> >What I am looking for is some statistical measurement which would tell us
> >how much faith to put into the ranking.
>
> I can give you a percentage of total connections, but I can't translate that
> number to tell you how much faith to have. If I say 34%, then is that enough?
> Is only 5% enough? I have no magic cut-off; I can only say that the more
> connections, the better the rating.
>
> I was already working on a "connection count" (or in hockey terms, the number
> of different opponents) for each team for a future TCHCR set of notes, so the
> total connections will be a breeze to calculate. Look for it at a computer
> near you.
>
 
Thanks Keith,
 
I don't know what level "faith" to put into a number like you a proposing
either.  Its sort of like being told the temperature is 30 degrees when you
have never felt that temperature before.  Once you have felt 30 degrees a
few times, you begin to get a feel for what the number means...
 
Having some number would be good though, especially if you could could give
us the number as a fraction of "complete connectivity" as you have defined
it, or as a fraction of the "end-of-regular-season connectivity".  I think
would like to see two numbers:
 
	1. this season's end-of-year connectivity as a fraction of "total
	   connectivity,
 
	2. the year-to-date connectivity as a fraction of end-of-year
       connectivity.
 
"1." could tell us how well this season's schedule is as compared to other
seasons.  This would tell us something about the relative accuracy of the
ranking year by year.  "2." would let us know how close we are to the
maximum connectivity we can expect this year.
 
Another question: what effect on "connectivity" does a second game between
two teams have?  Keith seemed to imply that you were 100% connected to
another team when you played them once.  After you play them again are you
connected > 100%?
 
What about the case where teams are "well connected" within their
conferences, but the conferences are not "well connected" to each other?
How much does the proposed connectivity number increase with a second game
between two teams compared to the first game?
 
--
Craig McGowan
[log in to unmask]