SPRAGGETT ON CHESS
WHO IS ”REALLY” THE BEST PLAYER OF ALL TIME?
It has always fascinated me that even though the 1st ever FIDE elo rating list was published only in 1971, many amongst us–mostly mathematicians I suppose— try to justify comparing the ELOs of players in the past such as Steinitz, Lasker, Capablanca, Alekhine and so on with players of today…EVEN THOUGH none of the former ever played a single elo-rated player or rated game in their lives!
Neither Alekhine (left) nor Capablanca (right) ever played a rated game in their lives!
It kind of reminds me of the accepted practice of MORMONS of converting the long dead relatives of their congregation over to the Mormon faith…even though some of them were devout catholics, protestants, jews or just plain athiests at the time of their death!
I suppose it really comes down to a matter of FAITH…in the ELO rating system. If you believe that ELO is not only a precise and accurate reflection of the performances of players but is also an excellent indicator of the relative strengths of said players…then why not? BUT we all know from experience–if not intuitively— that the ELO rating system is hardly perfect and certainly does NOT justify that kind of religious zeal. Besides, for many players of the past, there is simply not enough accurate data (ie. rateable games) to make accurate calculations. To go ahead and do so anyway would require a LEAP of faith! Yet, for some high-priests of mathematics, unfortunately, their faith often jumps ahead of the science…This book attempting to ”Elo-ize” the past masters was published in 1989 and has received scathing criticisms: ”Swill.” wrote Winter of ChessHistory. ”There are literally hundreds of problems with this book, in fact there are so many flaws it is hard to take it really seriously.” scoffed another.
A more recent example: Mark Weeks exposed on his excellent blog one other such leap of faith :
”The only resource that I know equivalent to Elo’s is Chessmetrics, where we find the Chessmetrics Player Profile: Saviely Tartakower. In the past I’ve been critical of the Chessmetrics methodology… because it draws too many conclusions on skimpy data, but it still manages to paint pretty pictures. On the ‘Ratings only’ section of the Tartakower page, he jumps from no.44 in the world on the September 1941 list (rating 2586) to no.10 in the world on the January 1946 list (rating 2688), despite being inactive between those months, as were most players.”
IT IS A TRUE MIRACLE!
Tartakower improved 102 ELO rating points without playing a single game!
AND THEN there is a problem in that the way the ELO is calculated has changed over time: it is easier to gain points today. For instance, today the ELO is calculated on a game by game basis; in the past it was calculated only AFTER the event was over–and using a rating average. What is the difference you may ask? It can make a SIGNIFICANT difference to a player’s rating! It is , from my observation, one of the two reasons why so many grandmasters have seen their ratings jump 50 points or more in the past few years!
THE OTHER problem has to do with rating team events: FIDE rates team match games just like ordinary tournaments, without any regard to colour alternation. This has lead to a lot of abuse by my fellow grandmasters, some of whom are able to convince the team owner to give them as many whites and as few blacks as possible. The result can often be a significant ELO gain…
In the 2010 Youngor Chinese Chess Division A team championship grandmaster Ni Hua played white a total of 16 times in 18 games!
In my view, these two problems go a long way to explain why the ratings of many of the top players have jumped so much in the past few years. In 2007, for example, my own rating was 2633, which was good enough for about 80th in the world. Today a rating of 2633 is not even in the top 140 in the world!
Has the chess elite simply got stronger, or , have the top players simply taken advantage of the changes that FIDE has implemented with regards to ratings? I think you know my answer…
Making things worse, many top grandmasters refuse to play in open tournaments any more in order to protect their ratings. Instead, they concentrate on playing team events and many are not adverse to manipulating their colour allocation. I think that FIDE should simply not rate team events where this abuse of colours takes place…the ELO rating system is based on some sort of normal distribution; lop-sided colour allocation is not a normal distribution.
IN ANY CASE, you won’t find anyone other than myself writing about these issues. Sort of tabu…the top players don’t want to attract attention–and just love all of those extra rating points– and the mathematicians who like to get published on chessbase (and elsewhere) would have a more difficult time to rationalize some of their already stretched out assumptions of the purity of the ELO system!
INFLATION , ANYONE?
This past week the german site CAISSA schach-chronik published an adjusted ”all time best” ELO list of the top players (above). I have no idea of the methodology behind the calculations, but I assume that it is as objective as any of the other so called experts out there! (Which does not say too much)
I am happy to see Fischer at the top of the list! I am less happy to see neither Alekhine nor Capablanca (or Lasker for that matter) making the top 20 list! But the ”adjusted” list does have the advantage that it has 3 times more deceased players than the current top 20 list! That rings in favour of objectivity, no? Unless, ofcourse, you believe that being dead puts you at a statistically significant disadvantage?!
Personally, I have no idea what is referred to by ”and our understanding of the game has developed since then (ed: late 1970’s)”. I hope he is speaking of his own personal chess development, because except for the openings, not much has changed!
As for Carlsen, no one doubts that he is the man of the moment, but he still has to go thru the motions and achieve all that Korchnoi achieved in his career. Probably the author (The Chess Mind) is not aware of the magnitude of Victor’s achievements….
How soon we forget!
Is the CMA dumbing-down your child?
The biggest private chess in schools program in central Canada , CMA (Chess and Math Ass.) has the policy of NOT teaching youngsters how to write down their chess moves when they play competitively! NOR does the CMA encourage the use of chess clocks when they play in tournaments…
This despite today’s youngsters being the most tech-savy generation in history!
It was recently admitted on the CMA message board–by Ex.Dir Larry Bevand– that this policy is considered best for the students in his program. Writing down the moves can take away fun of playing….(!)
This policy is in direct contrast to research studies that show that playing chess actually increases the child’s reading and writing skills!! Furthermore, the ‘expertise’ required to operate a chess clock is considered orders of magnitude less than that required to play video and computer games; or to work with any computer with WINDOWS environment.
SPRAGGETT ON CHESS