FIDE World Chess Championship thread

[Uri Blass]

It is astonishing that the match ended without Anand winning a single game. This has never happened before in a world championship, as far as I know.

I was a bit disappointed with the 12-game format - I hope in future FIDE will consider a somewhat longer match.

--Jon

Capablanca beat Lasker w/o any defeat, Kramnik to Kasparov, and Lasker did it too, but I do not remember to whom now, but he was the defending champ.
Very clear victory. MC played like a computer, not making blunders, and allowing Anand to self destruct. An extremely boring match, that will leave very little to remember, from the chess point of view.
Miguel

why do you think 43.Nd6 is not a blunder?
+2.81++ 43.Nd2

[d] 8/1pk3p1/p3p2p/P1K2p2/2P1NP2/1P2n1PP/8/8 w - -

Maybe because
43.Nd6 is a typical move of players with rating above 3000

Houdini3 single cpu(tactical mode) with no tablebases that obviously has a rating above 3000 suggest the move Nd6 in your fen even after searching
more than 6,600,000,000 nodes

It considers the move 43.Nd2 at depth 28/59 for a long time so maybe it can find it in this iteration but it cannot search 6,600,000,000 nodes at tournament time control.

I have SF searching this position with 8 cores (for 8MNPS), and at depth 53 it gives Nd2 +2.90 (this score is constantly improving). I'll let it search more over the night and publish the results tomorrow.

I did not claim that Nd2 is not winning but only that Nd6 is not a blunder because a blunder is not only a bad move and a bad move that part of the top programs can make is not a blunder because it is certainly not easy for humans to understand why it is not the best.

[michiguel]

It is astonishing that the match ended without Anand winning a single game. This has never happened before in a world championship, as far as I know.

I was a bit disappointed with the 12-game format - I hope in future FIDE will consider a somewhat longer match.

--Jon

Capablanca beat Lasker w/o any defeat, Kramnik to Kasparov, and Lasker did it too, but I do not remember to whom now, but he was the defending champ.
Very clear victory. MC played like a computer, not making blunders, and allowing Anand to self destruct. An extremely boring match, that will leave very little to remember, from the chess point of view.
Miguel

why do you think 43.Nd6 is not a blunder?

Why should be? at minimum it secured a draw and a WC.

Miguel

+2.81++ 43.Nd2

[d] 8/1pk3p1/p3p2p/P1K2p2/2P1NP2/1P2n1PP/8/8 w - -

[kgburcham]

It is astonishing that the match ended without Anand winning a single game. This has never happened before in a world championship, as far as I know.

I was a bit disappointed with the 12-game format - I hope in future FIDE will consider a somewhat longer match.

--Jon

Capablanca beat Lasker w/o any defeat, Kramnik to Kasparov, and Lasker did it too, but I do not remember to whom now, but he was the defending champ.
Very clear victory. MC played like a computer, not making blunders, and allowing Anand to self destruct. An extremely boring match, that will leave very little to remember, from the chess point of view.
Miguel

why do you think 43.Nd6 is not a blunder?

Why should be? at minimum it secured a draw and a WC.

Miguel

+2.81++ 43.Nd2

[d] 8/1pk3p1/p3p2p/P1K2p2/2P1NP2/1P2n1PP/8/8 w - -

why should it be? because if you play it out with a strong program with strong hardware the score increases with each move if you play bm for both sides.

no not true, it did not secure a draw, blunders by both humans secured the draw.
no not true it did not secure a WC, blunders by Anand allowed the human opponent to secure the WC.
kgburcham

[Uri Blass]

It is astonishing that the match ended without Anand winning a single game. This has never happened before in a world championship, as far as I know.

I was a bit disappointed with the 12-game format - I hope in future FIDE will consider a somewhat longer match.

--Jon

Capablanca beat Lasker w/o any defeat, Kramnik to Kasparov, and Lasker did it too, but I do not remember to whom now, but he was the defending champ.
Very clear victory. MC played like a computer, not making blunders, and allowing Anand to self destruct. An extremely boring match, that will leave very little to remember, from the chess point of view.
Miguel

why do you think 43.Nd6 is not a blunder?

Why should be? at minimum it secured a draw and a WC.

Miguel

+2.81++ 43.Nd2

[d] 8/1pk3p1/p3p2p/P1K2p2/2P1NP2/1P2n1PP/8/8 w - -

why should it be? because if you play it out with a strong program with strong hardware the score increases with each move if you play bm for both sides.

no not true, it did not secure a draw, blunders by both humans secured the draw.
no not true it did not secure a WC, blunders by Anand allowed the human opponent to secure the WC.
kgburcham

1)The bm of houdini3 tactical that is a strong program is Nd6 for a long time.
Later it changes its mind to Nd2 but later it changes its mind again to Kd4 that is probably a draw.

Nd2 may be a winning move but the win seem to be too deep for most chess programs.

2)You say that 43.Nd6 did not secure a draw but blunders by both players secured a draw.
I do not agree and I see no blunder by anand after 43.Nd6

3)I do not agree and no blunder by anand after 43.Nd6 caused Carlsen to win the world championship.

[gordonr]

because if you play it out with a strong program with strong hardware the score increases with each move if you play bm for both sides.

The fact that you mention the need for a strong program and strong hardware highlights the complexity of this endgame. And therefore, in human terms it is not a blunder to get it wrong. We need to judge humans relative to other humans when assigning terms such as "blunder".

[michiguel]

It is astonishing that the match ended without Anand winning a single game. This has never happened before in a world championship, as far as I know.

I was a bit disappointed with the 12-game format - I hope in future FIDE will consider a somewhat longer match.

--Jon

Capablanca beat Lasker w/o any defeat, Kramnik to Kasparov, and Lasker did it too, but I do not remember to whom now, but he was the defending champ.
Very clear victory. MC played like a computer, not making blunders, and allowing Anand to self destruct. An extremely boring match, that will leave very little to remember, from the chess point of view.
Miguel

why do you think 43.Nd6 is not a blunder?

Why should be? at minimum it secured a draw and a WC.

Miguel

+2.81++ 43.Nd2

[d] 8/1pk3p1/p3p2p/P1K2p2/2P1NP2/1P2n1PP/8/8 w - -

why should it be? because if you play it out with a strong program with strong hardware the score increases with each move if you play bm for both sides.

no not true, it did not secure a draw, blunders by both humans secured the draw.
no not true it did not secure a WC, blunders by Anand allowed the human opponent to secure the WC.
kgburcham

That is not blunder. A blunder is a horrible and stupid mistake. Even if Nd2 wins and Nd6 does not, still not a blunder. No seeing a win 30 plies ahead is not a blunder by any stretch of the definition. If that is a blunder, it should be explained in few words. Why do you think it is a blunder?

Miguel

[kgburcham]

because if you play it out with a strong program with strong hardware the score increases with each move if you play bm for both sides.

The fact that you mention the need for a strong program and strong hardware highlights the complexity of this endgame. And therefore, in human terms it is not a blunder to get it wrong. We need to judge humans relative to other humans when assigning terms such as "blunder".

sorry yours does not meet my definition.
you are entitled to your own also.
the machines can tell us most of the time if the human has blundered.
of course it is very complex, that is why the humans blunder, they cannot see each line and cannot see the whole board, they are blinded by their human search methods which is limited.
I will not judge the score of a position by another human.
there are many positions by the biggest names in chess history that have been found to be incorrect using the machines.
this whole blunder check issue and OTB humans being offended by machines is only going to get worse as the machines approach 3500elo.

this is why I call the human move a blunder:
program move score 3.25
2800 Human move score .66
+3.25 43.Nd2
+0.66 43.Nd6
[d] 8/1pk3p1/p3p2p/P1K2p2/2P1NP2/1P2n1PP/8/8 w - -

after this line
43.Nd2 g5 44.fxg5 hxg5 45.Kd4 Nc2+ 46.Ke5
this score +3.46 and increasing
[d] 8/1pk5/p3p3/P3Kpp1/2P5/1P4PP/2nN4/8 b - -


after this line
43.Nd2 Nd1 44.Kd4 Nf2 45.h4 Ng4
this score +4.62++ and increasing
[d] 8/1pk3p1/p3p2p/P4p2/2PK1PnP/1P4P1/3N4/8 w - -

I welcome a correction
kgburcham

[kgburcham]

It is astonishing that the match ended without Anand winning a single game. This has never happened before in a world championship, as far as I know.

I was a bit disappointed with the 12-game format - I hope in future FIDE will consider a somewhat longer match.

--Jon

Capablanca beat Lasker w/o any defeat, Kramnik to Kasparov, and Lasker did it too, but I do not remember to whom now, but he was the defending champ.
Very clear victory. MC played like a computer, not making blunders, and allowing Anand to self destruct. An extremely boring match, that will leave very little to remember, from the chess point of view.
Miguel

why do you think 43.Nd6 is not a blunder?

Why should be? at minimum it secured a draw and a WC.

Miguel

+2.81++ 43.Nd2

[d] 8/1pk3p1/p3p2p/P1K2p2/2P1NP2/1P2n1PP/8/8 w - -

why should it be? because if you play it out with a strong program with strong hardware the score increases with each move if you play bm for both sides.

no not true, it did not secure a draw, blunders by both humans secured the draw.
no not true it did not secure a WC, blunders by Anand allowed the human opponent to secure the WC.
kgburcham

That is not blunder. A blunder is a horrible and stupid mistake. Even if Nd2 wins and Nd6 does not, still not a blunder. No seeing a win 30 plies ahead is not a blunder by any stretch of the definition. If that is a blunder, it should be explained in few words. Why do you think it is a blunder?
Miguel

not a blunder to you, I am ok with that.
to me it seems a blunder when a 2800 GM misses a 2 or 3 point move.
most of the time a two point move would be a loss against a program but against a human it might hold with opponents errors.
the fact is the 2800 human could not see the advantage, they are limited.
kgburcham

[Uri Blass]

Carlsen played at his level, with a stratospheric 2882 perf....

but actually 2200 if you factor in the blunders Carlsen made.
kgburcham

No, Carlsen's 2800+ rating already factors in his mistakes.

Yes of course I am aware of that.
My point is, my opinion is, something is wrong with our system.
The blunders should carry an extra penalty as I state above in my last post.
kgburcham

My opinion is there is something wrong with the opinion that insult humans and want to give them smaller rating.

I play chess in tournament and have fide rating near 2000.
If you reduce the rating of carlsen to 2200 you also reduce my rating
with no reason.

Mistakes are part of the game and it is not fair if they reduce the rating of the players unless you find some method when good moves that are not easy to find increase the rating of the players so the average remain the same when rating is more accurate but nobody invented a method that does it.

Uri

[Uri Blass]

It is astonishing that the match ended without Anand winning a single game. This has never happened before in a world championship, as far as I know.

I was a bit disappointed with the 12-game format - I hope in future FIDE will consider a somewhat longer match.

--Jon

Capablanca beat Lasker w/o any defeat, Kramnik to Kasparov, and Lasker did it too, but I do not remember to whom now, but he was the defending champ.
Very clear victory. MC played like a computer, not making blunders, and allowing Anand to self destruct. An extremely boring match, that will leave very little to remember, from the chess point of view.
Miguel

why do you think 43.Nd6 is not a blunder?

Why should be? at minimum it secured a draw and a WC.

Miguel

+2.81++ 43.Nd2

[d] 8/1pk3p1/p3p2p/P1K2p2/2P1NP2/1P2n1PP/8/8 w - -

why should it be? because if you play it out with a strong program with strong hardware the score increases with each move if you play bm for both sides.

no not true, it did not secure a draw, blunders by both humans secured the draw.
no not true it did not secure a WC, blunders by Anand allowed the human opponent to secure the WC.
kgburcham

That is not blunder. A blunder is a horrible and stupid mistake. Even if Nd2 wins and Nd6 does not, still not a blunder. No seeing a win 30 plies ahead is not a blunder by any stretch of the definition. If that is a blunder, it should be explained in few words. Why do you think it is a blunder?
Miguel

not a blunder to you, I am ok with that.
to me it seems a blunder when a 2800 GM misses a 2 or 3 point move.
most of the time a two point move would be a loss against a program but against a human it might hold with opponents errors.
the fact is the 2800 human could not see the advantage, they are limited.
kgburcham

Houdini3 also does not see the advantage with Nd2 so I see no reason to insult humans who did not see it.

there are clearly better example for blunders but for some reason you do not choose positions when all top programs can find the right move in less than 0.1 seconds and even 2000 players can understand easily why the move is bad and carlsen missed a win in the same game by 30.exd6
that I expect even myself to avoid in a game(I can explain it by the fact that carlsen cared first not to lose so did not search for a better move after being sure that 30.exd6 is not losing) but for some reason you do not choose this move as a blunder but the mistake that it is hard to find
when houdni3 tactical fails to find it even after more than 70,000,000,000 nodes and houdini3 not tactical fails to find it after more than 4,000,000,000 nodes

Note that I used 1 cpu so the results are going to be reproducable.