Author: Jeff Lischer
Date: 12:33:40 01/05/06
Go up one level in this thread
On January 05, 2006 at 09:09:33, Maurizio De Leo wrote: >On January 04, 2006 at 16:16:44, Jeff Lischer wrote: > >>>www.deleosorice.altervista.org/altro/comparison.jpg >>> >>>As I tought the rating lists are very reliable and there are only a few cases >>>where the error ranges are incompatible. >>> >>>Maurizio >> >>Thanks for the chart! It's nice to see the various lists agreeing with each >>other. From a quick glance at the chart, however, it looks to me like the KURT >>ratings are pretty consistently 25-100 points below the others. How did you >>optimize the scaling from list to list? > >I scaled the rating so that every list had the same average than Kurt's list >when considering only engines present in both. I could put up the excel file I >used if you want. >It may be that for some lists the overlap with Kurt's list is small, so the >scaling is skewed. What method would you suggest using ? > >Best Regards >Maurizio Here's a simple idea I just thought of. You could try scaling the lists so the Shredder 9 rating is some fixed value, say 2750. Shredder has been a standard for awhile, is included on all the lists, and usually has played many games so its error bars are smaller. I think CEGT even published some lists with Shredder 9 scaled to 2750. I realize it's not as good as using information from multiple programs when scaling the lists, but I suspect it will show even closer agreement between the rating lists than you're seeing now.
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.