Hello Robert-Jan,

Il 28 settembre 2025 alle 23:35 Robert-Jan Veldhuizen ha scritto:
> Hi Francesco,
> 
> Your formulas are correct.

Very good!

> An important distinction  between error analysis and luck analysis, is that
> the latter is unbiased. Which means that in the long run, the luck analysis
> numbers will approach the real values (inaccuracies of the bot will cancel
> out).This is not the case for error analysis, where inaccuracies of the bot
> may not cancel out and sometimes  even pile up.

Many thanks for the explanation. I fetched pencil and paper, and indeed
it is easy to see L is an unbiased estimator, by construction. [1]

[1] Detailed in https://www.bkgm.com/articles/Zare/HedgingTowardSkill.html

> With GnuBG, I believe the default luck analysis is unfortunately (still)
> set to 0-ply and not changeable from the GUI. 0-ply luck analysis is quite
> inaccurate. With a command "set analysis luckanalysis plies 2" (or even
> higher, although that might be slow) you can improve the quality of the
> luck analysis significantly. You'll probably find that doing luck analysis
> at higher settings reduces (but does not remove) the discrepancies between
> error analysis and luck analysis.

This would indeed be very useful.
Browsing the code, I think this is the place

    https://cgit.git.savannah.gnu.org/cgit/gnubg.git/tree/analysis.c#n205

Alas, I am far from proficient in C.

Thanks again!
—F

Il 29 settembre 2025 alle 01:36 Murat Kalinyaprak ha scritto:
> BTW: This is not DailyGammon owned by your "control freaks"...
> You may get deservedly slapped for any garbage you spew...

Please, let us not abuse people answering questions!

Reply via email to