Re: FW: [Bug-gnubg] Update: Checkerplay vs cube decision errors

[kvandoel]

Hi,

Apart from noticing the misspelling of my name I have more comments:

Since chequer errors  and cube errors really have  different "units" the
ratio of the coefficients a2 and b is meaningless.

A meaningful measure is:

M = a2(N)*EPM / b(N)*EPC

M represents ratio  of rating points lost due to  chequer error adn cube
error. Clearly it is player dependent. For me it is usually 10 in actual
games I play.

Now in  the article this is  rewritten in the  form of Eq. 5.  The table
below Eq. 5 plots something undefined but I guess it is

(a2(N)/UFM(N))/(b(N)/UFC)

with UFM  # unforced  moves and  UFC # unforced  cubes as  obtained from
rollout.

This is the  ratio of the coefficients of a bilinear  fit to total error
rates of chequer and  cube play. It is not clear to  me this is really a
better measure than a2/b. We still have the fact that cube decisions are
really of a different kind  than chequer play decisions even though both
can be expressed in equity losses.

I don't understand the ssecond paragraph of the "Results". If the cutoff
for  actual cube  decision is  decreased  to .05  the #  of actual  cube
decision goes  down and the coefficient  b(N) will also go  down (it was
measured assuming a fixed cutoff) so nothing changes.

In  the 3d  paragraph I  read  "... doubtful  to clain  cube error  were
less..".   I  don't see  any  arguments  that  support that  claim.

The  opposite statement  "cube errors  are less  important  than chequer
errors" has no rigorous meaning  either.  It just seems in practice that
the rating  points you lose  due to cube  errors are much less  than the
loss due to chequer play.


Kees
--
On Tue, 18 May 2004, Albert Silver wrote:

> Hi Joachim,
>
>       I'm taking the liberty of forwarding this to Kees so that he may
> read your comments and respond to them if he so desires.
>
>                                       Regards,
>
>                                               Albert
>
> -----Original Message-----
> From: address@hidden
> [mailto:address@hidden On Behalf Of
> Joachim Matussek
> Sent: Tuesday, May 18, 2004 9:51 AM
> To: address@hidden
> Subject: [Bug-gnubg] Update: Checkerplay vs cube decision errors
>
> Hello,
>
> it didn?t let me sleep.
>
> After thoroughly examining Kees van Doel?s scripts I have found an
> important misunderstanding. Within his item he uses the term "unforced
> cube decision" but in fact means "actual or close cube decision". This
> lead to many changes. Thus I had to rewrite my article.
>
> http://home.arcor.de/joachimmatussek/CheckerVsCubeErrorsImproved.zip
>
> Sorry for any inconvenience,
>
> Joachim Matussek
> _____________________________________________________________________
> Endlich SMS mit Bildern versenden! Das Bild selbst ist dabei gratis,
> Sie bezahlen lediglich den Versand. http://freemail.web.de/?mc=021195
>
>
>
> _______________________________________________
> Bug-gnubg mailing list
> address@hidden
> http://mail.gnu.org/mailman/listinfo/bug-gnubg
>
>
>
>
>