>A couple of questions I have are:
>
>* What is a "reasonable" number of minutes played?  HE teams have
>played, on average, about 1200 minutes.  The goalies listed above have
>played anywhere from 477 minutes (Noble) to 1210 (Allison).  (Maine
>has played 21 games including 6 overtime games.)
 
In my mind, a reasonable number would be somewhere around 180-200 minutes,
but I'm not really sure. Basically, one wouldn't say that a goalie's save%
or GAA is truly representative after a game or two. The same is true for
perseverence. By 477 minutes, I would suspect that a goalie has found his
level.
 
>* WHY is this formula supposed to mean anything?  I ask because anyone
>can come up with a formula and claim it means something.  And why is
>the formula set up the way it is - i.e., why multiply save% by 6?
 
True, anyone can come up with a formula. This argument is rather like the
ones in favour of or against RPICH/CHODR/KRACH, etc. It comes down to
whether one rating system makes sense to the reader or not.
 
It makes sense to me that workload is something that should be taken into
account when measuring a goaltender's performance. The reason the formula
is set up as it is is because that's what the writers of the book thought
made the most sense.
 
Multiplying save% by 6 is done to ensure that the save% is still the
dominant part of the rating. If we assume that a goalie's save% will be
anywhere from 80.00 to 90.00, and his shots faced/60 min is from 20-35,
it turns out that perseverance is 13-14 parts save% and one part Sh/G.
 
See you later,
John
 
--
John C.K. Edwards        Stats Geek, Ottawa Jr Senators (23-22-3 51pts) (CJHL)
Carleton U., Law IV                   I don't give a damn about being liked,
[log in to unmask]           but I sure as hell intend to be respected.
All Canadian [Tier II] Junior A standings: http://chat.carleton.ca/~jedwards