[blml] IMP-scales for 2/3/4 boards

[Harald Skjæran]

On 10/02/07, Herman De Wael <hermandw at skynet.be> wrote:
> Tony Musgrove wrote:
> >>
> >> IMPs otoh are rounded to integers. Normal scores 20-40 give the same IMP.
> >> Then, it is not needed to use more precise formulas when awarding
> >
> > So my question is..why not expand the VP scale  (say 0-100 should do) so
> > that every integral IMP converts to an integral VP.  Different number of
> > boards can quickly be accomodated via
> > erf(x), sqrt(2), n/n-1 or whatever.
> >
>
> Yes Tony, that would be a sensible solution. But it means changing the
> whole way we deal with the calculation. My argument was within the
> current system. If we use a 30-step VP scale, then all subsequent
> regulations must also deal with integer (30-scale) VP's.

Why?

As a matter of fact we don't in Norway.
If a team get an unexpected bye (opponents don't show up) they are
awarded 18 VP's preliminary. If their average score in the matches
played is above 18 (or the non-show's average score aginst the
non-show is above 12) their score for the bye is adjustet upwards,
using non-integer VP's (one digit after decimal point). They are
awarded the best of 18 VP's, their own average and the non-show's
opponents average vs the nonshow.

-- 
Kind regards,
Harald Skjæran


>
> > I must get out more.  Vowed 10 years ago never to enter a blml argument
> > with Herman, but
> > last night had to watch England beat Aust at pyjama cricket.  The pain
> > killers were just
> > kicking in.  I am going out, I may be a little while,
> >
>
> Which just goes to prove that ODI is mere (well, you said it: pyjama-)
> cricket. Saw the highlights of that match yesterday evening. How one
> man can beat a whole team!
>
> > Tony(Sydney)
> >
>
>
> --
> Herman DE WAEL
> Antwerpen Belgium
> http://www.hdw.be
>
> _______________________________________________
> blml mailing list
> blml at amsterdamned.org
> http://www.amsterdamned.org/mailman/listinfo/blml
>