Friday, October 20, 2006

ICCF congress in Dresden

This year I participated for the first time at the ICCF (international correspondence chess federation) congress.

I had always dreamed that I would one day be invited and get a nice IM title. Alas, this will have to wait for some time, as I have too much work with the server and not enough time to play correspondence chess on the level required for an IM title. But I was invited to the ICCF congress to make a presentation about the server.

On Saturday my wife and I visited Dresden. Here are some photos (you can click on the photos to see them bigger):

The photos show: (1) me in front of the Zwinger, (2) Semper opera, (3) Hofkirche and Residenzschloss , and (4) Frauenkirche.

It was impressive to see that many of the buildings which had been completely destroyed in the second world war have been rebuilt using the original plans. This includes the Semper opera and the Frauenkirche.

On Sunday there was the opening of the congress with several speeches and votes. Here you can see me sitting at the computer during congress.

This picture was published by Uwe Bekemann at the congress website.

I did the presentation of the server together with Alan Borwell and Gerhard Binder. I read on TCCMB that some people had problems with the internet access in the hotel. I did not have any problems. In fact, there was wireless access in the hotel, and it must have been broadband, too. I was able to make a backup of the server and run a report against it - I could then present the very latest server statistics.

In the evening there was the opening banquet. Six correspondence world champions were present:


(1) From left to right: Gert Jan Timmerman, Ivar Bern, Fritz Baumbach, Tunc Hamarat, Horst Rittner, and Grigorij Sanakojew, (2) Gert Jan Timmerman, Ivar Bern and myself discussing the gambit match between Timmerman and Umansky, (3) congress postcard with autographs of all six world champions present.


Wednesday, October 11, 2006

Kramnik-Topalov game twelve preview

In game ten Topalov played the Catalan again, and Kramnik again reached a better position. I won't do analysis of concrete moves today - there is an excellent analysis of the game at, where Mikhail Golubev replaces Peter Svidler, who is now playing at the European Club Cup. I think Topalov is somehow playing the Catalan when he wants to draw, and the Slav when he wants to win with Black.

Game twelve is the last game in the match, and Kramnik has White. While I generally think Topalov should play for a draw and not risk too much, I can only recommend the Slav to him.

Out of Topalov's four games with Black so far (not counting the forfeit), he played the Slav once, and the Catalan three times. Topalov has scored one point out of the Slav game, and half a point ouf of the three Catalan games together. He also had a good position out of the opening in the Slav game, but three times a worse position in the Catalan games. Enough said.

I hope for an exciting game twelve and am not thinking about the tie breaks yet - after all we have seen already five decisive games in the match. We haven't seen so many decisive games in a world championship match for some time.


Monday, October 09, 2006

Topalov-Kramnik game eleven preview

Kramnik equalized in game ten. Game eleven is Topalov's last White game in the match. He has to create winning chances with White to avoid a Brissago-like last game situation: facing Kramnik in the last game with Black is tough, as Peter Leko can tell you. Topalov should be used to this situation by now, he was under pressure to generate good chances in practically every one of his White games.

Lets have a look at game nine, where Topalov managed to break through Kramnik's black wall for the first time in the match:

1.d4 d5 2.c4 c6 3.Nf3 Nf6 4.e3

Topalov plays the same move order as in game seven. He does not seem to mind transposing to the same queens gambit accepted line we saw in game seven. Or maybe he wants to play another Meran?


Kramnik deviates from game seven. As I said in my game ten preview, Kramnik had so far played like a wall, and a wall should stand still. I think Kramnik should have played 4..e6 just as in game seven.

5.Nc3 e6 6.Nh4

I think this line is dangerous for Black. There are a lot of different ways to play this line for White.

6..Bg6 7.Nxg6 hxg6 8.a3

The main lines here are 8.g3 and 8.Bd2. I think the idea of this move is to have the option of playing c5.

8..Nbd7 9.g3

This is the new move. Topalov said in the press conference:

I used a good novelty, but one cannot say it gives White a decisive advantage. The move 10.f4 is invented by my second Vallejo – he specializes on this variation and worked a lot in Elista. So, you see the fruits of our work!

9..Be7 10.f4

The problem for Kramnik after this novelty is not that he has to calculate many forced variations, but that he is faced with a completely new concept in this opening. He first has to find a good plan for Black, which is difficult over the board, especially when you know that your opponent and his team have probably spent a lot of time analyzing everything at home.


I don't like this move. Susan Polgar writes

Why not to wait with this trade until White moves the Bishop from f1?

Actually, I am not sure if Black has to give up the center at all.

Mihail Marin proposes

Giving up the centre so easily is a risky decision. 10...a5 deserves attention, with the idea to answer 11.c5 with 11...b6.

If somebody plays f4 against me, and I have an open h-file available, I am always thinking about capturing the g3 pawn. But 10..g5? does not work, because after 11.fxg5 Nh5 12.g6 is strong, and after 10..Nh5 White can just defend with 11.Qf3.

Maybe Black can really just play the waiting game Susan suggests:

10..O-O and I think Black does not need to fear 11.c5 b6 12.b4 a5 13.Bd2 axb4 14.axb4 Rxa1 15.Qxa1 Qa8. If White moves his bishop, say 11.Bd3 Black can play 11..dxc4 and win a tempo upon the game. Maybe White will start rolling the pawns on the other side of the board though, 11.g4 or 11.h4.

Of course Black could even consider castling long in this position - you will have the rook h8 still on the open h-file. Or maybe Black should wait with castling, and decide wether to castle short or long only after White has revealed a bit more of his plan?

You can see Kramnik's dilemma - deciding between all these lines over the board is a difficult decision at home, and much more so over the board.

I also noticed, that the plan with attacking g3 may be not so bad if Black had put his Bishop on d6 on the ninth move instead of on e7, e.g.

9.g3 Bd6 10.f4 Nh5 11.Qf3 g5!? or

9.g3 Bd6 10.c5 Bc7 11.f4 g5 12.fxg5 Bxg3 13.hxg3 Rxh1 14.gxf6 Qxf6 15.Qe2 O-O-O 16.Bd2 Rdh8 17.O-O-O e5 with a complicated game.

The game continued

11.Bxc4 O-O 12.e4 b5?

After Black has given up the center, he should think about attacking White's center immediately - with 12..c5 or 12..Nb6 and 13..c5.

13.Be2 b4 14.axb4 Bxb4 15.Bf3

Here Kramnik had the last chance to play 15..c5, when he could at least have caused complications after 16.e5 cxd4. The game continued

15..Qb6 16.O-O

and Kramnik had a terrible position.

He had to say this in the press conference:

It is difficult to play well in positions such as the one I got today. Topalov’s novelty turned very strong – at least for one game. I didn’t manage to find adequate response to it. Already 12…b5 was played because I could not find anything satisfactory, especially bearing in mind that it was all opponent’s preparation. After the opening, I continued resisting because it was inconvenient for me to resign that early, but the game was basically decided by the move 17.

In game eleven, I think Topalov will try the same move order for the third time. Even if Kramnik or his seconds find a solid plan against Topalov's novelty at home, I think he should avoid the line after 4..Bf5 5.Nc3 e6 6.Nh4. Topalov may just have another new plan in that line and surprise Kramnik again.

It should be much safer for Kramnik to play 4..e6 like in game seven. I think the queens gambit accepted was not that dangerous, and Topalov would go for another Meran after that. As I said in some of the other previews, the sharp lines where White plays Qc2 and g4 are still left to be explored, and would certainly bring us another exciting game.

I think neither of the players will switch to 1.e4 at this point in the match - there are too many unexplored options. They will concentrate on trying to show an advantage for White in the setups that have already been played.


Saturday, October 07, 2006

Kramnik-Topalov game ten preview

With Black Kramnik played like a wall so far in the match. The wall stood still, and when Topalov threw something at it, it came back to him. In game nine however, after

1.d4 d5 2.c4 c6 3.Nf3 Nf6 4.e3

Kramnik deviated from game seven (4..e6) and played


allowing Topalov to play the line

5.Nc3 e6 6.Nh4

where Topalov had prepared a new line and won in a very convincing way.

The wall had moved and collapsed. I don't understand why Kramnik did not continue to play like a wall and repeat the line from game seven. I think this was a bad idea, and he should have repeated the line from game seven. I will come back to this for the game eleven preview. If you have any idea why Kramnik played like this, please let me know - you can send me feedback at

Now Kramnik faces a difficult situation. He is behind for the first time in the match, and there are only three games to go. It will be crucial to create winning chances with White in game ten.

But first lets have a look at game eight, and what this means for game ten:

In game eight Topalov abandoned the Catalan and switched to the Slav. While the Catalan had not been a desaster, Kramnik was able to steer the games away from complications and towards the technical positions he plays so well. Therefore I think Topalov made a wise decision by switching to the Slav.

1.d4 d5 2.c4 c6 3.Nf3 Nf6 4.Nc3 e6

Kramnik with Black played 4..dxc4 in this position in games two and six, here.

5.e3 Nbd7 6.Bd3 dxc4 7.Bxc4 b5 8.Be2

Topalov played 8.Bd3 in this position in game four.

8..Bb7 9.O-O b4 10.Na4 c5

Topalov has prepared a rare line. Now 11.Nxc5 Nxc5 12.dxc5 Bxc5 13.Qa4+ Ke7 is ok for Black.

11.dxc5 Nxc5

Now 12.Nxc5 Bxc5 would transpose to the last note.

12.Bb5+ Ncd7 13.Ne5 Qc7 14.Qd4 Rd8 15.Bd2

Hannes Langrock at proposes the greedy 15.Qxa7 Bd6 16.f4 O-O 17.Nxd7 Nxd7 18.Bd2 and asks if Black has enough compensation. I think yes, for example 18..Nf6 (Black has to make room for the queen) 19.Rfc1 (or Rac1) Qe7 and the White pieces on the queenside have gone astray. Black will make moves like Rd8-a8-a5, Rf8-d8 or a8, Nf6-d5 or e4.


This is Topalov's new move.

16.Bc6 Be7 17.Rfc1

Here 17.b3 or 17.Rac1 are alternatives.

After 17.b3 O-O 18.Bxd7 Nxd7 19.Nxd7 Black can choose between 19..e5!? 20.Qxe5 Qxe5 21.Nxe5 Rxd2 with a very nice bishop pair for the pawn (Hannes Langrock) or 19..Bc6, which is also fine.

17..Bxc6 18.Nxc6 Qxa4

Now Kramnik played 19.Nxd8 and lost the endgame after 19..Bxd8 20.Qxb4 Qxb4 21.Bxb4 Nd5 22.Bd6 f5.

There are some interesting alternatives on move 19:

Stefan Löffler looks at 19.Bxb4 e5 20.Qh4 Bxb4 21.Nxd8 O-O 22.Nxf7, but after 22..Rxf7 23.a3 e4 24.axb4 Qxb4 25.Rxa7 Qxb2 I think Black's two knights may be better than White's rook.

But very interesting is the sharp

19.Nxe7 Kxe7 20.Bxb4+ Ke8 21.b3 Qb5 22.Rd1

Now Vladimir Below analyzes

a) 22..Qe5 23.Qxa7 Nd5 24.Ba5 Rc8 25.e4 Qxe4 26.Rd4 Qf5 27.Re1 "and it is difficult for Black to untie".

b) 22..Qb6 Peter Svidler thinks "sooner or later the extra piece will tell", but Belov gives 23.Qc3 "and it is not easy to find the way for Black to unpin, while White’s attack develops naturally".

c) I think maybe 22..a5 is the best try to untangle, e.g. 23.Ba3 Qe5 and now

c1) 24.Qc4 Nd5 25.e4 Nb4 or

c2) 24.Qa7 Nd5 25.Rac1 f5

but this is all very complicated. Black's problem is that he cannot castle, because the King has already moved, and that all his pieces are in various pins. But can White do something before Black manages to untangle?

I doubt that we will see this line again.

Interestingly, therefore Kramnik now faces the same situation that Topalov has faced a couple of times in the match: he is trailing in the match, and if he plays 1.d4 could face the mighty Slav again.

So maybe it will be Kramnik's turn to play something sharp after all? Another Meran, perhaps? What will happen if Kramnik decides to play 1.e4? Would Topalov play the Najdorf? But now for Topalov a draw with Black would be fine. As I briefly discussed in my game one preview, he usually plays the Berlin in these situations, however against Kramnik, who is a Berlin player himself, he may have prepared another line - maybe the Marshall.

Lets hope for three more interesting games.


Friday, October 06, 2006

Topalov-Kramnik game nine preview

In game eight Topalov switched to the Slav (as I predicted in all my previews :-)and scored his first win. Because of the defaulted game five the score is now equal at 4-4. There are only four games left and we can hope to see an exciting finish to the match.

But now lets have a look at game seven, and what conclusions we can draw for game nine:

1.d4 d5 2.c4 c6 3.Nf3

In game two Topalov played 3.Nc3 Nf6 4.Nf3 and in game four 3.Nc3 Nf6 4.e3 e6 5.Nf3.

Nf6 4.e3

In game six Topalov played 4.Nc3 here, transposing to game two. Kramnik could play 4..Bf5 here, but he would like to transpose to game four:

4..e6 5.Bd3

Topalov could have transposed to game four with 5.Nc3, but he plays a clever move order. If Kramnik now continues 5..Nbd7, White has the alternative 6.O-O Bd6 7.Nbd2, when Black cannot play dxc4, b7-b5-b4 with tempo. White will then be able to get in e4. Therefore Kramnik played

5..dxc4 6.Bxc4 c5

when a position from the Queens gambit accepted is reached (the usual move order to reach this position would be 1.d4 d5 2.c4 dxc4 3.Nf3 Nf6 4.e3 e6 5.Bxc4 c5, taking one move less).

7.O-O a6 8.Bb3

In the worlchampionship match between Kramnik and Kasparov, Kasparov had abandoned the Grünfeld, and this position was reached twice. Kramnik played 8.dxc5 in game four, while game six saw 8.a4 Nc6 9.Qe2 cxd4 10.Rd1.

Please note that the common line 8.a4 cxd4 9.exd4 Nc6 10.Nc3 Be7 11.Re1 O-O would transpose to the game after 12.Bb3 - however White seems to have more useful moves in this position and almost never plays 12.Bb3.

Because of this observation I cannot believe that Topalov's novelty in the game can be that dangerous. I think the idea was more to get a playable position with fighting chances that Kramnik could not possibly have prepared.

8..cxd4 9.exd4 Nc6 10.Nc3 Be7 11.Re1 O-O

We have reached a typical position with an isolated queen pawn. This position can also be reached from other openings by transposition.

It is interesting to note that positions with isolated queen pawns have been played in many, if not in all worldchampionship matches. Sometimes the handling of this type of position was even decisive. The first world champion Steinitz developed the strategy of playing against an isolated pawn and used it to good effect against Zukertort. Later Botvinnik developed attacking plans for the side with the isolated pawn.


This move is new although the same position was reached in a game Gershon-Papatheodoru, Korinthos 1998 by transposition. The moves 12.a3, 12.Bf4 and 12.Bg5 are all more common.

12..Bd7 13.Ne5 Be8

I think the position of the bishop is a bit artificial here, because the rook on f8 has not moved yet. Furthermore the bishop moves again a few moves later. Maybe Kramnik could just play 13..Nb4 here. I think the loss of the pair of bishops after 14.Nxd7 would not pose Black real problems. In the game Black exchanges this bishop against a knight anyway, so 13..Be8 looks just like a move lost.

14.Be3 Rc8 15.Rc1

This rook moves to d1 a few moves later, maybe it could have gone there at once?

15..Nb4 16.Qf3 Bc6 17.Qh3 Bd5 18.Nxd5 Nbxd5 19.Rcd1 Rc7 20.Bg5 Qc8

Again the placement of Black's pieces looks artificial to me.

21.Qf3 Rd8 22.h4 h6 23.Bc1

As several commentators have shown, after 23.Bd2 White could have developed a nice initiative.

What does this mean for game nine? Topalov has a difficult decision - should he go for an all out attack but risk losing? On the other hand, if he plays too calm and makes a quick draw, Kramnik will have one more White game remaining. This can be very dangerous against Kramnik. In the last world championship match, Peter Leko was leading with one point before the last game - but Kramnik had White and won.

Topalov does not have so many options left for the remaining two White games. If he repeats the line from game seven I think Kramnik would handle the position better. After 1.d4, I think this just leaves the Meran as a good option - and then it seems time for the sharp lines where White plays Qc2 and g4.

Or is this to risky? Maybe Topalov should switch to 1.e4 now. Somehow after all these Slav games the Petroff may not look so bad any more.


Tuesday, October 03, 2006

Topalov-Kramnik game seven preview

Thanks to every reader who made me aware that they are actually changing the order of the games in the second half of the match (you can send me feedback at Therefore game seven will see Topalov with White again - actually, because of the defaulted game, for the third time in a row.

So lets have a look at the opening of game six, and see what this means for game seven.


1.d4 d5 2.c4 c6 3.Nf3 Nf6 4.Nc3 dxc4 5.a4 Bf5

Topalov played


In the second game he played 6.e3, but got nothing out of the opening - although he managed to conjure an attack later (see my game four preview). Now Kramnik did not go for 6..Nbd7, but after

6..e6 7.f3

he didn't go for the sharp 7..Bb4 8.e4 Bxe4 either. Instead he went for the side line


and after


he surprised Topalov and played the rare move


I have some games from Smyslov with this move in my database, but it has not been played very often. There are more games with 8..cxd4 9.exf5 Bb4 with some unclear complications.

In the game, Topalov did get an endgame that looked a little bit better for White, but Kramnik never was in any real danger:

9.Be3 cxd4 10.Qxd4 Qxd4 11.Bxd4 Nfd7 12.Nxd7 Nxd7 13.Bxc4 a6 14.Ke2 Rg8

and so on.

Can Topalov improve White's play?

A) Susan Polgar shows that after

9.d5 exd5 10.Nxg6 hxg6 11.e5 Nh5 12.Nxh5

Black can (instead of 12..Ng3 13.Bxc4 Nxh1? 14.Nf6+ Ke7 15.Ng8+) just play


and White has only managed to weaken his own position.

B) Of course White could try to avoid the exchange of Queens with


but is that really an improvement? I have one game Pelletier-Smyslov, Zürich 1998 in my database, which continued

11.Nxc6 bxc6 12.Bxc4 Qa5 13.Qe2

and a draw was already agreed. Indeed, an advantage for White is not in sight, and if Black plays


maybe it is already White who should think about equalizing.

I think Kramnik did a fine homework here, this rare line seems to be safe enough for Black. I don't see any good way for White to improve. Maybe this will be played more often by Black in the future.

I think Topalov will have to return to 4.e3 and play another Meran, as I discussed in my game six preview.


Monday, October 02, 2006

Kramnik-Topalov game eight preview

Update: This is now game eight preview, because of the changed order of games in the second half of the match.

I am glad we have more chess. I already had nightmares of a certain match between Karpov and Timman.

Today Topalov played the Slav line with 6.Ne5 and 7.f3, but did not get real winning chances. I will come back to that line for the preview of game eight.

Thanks again to the readers who sent me feedback (you can send me feedback at

Mark Bowen from Jamaica sent me the following question:

I'd also love to hear your thoughts on the whole design of one's opening repertoire and discussion of the player's overall styles and how their repertoires take advantage of this.

For your own opening repertoire of course it depends on how strong a player you are. If you are an average club player and you don't have a lot of time for studying openings, I would recommend to use some of the repertoire books that are out there, and spice it up with some lines that are either rare, for example from an SOS book from New In Chess, or some main lines that you are particularly interested in.

Of course I am interested in opening theory more than in my over the board results, so I am reading the hardcore books about the lines I am interested in, but I am regularly forgetting my own analysis, and have big holes in the openings I am not so interested in. So I cannot really recommend this, unless you are doing this like me just for fun, not for results.

With regards to the styles of Kramnik and Topalov, and how this affects their openings in this match, Kramnik excels in technical positions and middle games without queens, while Topalov plays his best in dynamic or even unclear positions that offer a lot of different ways for fighting.

But today's opening preparation at world class level goes so deep, that you cannot just choose your openings by style. For example, Topalov may in principle like the Petroff positions with White. While the Petroff has a drawish reputation, usually you keep enough pieces on the board for a full fight, and White often has several ways to anvance his center pawns (c3,d4). But the opening is analysed very far, and you have to come up with some strong new moves to actually get to a position where you can pose Black some problems and get to this fight. The same goes for the Sveshnikov - in principle this is a very interesting and dynamic position with a lot of different ways to fight for both sides. But there is so much analysis on this opening that you have to find something new to pose any problems.

That is why it is an advantage if you can play both 1.e4 and 1.d4, especially in a match where the openings can be repeated several times. If Topalov has not found enough strong moves against the Petroff, maybe he has some new ideas in the Slav. That is why Kramnik has learned to play 1.e4, too, even if it fits less to his style than 1.d4. That is why Leko prepared 1.d4 for his match against Kramnik, although he almost always played 1.e4. And he was glad he did, after banging his head against Kramnik's Petroff.

But one new move can make a difference. Actually I have to revise my so far unused preview of game five a little bit, because Peter Svidler showed a move that I did not consider. After

1.d4 Nf6 2.c4 e6 3.Nf3 d5 4.g3 dxc4 5.Bg2 Nc6 6.Qa4 Bd7 7.Qxc4 Na5 8.Qd3 c5 9.O-O Bc6 10.Nc3 cxd4 11.Nxd4 Bc5 12.Rd1

the alternative mentioned by Topalov in the press conference

12..Qxd4 13.Qxd4 Bxd4 14.Rxd4 Bxg2 15.Kxg2 Nc6 16.Rd1

seems to be equalizing after all. I did only consider 16..Ke7, when White can play 17.b3, however Svidler just castles:

0-0 17.Bg5 Rfd8 "and it is hard to believe Black's problems are not temporary" (Svidler).

Svidler also has the following explanation why Topalov did not play this line, and why it looked as if Topalov did not prepare this game well:

My feeling, as I sat watching the game live, was that Veselin came to the board with the idea to trade as many pieces as possible as soon as possible and make a draw, but couldn't bring himself to do it - the whole concept is, as we've seen on numerous occasions, alien to him.

I think that makes a lot of sense. As you see, one move can make a difference.

It will be interesting to see if this line is repeated in game seven. Maybe this position is not enough even for Kramnik. Or Topalov, if this concept is so alien to him, will choose another line himself?