Re: [Bug-gnubg] Empirical validation of rating prediction formula

[Holger]

At 21:32 05.10.2003, address@hidden wrote:
> As  a reminder:  a while  ago I  did some  simulations to  correlate ELO
> rating  with checker  and cube  error rates,  resulting in  a predictive
> formula for the rating of a player from a GNUBG error analysis.

> An action item would now be to change the verbal playing level indicator
> ("beginner",  etc.) to  also  be based  on  the estimated  rating as  it
> currently gives inconsisten results.

Having had some time to get used to the new rating calculation I have to 
say that I'm not sure that it's inconsistent (at least for Intermediate and 
above). I rather tend to believe we have to think over our perception of 
FIBS ratings. It seems to me that e.g. a rating of 1750 is not nearly as 
good as I formerly thought. I think it just shows some understanding of 
game principles and is, how gnubg puts it, "Intermediate". (The 
categorization depends quite a bit of the relation of chequer play and cube 
errors, though.)

Regards,

        Holger