Biel Interzonal 1993 also had a large number of games, with players ranging from Kramnik and Anand to Gluckman and Kalesis.

Nunn looked for serious errors to negate 80 years of chess Development.

John mentioned Vidmar's book was completely useless! Vidmar overlooked about 90% of the errors. Nunn did run the games through the late 90s version of Fritz in blundercheck mode -- the first edition of his puzzle book was published in 1999, so Biel IZ 1993 was recent then too -- and then looked at the games "by hand" which turned up serious errors, eg. overlooking the loss of a piece and overlooking the win of that piece in return!

No mention about time trouble/pressure errors though. I presume the number of such errors was not relevant compared to the overall picture.

Tal in his 20s was as good as any GM today as was Karpov or Emmanuel Lasker or Rubinstein.

The difference in openings etc and the numbers of players and the stakes makes up the "inflation" but that will fall away. In fact there may well be an adjustment. I would say Kasparov's true rating was equivalent if not higher that Carlsen's and the same applies to all the other strong masters. Given that Steinitz and others required a lot of catching up.

But ironically it is because of the ideas adn games of the above last two we have such knowledge.

I think books as Kasparov's predecessors is more useful to read than anything by Nunn.

I DONT mean any of his books as of course hie puzzle books are great, and I have the book of his own games and his bio. partly and that he is or was a maths boffin, and he works with NZ's Murray Chandler for Gambit books...I think these books (like John Watson's) as long as we realise they are something we need to dispute or keep critical of. Watson similarly contradicted himself as there is no way to write about not learning basic chess ideas being wrong without learning the basic ideas.

The time situation, the psychology involved, these and many other factors make this kind of investigation slightly dubious.

Chess is won not by "perfect" moves. I've noticed that in many cases GMs - including many famous World Champions lose through relatively avoidable errors, or win through "unsound" attacks.

Often they simply become confused in complex positions.

<Nunn wouldn't risk a comparative analysis of today;s players. He might prove that e.g Kasparov and Fischer were really only 2100 players and then Chandler might have to sack him!>

Well, no, that wouldn't happen. That was one of Nunn's points -- he compared the Biel 1993 games against the 1911 Carlsbad games and found that the former were played at a much higher level. No one is going to take an engine and prove that Kasparov or Fischer played at a 2100 level.

I am sympathetic to Nunn's view. I haven't done the kind of analysis he did, but I have looked at the games from London 1851 v. Hastings 1895, and I can tell you the level of play at Hastings was much higher. Similarly if you compare Hastings 1895 and Zurich 1953.

On the other hand, analyses have been performed that purport to show that Capablanca was the most accurate player of all time, and on this website bridgeburner did an analysis that indicated that Schlecter-Lasker 1910 was played basically on the same level as Anand-Kramnik.

I have heard that kind of nonsense before. (Older players were weak, modern players could give them Knight odds ... modern players play/plays hundreds of points higher than the old guard.)

It is all smoke and mirrors and pure rubbish to me.

And I have annotated many of these games.

I have also done tests. Famous game, Anderssen-Lange. Gave it to about a dozen FM's over the years, so far none of them have solved it. (Done the same thing with about 30 of Morphy's games.)

And you wander why you got your head kicked in??
Whereas mine had chicks phone numbers, AJ'S had puzzles!

John mentioned Vidmar's book was completely useless! Vidmar overlooked about 90% of the errors.>

It seems to me that Nunn and bridgeburner can't both be right. If an average player at Carlsbad was about a 2100, a modern super-GM would be expected to win every game or almost every game against the field. Schlechter (who finished tied for second at Carlsbad) fell well short of that.>

I think it probably all depends on whether the old players were playing openings in which a lot of theoretical progress has been made since then. If not, there's no reason why the modern players should outperform them unless they're more intelligent, which I strongly doubt.

This is the age-old <nature versus nurture> question. What has Magnus Carlsen got that Gioachino Greco hasn't got?

<visayanbraindoctor> has posted extensively on the <nature> side (perhaps relying on some brain science but I haven't seen the details), positing that nobody will ever play better than Lasker, Capablanca and Alekhine, due to the limits of human brain capacity.

I've posted on the <nurture> side (citing research by De Groot and others about the pattern-based nature of chess skill), suggesting today's players are stronger because they have learned more patterns.

I think it probably all depends on whether the old players were playing openings in which a lot of theoretical progress has been made since then. If not, there's no reason why the modern players should outperform them unless they're more intelligent, which I strongly doubt.>

After implying that the subject has already been exhausted by the great minds (including your own), you say:

<It's not about the openings. Nobody ever shellacked Capablanca with an opening novelty. It takes a better understanding of the position.>

I don't think that's a valid refutation of my argument because:

1) You yourself recently admitted in another post that the Qxg4 line of the fried-liver attack (I believe) was considered to be bad but a single game of Kasparov vs. Shirov changed everything and now that variant is considered good for black. That means that opening theory progresses.

2) Knowledge of progress in opening theory constitutes an enormous advantage for modern players. Capablanca is relatively recent but say Carlsen or Giri or any of today's 2700 or 2800 players was projected back in time to the age of Philidor or even Steinitz and Lasker. It's extremely likely that he would trounce those players in an opening in which a lot of cumulative progress has been made or that didn't even exist at the time, like the Reti (I presume). You could argue, but who would bet on Steinitz in a game under those conditions?

3. I think your distinction between "a better understanding of the position" and "openings" is flimsy. Most openings these days are theory up to move 18 or so and the average number of moves in a tournament game is probably about 40 (just guessing), so the question is when does knowledge of the position break off from opening theory? Modern players often just memorize the first eighteen move or so (although there are rare innovations in the openings still) so the only positions that they have to have a feel for are from move 19 on, which would already be the mid-game back in the day.

Conversely, if you could somehow tip Steinitz off about the all innovations in opening theory (and I'm not saying it wouldn't be a daunting and time-consuming task for him to catch up), then there's no reason why Steinitz should be outplayed by Carlsen based on the latter's "better understanding of the position".

Position after Black's 32nd. move.

click for larger view

Boom! Bang! Boom! Bang! Boom! Bang! and three moves later...

click for larger view