Author: Dann Corbit
Date: 17:07:58 05/31/02
Go up one level in this thread
On May 31, 2002 at 19:48:12, Chris Carson wrote: [snip] >>Trying to show exactly what is happening is like herding cats. Get the darn >>thing to hold still, and maybe you could formally prove something. > >Not at all. Simple design. Simple two tailed analysis (one tailed if only >checking for inflation). I have research assistants that can set this up and do >the analysis, it is routine, but I am skeptical and see no practical use in >testing this, if you think there is significant drift (use 95% confidence >level), then test it. The hardest part is getting the FIDE lists, but they are >all there if you look for them, the stats are easy, Excell or SPSS will do. What is significant drift? 100 ELO over 200 years? I disagree that you can draw any meaningful conclusions with a constant flux of data points. In fact, I don't think you could even model it accurately enough to be useful. Players learn as they play. Someone might be 150 ELO points better in two years time. Players lose their skill over time as well. With huge variation for a single player possible, and even at that we are making no assumptions about longevity in the pool. I think it is herding cats. Do any math you like, but I won't believe your numbers. There are far too many variables that are not held constant. I think a simulation is the only answer to even get an approximation of what might possibly occur. Once we know how the changes to the population (and individuals within the population) affect the scores, then it might be possible to make some extrapolations about what happens in real life. I am totally sure that the absolute number drifts either up or down and also totally sure that it does not matter a lick which way it goes. Therefore, I suggest putting a saucer of milk on the floor and letting the cats have at it.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.