa5?! (Dia)

6.Nxd4
6.Qxd4 appears less logical. After 6...Nc6 7.Qe4 e6 8.Bd3 f5 9.Qe2 Qa4 10.h3 Nf4 11.Bxf4 Qxf4 chances were equal in Tzoumbas-Skembris, Liosia 1991.

6...e6
I think I had a faint hope of 6...Nc6 7.Nb3, which actually happened in Arslanov-Nozdrachev, Russian Ch U12 2004.

7.Nd2!
This appears more logical than 7.Bc4 Nb6 8.Bb5 a6 (8...Nc6 9.a4 Nd5 10.0–0 a6 11.Bxc6 bxc6 12.Na3 += Donets-Khatenever, St Petersburg 2005) 9.b4 Bxb4 10.cxb4 Qxb4+ 11.Qd2 Qxd2+ 12.Bxd2 axb5 13.Nxb5 Na6 14.Nd6 += Stojic-Edwards, Canberra 2003.

7...Nc6
Burgess notes that White is better after 7...a6 8.Nc4 Qc5 9.Qg4 as well as 7...d6 8.Nc4 Qd8 9.Nxd6+ Bxd6 10.Bb5+.

8.Nc4 Qd8 9.Nb5 Bc5
9...Be7 10.Nbd6+ Kf8 11.Qh5 g6 12.Bh6+ Kg8 13.Qf3 Bxd6 14.Nxd6 transposes to 9...Bc5 10.Nbd6+ Kf8 11.Qh5 g6 12.Bh6+ Kg8 13.Qf3 Bxd6 14.Nxd6 (note to Black's 11th move).

10.Nbd6+ Kf8 11.Qh5 (Dia)

11...Qe7?
This loses a piece. Black had to try 11...g6 12.Bh6+ (12.Qh6+ Kg8 13.Bg5 Qf8 14.Qxf8+ Kxf8 15.Ne4 Be7 16.Bh6+ Kg8 17.f4 is good too) 12...Kg8 13.Qf3 Bxd6 when Burgess considered 14.Nxd6 dubious because of 14...Nxe5 but actually 15.Qe4 Nc6 16.0–0–0 is very good for White. An important line is 16...Qe7? which surprisingly loses to 17.Rxd5! exd5 18.Qxe7 Nxe7 19.Bd3 when Black is helpless. Instead Black must try 16...f5 but 17.Qc4 Nf6 18.Be2 clearly is better for White. Thanks to Rybka for assisting me with these lines!

12.Bg5 f6 13.exf6 1–0
(after 13...Nxf6 14.Bxf6 gxf6, White picks up the loose piece with 15.Qxc5)

Love Not Returned

I can understand why I score badly when I am tricked into opening variations that I don't like and have not prepared. It's more surprising when the same happens in my favorite variations.

As far back as I can remember I have had a weak spot for the Kan variation: 1.e4 c5 2.Nf3 e6 3.d4 cxd4 4.Nxd4 a6!? Or actually for three sub-variations that can arise from it:

a) 4.Nc3 b5 5.Bd3 Bb7












b) 4.Bd3 Bc5 5.Nb3 Ba7












c) 4.c4 Nf6 5.Nc3 Bb4












I cannot recall exactly when I first saw saw these variations but they must have been among of the first Sicilian variations I came across.

Unfortunately these very variations seem to be weak spots in my chess understanding. I have played them in a few tournament games with catastrophic results and my blitz results have not given me any reason to give them a new chance.

I cannot say exactly why my score is so miserable. It's hardly the variations' fault as they according to MegaBase 2008 score very well for Black. It more likely is due to the fact that those bishops in the long diagonals - normally pointing towards White's king - tempt me to play for a king's attack with moves like ...h5 and ...Nf6-g4 when patient consolidation and queenside play are called for.

Maybe some day I will find the time to write a repertoire book based on these lines - just for myself. In the meantime I will start posting a few annotated games here.

What's the Kan Variation?

I recently noticed this upcoming book on the Sicilian Kan from Everyman:

Judging from its title I expect the book to examine the variation 1.e4 c5 2.Nf3 e6 3.d4 cxd4 4.Nxd4 a6 (Dia) which is what I call the Kan variation.

This expectation is even more reasonable because Everyman offers another older book, called 'Sicilian Kan' by Emms which I used to own and which treated this very line.

Yet I am not entirely certain as the publishers offer no moves and there is some confusion between the Sicilian lines Paulsen, Kan and Taimanov. I know for a fact that quite a few players and authors would call this 4...a6 line the Paulsen variation. And that indeed makes sense as Wilfried (not Louis as far as I know) Paulsen played the line a long time before Ilia Kan and with quite modern ideas. His stronger brother Louis probably contributed to the development of the line but himself preferred 4...Nf6 (generally followed by a quick ...Nc6) and 4...Bc5 (now known as the Basman-Sale variation).

This doesn't really explain why the 4...Nc6 variation (Dia) quite confusingly is occasionally called the Paulsen variation. The explanation may be that Louis Paulsen reached an important position from that variation repeatedly via the 3.Nc3 Nc6 4.d4 cxd4 5.Nxd4 move-order. Can it be that he after 2...e6 3.d4 cxd4 4.Nxd4 Nc6 feared 5.Nb5!?

So, what about the Taimanov variation then? Well, first of all it's worth noting that Taimanov early in his career played a lot of 4...a6 games. Then, around 1971 he started playing 4...Nc6 - usually followed by ...Nge7, ...Nxd4 and ...Nc6. This system - which now is rather rare - was what Taimanov himself called the Taimanov variation. What I don't really know is why the entire 4...Nc6 system is more and more frequently called the Taimanov system. It's a quite confusing situation and occasionally forces writers to distinguish between the Taimanov variation (where Black usually plays an early ...Qc7) and the 'pure Taimanov' (with ...Nge7).

Still not confused? Then take into account the hybrid variation below, which can equally well arise from 4...a6 5.Nc3 Nc6 and 4...Nc6 5.Nc3 a6 (Dia).


Is it a Kan, a Paulsen or a Taimanov? Well...I find 8129 games in MegaBase 2008, 1159 from the 4...a6 move-order and 6540 from the 4...Nc6 move-order (and obviously some from other move-orders too) ...so maybe it's a Taimanov.

Anyway, the move that interests me the most is 4...a6. That is also clearly the move that Hellsten has played the most, so I expect to buy the book as soon as it's available.

A Favourite Game

Although I consider myself a chess player, I have for the last 15 years been more of a children’s coach. On occasion I give lectures for grown-up players too, but generally I prefer a younger audience as children are somewhat more likely to pay back by becoming new chess stars.
One of my favourite lectures is based on this brilliant game, which almost perfectly illustrates the concepts of rapid development and forcing moves. In addition it’s a nice advertisment for the Rossolimo variation by its originator. I was really surprised when I couldn’t find the game in MegaBase 2006:

Rossolimo-Romaneko 1948
1. e4 c5 2. Nf3 Nc6 3. Bb5
This is known as the Rossolimo variation. White normally develops his pieces quickly and tries to build a pawn centre with a delayed c3 and d4.

3...g6

This is Black's most popular reply, but 3...e6, 3...Nf6 are respected alternatives. Even moves like 3...Na5 and 3...Qb6 are possible.

4. O-O Bg7 5. Re1 Nf6 6. Nc3
Two more popular lines are 6.c3 and 6.e5.
6...Nd4?!
After 6...O-O Black would be perfectly OK.
7. e5!
Now Black will not be able to castle for a few more moves.
7...Ng8 8. d3 Nxb5 9. Nxb5 a6?!
Naively expecting Black to retreat his aggressively posted knight.


10. Nd6+! exd6?
10...Kf8 11. Ne4 was not very tempting but still Black’s last chance to play on.
11. Bg5!
This is considerably more accurate than the obvious 11. exd6 (which should give roughly sufficient compensation for the knight but no more).
11...Qa5

Except for 11...Qb6, which comes more or less to the same, all alternatives lose material immediately.

12. exd6+ Kf8

13. Re8+!
A rook for a tempo! The point obviously is that the queen can do more damage in the e-file than the rook can.
13...Kxe8 14. Qe2+ Kf8 15. Be7+ Ke8

15...Nxe7 16. Qxe7+ Kg8 17.Ng5 loses even more quickly.

16. Bd8+!

Other checks are only good for a draw by repetition.
16...Kxd8 17. Ng5 1-0
The only way to stop Nxf7#, is 17...Nh6, which allows 18. Qe7#.