English Chess Forum

Or at least that study of chess improves the mathematical performance of children generally, strugglers or not.

Arshad Ali wrote:

JustinHadi wrote:Don't know about Speelman but Nunn was admitted to Oxford to study Mathematics as the youngest undergraduate since Cardinal Wolsley long before he became a GM. He is at the top end, there are many others with a similar correlation between chess and maths abilities.

Other than Nunn, who could have been a successful academic mathematician had he chosen that route, there's Noam Elkies in the Harvard math department, who's a well-known study composer as well as having an OTB rating of around 2200. Otherwise I can't think of any other top-flight mathematicians who've also been high-rated chess players -- or the other way around. There is Kenneth Rogoff, who's GM strength and a professional economist ....

I'm pretty sure we've discussed the chess-maths connection on here several times before.

Another Oxford student one, less well known, would be H Dugald Macpherson, who was 2300 ish rated as a University player and is an academic mathematician (Professor at Leeds).

I'm sure there are many other examples, at various levels. For instance, one of Chorlton's best players, though now probably in his early 70s, is Manchester Maths Professor Ron Doney, who I'd guess must have been ECF 200+ at some stage.

Of course early ability at chess is not always associated with academic excellence

Comes back to the definition of maths I guess I suspect a good majority of active chess players did A level maths.

A good number will have then gone on to do maths/stats/ comp sci/ engineering or such like at University and then onwards in life. They can do it quite easily and its useful enough. 'Proper' academic maths really is a very different animal indeed though.

Oh, Mestel seems to maybe get into the top 100 in the country for bridge as per the bridge grading system.

Not sure if this is related, but I was asking some children in my ACES chess class how they did in their 11+ tests (they come from all over Kent) and this year the test was made a little trickier such that only 8 children (out of 13,000 or so) scored 100%. In my class, there were two of them and three others who were in the top 37 (in total, 5 out of the top 37 in the entire county all in my chess class). It is particularly impressive in that there were only 6 of the class who sat the test.

What are the odds of 5 out of the 6 being in the top 37 in the entire county? I think there is a very strong link between being good at maths (and also academic success) and having an aptitude at chess.

We could run a test. Everyone can report their times to solve the above problem and then we run a correlation between their times and their current grade. Problem solved..

[No, only joking].

MartinCarpenter wrote:Comes back to the definition of maths I guess I suspect a good majority of active chess players did A level maths.

A good number will have then gone on to do maths/stats/ comp sci/ engineering or such like at University and then onwards in life. They can do it quite easily and its useful enough. 'Proper' academic maths really is a very different animal indeed though.

I strongly suspect there's a high correlation between the ability to handle the standard kinds of chess positions (tactical, attacking, positional) and the ability to tackle the kind of problems that occur in 'A' level math (say, calculating a tricky integral or proving some trig identity). Both involve practice, pattern recognition, rapidly going over different options, and the 'Aha!' flash of insight.

Arshad Ali wrote: Both involve practice, pattern recognition, rapidly going over different options, and the 'Aha!' flash of insight.

I expect that's true, but the Conference seems aimed at using chess at a much more basic level.

http://londonchessconference.com/chess- ... /#more-220

The sort of question they would pose is
If a Rook is worth 5, a Queen 9 and a pawn 1, what is a Queen and pawn worth in terms of Rooks? That might be good for arithmetic, but ultimately is bad for chess as where the pieces are and what they are doing is as valuable.

I am a firm believer that chess improves the performances of children at school, but also that this is not confined to mathematics but applies to other subjects, including Latin and other languages. Proof may be another matter.

Which is what we're looking for, really, isn't it? Proper scientific research showing that teaching chess in schools improves academic performance in one way or another - and if possible, that it is a more effective means of doing so than alternative uses of the same amount of teaching time.

Not that there's a correlation between interest in and aptitude for chess and interest in and aptitude for mathematics, which is pretty much a matter of simple observation.

I agree that some scientific research is highly desirable.

John Townsend wrote:I agree that some scientific research is highly desirable.

Here you go. Here's something relating to a board game, with included scientific research.

Roger de Coverly wrote:
I expect that's true, but the Conference seems aimed at using chess at a much more basic level.
http://londonchessconference.com/chess- ... /#more-220
The sort of question they would pose is: If a Rook is worth 5, a Queen 9 and a pawn 1, what is a Queen and pawn worth in terms of Rooks? That might be good for arithmetic, but ultimately is bad for chess as where the pieces are and what they are doing is as valuable.

Roger, I would have expected a man of your intellect to have detected the point of the question posed. It is not about arithmetic, it is about parity. Nor strictly is it about chess - it is about logical thinking. Given conventional piece values (9,5,3,1) there is no way to construct 9 points from four pieces. It is understanding the argument from parity (four odd-valued pieces will always sum to an even number) that gives this particular exercise its educational significance. It is a small exercise but children (and adults) love it. It involves cognitive flexibility - being able to shift from one domain to another. This is a critical foundation to developing thinking skills. You might also credit this simple question with being original - something which is difficult to attain in the well worked interface between chess and mathematics.

John Foley wrote: Nor strictly is it about chess - it is about logical thinking. Given conventional piece values (9,5,3,1) there is no way to construct 9 points from four pieces.

So why use chess as an example? One of the things you need to know as a "strong" chess player is to break free of the notion of fixed piece values.

Exploiting an old joke, a physicist would explain away that two bishops, a knight and a pawn were equivalent to a queen as "experimental error". Indeed on a chess board, the extra pawn might well be irrelevant.

Arguably though, a rook is often worth less than the theoretical 5 and bishops as pairs might be worth 6.5 to 7.

Back in the mists of time, one of my fellow students (of mathematics) had written one of the earlier chess programs. I argued the case that a program couldn't be better than the knowledge of its programmer(s) played optimally. I thought the Deep Blue brute force method disproved this until the development of the "Fruit" series of programs (Rybya, Stockfish, Houdini etc.) demonstrated the importance of "chess knowledge".

JustinHorton wrote:Of course early ability at chess is not always associated with academic excellence

Note, though, that The Simpsons' is written by mathematicians, and Simon Singh has written a book about it:
http://www.theguardian.com/tv-and-radio ... imon-singh

In all these discussions we should be aware that "correlation does not imply causation"