Prof Elo Ratings of Chessplayers: Past&Present - Who was the best?

JamieDelarosa

Based on the seminal work by Professor Arpad Elo, the inventor of the Elo Rating System, I will open this discussion with a list of players who, from 1855 to 1975, rated as the strongest player in the world.

These data come from Prof. Elo's book, and particularly his graphs and tables.  The main graph ("Lifetime Ratings, Selected Grandmasters" Pages 88-89), which is partially reproduced on the dustcover of the hardback version of his book, provides the following information.  To obtain "Elo Ratings" for masters who pre-dated their invention, Prof Elo meticulously back-calculated tournament and match results.  After publication, the database was found to have some errors, however, they were not serious enough to compromise the entirety of his work.

 

Adolph Anderssen of Prussian Silesia won the famous London (knockout) Tournament of 1851.  Elo's estimate of Anderssen's rating, normalized to 1975, was approximately 2595 (these are estimates from reading the graph).

1855-1858, 2600 (peak), A. Anderssen

1858, 2690 (peak), P. Morphy

1859-1866, A. Anderssen

1866-1891, 2650 (peak), W. Steinitz

1891-1914, 2720 (peak), E. Lasker

1914-1927, 2730 (peak), J. R. Capablanca

1927-1942, 2685 (peak), A. Alekhine

1942-1954, 2730 (peak), M. Botvinnik

1954-1961, 2700 (peak), V. Smyslov

1961-1963, 2680 (peak), T. Petrosian

1963-1975, 2780 (peak), R. Fischer

Notably missing from Elo's "selected grandmasters" are M. Tal and B. Spassky.  The exclusion of Tal (best 5-year average = 2700) is understandable given his health problems and erratic results, though I would suggest he was the top dog in 1959-61.  Why Spassky (best 5-year average = 2680) was left out is anyone's guess, other than he may never have been the strongest player in the world, given Fischer's rating after 1963.

Fischer's curve is the most anomalous of all.  His rise was meteoric, and showed no decline.  Other's curves were, at times affects by layoffs due to wars, etc.

JamieDelarosa

Here is another graph, showing the succession of strongest players.

Interestingly, it begins with Howard Staunton, omits Adolf Anderssen and Morphy, includes Victor Korchnoi(!), but does not include Mikhai Tal or Tigran Petrosian. (From chessbase.com, 2010)

These data present "performance ratings."  Fischer's peak at 2895.

JamieDelarosa

He does show up again, 20 years later, at the 2600-2650 level (Spassky II).

The question becomes, had he continued to play, would he have continued to improve?

Fischer certainly felt he was the best player in the world after the 1962 Stockholm Interzonal.  Hence, his frustration with the Soviet collusion in the Candidates Tournament the next year.

yureesystem

I think Morphy rating is too high, Alekhine understanding of chess was much more deeper than Morphy.

fabelhaft

Elo's back-calculated ratings look strange to me. Placing Morphy more or less equal with Petrosian more than hundred years later is just difficult to believe whatever one is supposed to measure. If it is dominance in relation to the other players of one's time, as Elo ratings should measure, Morphy was far ahead of everyone in his own time, while Petrosian may have been #1 with the slimmest of margins for the shortest of time. And if it would be playing strength, that Elo ratings are not supposed to measure, Petrosian and the other top players of the 1960s surely played on a much higher level than Morphy, for natural reasons.

Houdini_Genius

Those player were real brilliant. . in a sense incomparable to each other. But they dominated the world in their era....Legends. ..!!!

JamieDelarosa

I understand the issue revolving around Morphy.  His active period of play was so short, many of his games were played at odds or against teams of players, and the further back in time you go the more uncertain you are about the accuracy of the calculations.

The strength of the Elo system is that it is strictly based on results - wins, draws, losses.

Folks who use this statistical rating method, have been able to back-calculate rankings in other sports, such as tennis and American football.  Those results generally comport with the wisdom of the ages passed down.

JamieDelarosa

(EDIT - This post was in response to a troll who has since been banned)

Who let the dog out?!  LOLZ

What do you want to know, sweetie?

I suggest you put your hands on your knees, bend over as far as you can, and kiss your ass good bye, silly little boy.

MrDamonSmith

Alrighty then. That was out of the blue.

JamieDelarosa

That guy has an axe to grind with the ownership here.  I don't know the backstory, but he makes violent threats.  I'd turn it over to local law enforcement or the FBI.

I have zero tolerance for morons like him.

VierKazen89

I think Kasparov, Fischer and Karpov were the best. Still I think you cannot compare chessplayers  from different era's. Imagine Bobby Fischer working things out with a computer. I played games over from Fischer and they look so simple , so logical and they were a delight to play over.

JamieDelarosa

Professor Elo admitted it was difficult to compare players if different eras, but that looking at statistical performance was the best way to try.