Flutterby™! : Too Clever by Half

Too Clever by Half

2010-03-17 20:11:49.117527+01 by petronius 4 comments

Can you be too smart to be a chess champion? The world's top-ranked player thinks so.

[ related topics: Games ]

comments in descending chronological order (reverse):

#Comment Re: made: 2010-03-18 22:09:49.715071+01 by: Dan Lyke

Yep, I've found that with various games. Charlene and I are at that point with Scrabble. I can become a better Scrabble player, but primarily by memorizing lists of words that are only useful to Scrabble.

#Comment Re: made: 2010-03-18 21:11:45.598001+01 by: petronius

Discovered a great quote about Chess from Professor Volokh:

"Learning to play chess, he said, will help you learn to think better. But only up to a point. Past that point, it will only help you learn to think better about chess."

#Comment Re: made: 2010-03-18 15:53:32.317749+01 by: petronius

The fact is that there is no such thing as a single definition of intelligence. Some people excel in pattern recognition, others in mathematic reasoning, etc. I'm pretty good at crosswords but hopeless at sodoku. There seems to be a pattern issue in chess that rewards one kind of recognition. There are also odd talents that just pop up. My sister-in-law never forgets a telephone number, ever. Otherwise she is a pretty average person.

#Comment Re: made: 2010-03-17 23:39:46.285966+01 by: mvandewettering

I would make the claim that I possess a reasonably high level of intelligence in a number of different regimes, but I am usually fairly bad at board games. My son can beat me at chess. I find go baffling. I am becoming a middling checkers player as part of my work on a checkers program. I used to regularly get beaten at the various war and board games I used to play. Quite frankly, I lack the patience to study the back-and-forth of move and countermove.

There is a conceit that tells us that skill in chess is a sign of intelligence (a conceit more commonly shared among those who play chess). Levitt gave without any real justification or examination of data that a person's ELO rating was about (10*IQ) + 1000. This means that a person of average intelligence might rank a maximum of around 2000, and a person with an ELO score of around 2800 would have an IQ of about 180. But he actually makes no argument that these numbers are even broadly representative of reality.