Aging in Chess

e5master

Hyacinthe Henri Boncourt was the 3rd best player at the age of 73.
!!!

Hyacinthe Henri Boncourt was the 3rd best player at the age of 73. !!!

e5master

Viktor Korchnoi was still among the top 10 players at the age of 59.

RIHK1992LWK

born 1941, I know what means t be older in chess. But still I love Blitz -lighting-!!!!!!!!!!!!!

playingsomerapid

edited

@ebk1976 said in #31:

OFC this is not a completely decisive conclusion, but it is an observation from which we can draw conclusions. While they could be much more precise, they still can have value. OFC the ratings have also changed, but no one thinks that Topalov for instance is dropping rating every time he plays because of rating definition.
Thanks for sharing your thoughts!

Well, there are two major things which make the analysis very difficult:

ELO is only "locally" a measure of strength, adjusting for playing era (elo deflation/inflation, competition), activity, (at amateur level also the geographical region)

One of the flaws of elo as an objective measure without controlling for other variables becomes apparent when looking at the ELO-preserving strategies in the current era (even more relevant below the absolute top). That seems to be either:

i) barely playing any tournaments
ii) playing Senior Chess events against their own chess group

It's easy to note that those strategies are not ideal for actually preserving the playing strength. So player A has played 150-200 games in international opens during the last two years and lost 100 points with K= 10. Player B has not played at all. With the pure elo comparison player B appears to be aging better but it is not the case.

The second issue is that even if elo was somehow an objective measure for your question (let's continue with your example of Topalov and agree that he has indeed declined) the explanation is rather unclear. I would be surprised if he performed the same high quality and volume of training, has played in such high profile tournaments and had the same motivation as in early 2000s.

Even if we put this factor aside (which is hardly justifiable IMO), there are other explanations left, an important one being the general health (stamina, possible conditions).

So overall I think that with your chosen group you get very limited data which is not as clean as it initially seems and little control over many different variables which all influence performance to conclude that it is related to the aging of the brain.

To be fair, even a semi-serious study on this matter would require an effort which can't be expected in a context of a blogpost (involving clearer definitions of things to measure, more data, adjusting for data drift and different other quirks of the elo system, literature research, interviews with the inner circle of the players or the players themself to figure out their training regiment over the years and much more).

@ebk1976 said in #31: > OFC this is not a completely decisive conclusion, but it is an observation from which we can draw conclusions. While they could be much more precise, they still can have value. OFC the ratings have also changed, but no one thinks that Topalov for instance is dropping rating every time he plays because of rating definition. > Thanks for sharing your thoughts! Well, there are two major things which make the analysis very difficult: 1) ELO is only "locally" a measure of strength, adjusting for playing era (elo deflation/inflation, competition), activity, (at amateur level also the geographical region) One of the flaws of elo as an objective measure without controlling for other variables becomes apparent when looking at the ELO-preserving strategies in the current era (even more relevant below the absolute top). That seems to be either: i) barely playing any tournaments ii) playing Senior Chess events against their own chess group It's easy to note that those strategies are not ideal for actually preserving the playing strength. So player A has played 150-200 games in international opens during the last two years and lost 100 points with K= 10. Player B has not played at all. With the pure elo comparison player B appears to be aging better but it is not the case. 2) The second issue is that even if elo was somehow an objective measure for your question (let's continue with your example of Topalov and agree that he has indeed declined) the explanation is rather unclear. I would be surprised if he performed the same high quality and volume of training, has played in such high profile tournaments and had the same motivation as in early 2000s. Even if we put this factor aside (which is hardly justifiable IMO), there are other explanations left, an important one being the general health (stamina, possible conditions). So overall I think that with your chosen group you get very limited data which is not as clean as it initially seems and little control over many different variables which all influence performance to conclude that it is related to the aging of the brain. To be fair, even a semi-serious study on this matter would require an effort which can't be expected in a context of a blogpost (involving clearer definitions of things to measure, more data, adjusting for data drift and different other quirks of the elo system, literature research, interviews with the inner circle of the players or the players themself to figure out their training regiment over the years and much more).

ebk1976

@ingenuity9
Thanks, keep up the good work!
@QueenRosieMary
Thank you so much, really appreciate it!
@e5master
There are many players who were very good at an old age, I am not saying there were not. :D
@RIHK1992LWK
Keep it up!!
@playingsomerapid
You have some great thoughts, if you would write a blog about this subject, I would certainly check it out! However, I don't think anything you said invalidates my conclusion, as I said earlier.

@ingenuity9 Thanks, keep up the good work! @QueenRosieMary Thank you so much, really appreciate it! @e5master There are many players who were very good at an old age, I am not saying there were not. :D @RIHK1992LWK Keep it up!! @playingsomerapid You have some great thoughts, if you would write a blog about this subject, I would certainly check it out! However, I don't think anything you said invalidates my conclusion, as I said earlier.

Tainck

I really want to stop aging now.

Griffondort

Moro was ill. For me this is the explanation for his decrease so early

Ordans81

Very interesting blog post. Does anyone want to share advices on how to keep the brain in good shape even while getting older?

happychappie

"We can see that, as following science, chess players begin to falter around age 40."

Where is the science? So many factors not addressed, correlation doesn't equal causation. Drops in ELO cannot be directly linked to cognitive decline in the way you've described. These GM's are facing opponents who are on average much better than their average opponent was a decade ago, it doesn't make sense to just use ELO score in this way, let alone the multitude of other factors like social stressors, diet, sleep, etc. This article is genuinely misleading and poisonous to any impressionable young person reading this.

"We can see that, as following science, chess players begin to falter around age 40." Where is the science? So many factors not addressed, correlation doesn't equal causation. Drops in ELO cannot be directly linked to cognitive decline in the way you've described. These GM's are facing opponents who are on average much better than their average opponent was a decade ago, it doesn't make sense to just use ELO score in this way, let alone the multitude of other factors like social stressors, diet, sleep, etc. This article is genuinely misleading and poisonous to any impressionable young person reading this.

stephencz

Ahhh nothing like an existential crisis on Saturday morning!

Your network blocks the Lichess assets!