| Subject: |
|
Re: poor criticism |
| Name: |
|
Wim van Vugt |
| Date Posted: |
|
Jun 4, 07 - 7:25 AM |
| Message: |
|
Hi Gino and Ron,
There have been shown in literature plenty of examples of so-called linear relationships with high correlation, while in fact a third factor caused the relation between X and Y. A famous one was the relation between the number of radio sets sold and the increased number of psychiatric patients in England between 1920-30.
But the relation between the number of moves and used time is not an accidental one, as one might easily understand.
From this (including the spread within it) I had calculated a theoretical time distribution:
Last weekend I have tested this theoretical distribution to the practical values to be found on the webserver. I have selected about 100 tournaments (WCCC, CL, Joel Adler Mem, Olympiads, &tc) of which the players have ratings of 2200 and higher. Only below 2200 there was a significant difference: those tournaments were more quickly finished than expected. These I have excluded.
.jpg)
As one can see, the experimental data fit reasonably well to the theoretical curve.
Of one has any doubts about the goodness of fit, I have done the same for the so-called uniform distribution of random numbers between 0 and 1. Taking 200 random numbers and plotting these against the theoretical straight line, one also gets a "reasonable" curve-fit. But even here a considerable spread is visible.
.jpg)
So, I wonder with what argument our Ken might come to show that my analysis is of no value.
Best wishes,
Wim |
|
Replies:
|
|
|
|
|
Hey Webmasters! New Photo Album Service Launched - Check it out!
Pending TCCMB Exhibition Matches!
| 
TCCMB FAQ/Terms of Use |
Visit CC.COM's Online Store!
TCCMB
Video (Windows Media Player)
Reply
Home
bravenet.com
