Timothy J. Danzer <[log in to unmask]> writes:
 
> Since Erik Biever first posted the RPICH ratings to HOCKEY-L several
> weeks ago, there has been much discussion about the relative weights
> assigned to each factor in the rating.  It seems that giving 20%
> weight to a teams own winning percentage while giving 40% each to
> thier opponents winning percentage and their opponents opponents
> winning percentage is heavily biased toward strength of schedule.
> Noone has yet been able to explain this but I think I may have found
> the answer.  Read on....
>
...
> it is important to remember that most of the Division I teams play in
> leagues, and league games comprise the majority of their schedule.
> Some leagues play as many as 32 games per team, while others are
> lower at 21, but in all cases, they make up the majority of league
> games.  By necessity, the average winning percenatge of all league
> teams in league games is .5000.  What does this mean?
 
....
 
> Let's look at the opp win pct. factors for the CCHA, which plays the
> most league games at 32.  We see that the highest factor for a CCHA
> is 53.23% for Bowling Green, and the lowest is 50.99% for Western
> Michigan.  We note that all CCHA teams have factors above .5000, but
> we also note that there is a difference of only 2.24 points between
> the best and the worst.
 
I see this as a problem with RPICH. The western teams (CCHA in this example)
don't play many non-league games. So their opponents winning percentage
will be close to 0.500. Thus, while it may not help boost these teams, not
playing non-league games does not hurt these teams.
 
Lets take a random ECAC team who, in an effort to schedule a relatively
good schedule, schedules BC. BC is a team which generally does quite well,
and should help your schedule. Since schedules with non-league opponents
may be set years in advance (an agreement where `we' play `you' at your
home one year and switch the next year), it is not easy (or nice) to
unschedule someone like BC because it appears thay may have an off season.
Many would agree that this isn't in the best interest of hockey. And if
you could unschedule BC, wouldn't you also have wanted to do the same for
BU, who had a good season despite losing key players?
 
Now, BC has a sub-0.500 season, which hurts your oponents winning percentage.
So an ECAC team which schedules a team like BC who is having an off-year
is hurt in the RPICH compared to a CCHA team that doesn't play many (or any)
non-league games.
 
Since opponents winning percentage mostly depends on non-league games,
then you could try only using non-league games in this category. And
similarly only non-coference games played in the conference for
opponents opponents winning percentage. But how do you reconsile a team
that plays 2 non-conference games with a team that plays 9
non-conference games.  Should a mediocre non-conference schedule (a bit
below 0.500 non-conference opponents winning percentage) hurt a team
who plays more non-conference games (inevitably throwing in a few
weeker teams) compared to a team that plays just a few non-conference
games, choosing them very carefully?
 
The only other thing I'd point out is that `your' winning percentage is
included in your opponents opponents winning percentage (you are one
of your opponents opponents', right?). So, your winning percentage does
contribute more than 20% to the RPICH. The amount varies per league, but
should fall around 5% or so (example: a 9 team league playing each other
4 times. 4*8=32 league games. 4/32nd or the opponents opponents winning
percentage is your winning percentage, or 1/8 of 40% = 5%. A bit lower
due to non-league games). So RPICH may be considered 25%-40%-35%. While
this helps a bit, I still think its too biased toward schedule `strength'.
 
As a closing note, I'll admit to being a fan of ECAC teams. However, I
won't defend the ECAC. I think both Clarkson and SLU had the talent, but
neither was able to dominate the league (unlike Maine or Minnesota). If
either could have dominated and would still be ranked low, then there
might be something to complain about.
 
> There's my theory.  Fire Away !!!
>
> Timothy J. Danehy
> [log in to unmask]
 
Derek Snyder
GE Corporate Reseach & Development
[log in to unmask]