Re: [pbmserv-dev] Re: Backgammon Standings

[Lyman Hurd <lhurd@yahoo.com>]

Except what ratings are supposed to measure is the
probability that a player would beat another and
realistically this function is probably not
time-dependent.  I really think that inactivity should
be noted in some way other than chaning rankings.

I gave one suggestion.  Another might be expanding
Established/Provisional to
Established/Provisional/Inactive and have some time
out for transitioning to the third status.  The actual
cut-off time may be game-dependent as there are some
games that take many times longer than others.

Cheers,

Lyman

--- Yves Rutschle <pbm@rutschle.net> wrote:
> On Fri, Mar 12, 2004 at 09:37:38AM -0500, Richard
> Rognlie wrote:
> > Just remember, there are three kinds of lies...
> > 
> > lies
> > damn lies
> > and statistics
> 
> C3PO: "R2 says there is one chance in 2.78 million
> to come
> out of an asteroid field alive!"
> Solo: "you know, me and statistics..."
> 
> 
> I support the spirit of the suggestion (while
> finding it
> ironic that this topic has come up several time, but
> now
> it's Richard who's frustrated so hopefully something
> will
> get done ;-) ).   
> 
> Statistics degradation sounds good, however doesn't
> that
> change the balance of points? If I understand the
> rating
> system correctly, one starts off with 1700 points,
> then
> every game played effectively transfers points from
> the
> loser to the winner. As a result, the average rating
> should
> remain 1700. So, degrading one's rating should
> result in
> giving their points to active players to conserve
> the
> balance. Incidently, a player who's drifted down to
> 1700
> over months, could then be removed with no incidence
> (and
> they'd have been inactive for many months by that
> time).
> 
> Just my 2 hundredth of currency unit...
> 
> Y.
> 
> 
> To unsubscribe, send a message to esquire@gamerz.net
> with
> 	unsubscribe pbmserv-dev@gamerz.net
> as the BODY of the message.  The SUBJECT is ignored.
>