,
we are happy to add Kirill Kryukov to the team of CEGT testers. Apart from helping to test the deep versions on a dual machine he will now also offer valuable statistics and comparisons. So we got a table for number of games played between engines what is very useful to plan further matches and keep a better balance.
Moreover Kirill will regularly compare Bayesian ELO and EloStat. This was just done with the current CEGT rating list 40/40 and when you check you will see some differences like for example when you compare Fruit 2.2 and Fritz 9.
http://www.husvankempen.de/nunn/rating.htm
If there are any questions regarding this statistics, Kyrill offers to answer them here.
Best Regards
Heinz