On Wed, Aug 27, 2008 at 5:45 PM, Rémi Coulom <[EMAIL PROTECTED]>wrote:
> Don Dailey wrote:
>
>>
>> Yes, I believe it does generalize on average.
>> This data matches my 13x13 study pretty closely, about 62% give or take
>> for each doubling. That is about 90 ELO or so. I have heard that
>> 100 ELO is 1 stone which is what I was basing this on. But it's not
>> clear to me at all if that is true. So I can only guess that 4x in
>> Mogo is worth something like 1 or 2 stones or something between.
>> - Don
>>
>
> According to my experience with Go data, it is not possible to give the
> value of one stone in terms of Elo ratings. For weak players, one stone is a
> lot less than 100 Elo. For stronger players, it may be more.
>
> Also, it is very important to understand that the Elo model is very wrong,
> and Elo against humans has nothing to do with Elo against computers (and
> even less with Elo against the previous version). In games against GNU Go,
> Crazy Stone improved 200-300 Elo points in one year. On KGS, this translated
> into an improvement from 2k to 1k.
>
> Rémi
I made a relevant post about this awhile back, below I pasted it. Maybe
I'll make a senseis page for this. KGS varies the "k" factor according to
the strength of the players to account for the differences in winning
probabilities in even games for weak players vs strong players. The EGF
formula has a similar factor "a" that varies for the same reason. So the
EGF Elo ratings which set 1 stone = 100 points doesn't correspond directly
to the classic Elo probabilities.
So for KGS, improving one stone can be anywhere from 139 to 226 classic Elo
points.
For CrazyStone's improvement from 2k to 1k, this represents around a 182
classic Elo points improvement.
---------------------
See below I created a table that shows the transformation from KGS ratings
to the Elo that CGOS uses. I set 6k=1800 because I believe that is what GNU
3.7.10 is under both systems. Does anyone have more data points for bots
that play on both systems?
Also is there an "all times" list for 19x19? CS-9-17-2CPU is on top at
2297, but I don't think that's the strongest CrazyStone. I think other strong
19x19 bots are not on the current list either.
To create the table I calculated the probability for a player of rank X to
upset a player of rank X+1. Then I converted this to the equivalent number
of Elo points for the same upset rate. Finally I made a running total,
setting 6k to 1800. The transform could be done using calculus I suppose
but it's been a long time since I did that! :)
ELO set
rank k 1 rank 6k=1800
10k 0.80 139 1,244
9k 0.80 139 1,383
8k 0.80 139 1,522
7k 0.80 139 1,661
6k 0.80 139 1,800
5k 0.80 139 1,939
4k 0.88 153 2,078
3k 0.97 168 2,231
2k 1.05 182 2,399
1k 1.13 197 2,582
1d 1.22 211 2,779
2d 1.30 226 2,990
3d 1.30 226 3,216
4d 1.30 226 3,442
5d 1.30 226 3,667
6d 1.30 226 3,893
7d 1.30 226 4,119
8d 1.30 226 4,345
9d 1.30 226 4,571
_______________________________________________
computer-go mailing list
[email protected]
http://www.computer-go.org/mailman/listinfo/computer-go/
