Jason Patton writes:
 
> I found it very curious that you opted to put Winning %s down (is this
> because the analysis would have looked so odd if you had put the seemingly
> statistically insignificant records down instead?... both in what you did on
> the MAAC and on the ECAC).
 
No, I listed percentages rather than W-L-T because they're what my
program normally spits out (results are actually stored as PF-PA
rather than W-L-T).  For the modified strength of schedule measures
(opponents' percentage and opponents' opponents' percentage) there is
no corresponding W-L-T anyway.  Listing percentages may also have
obscured the fact that I was replacing the "last 16" record with
record against all non-conference opponents, so for instance
Princeton's was 5-2-0.  Slightly more significant than just including
the Princeton-Army game.
 
Yes, conference games provide us with a lot more data, but if we have
no baseline for the strength of those teams, the results can be
inaccurate.  Since MAAC teams don't play teams from other conferences,
the most direct way to gauge their relative strengths is via their
performance against the four eligible Division I independents:
 
                vs Indies    vs Army    vs Niagara   vs AFA      vs Mankato
      Avg RPI   PF-PA  Pct  PF-PA  Pct  PF-PA  Pct  PF-PA Pct    PF-PA Pct
 HE    .526     14- 2 .875  12-0 1.000   0- 2 .000   2-0  1.000   0-0  .---
 WCHA  .506     28- 8 .778   0-0  .---   0- 0 .---  10-0  1.000  18-8  .692
 CCHA  .505      5- 5 .500   0-0  .---   2- 4 .333   0-0   .---   3-1  .750
 ECAC  .497     28-10 .737  10-0 1.000  10-10 .500   2-0  1.000   6-0 1.000
 MAAC  .452      8-22 .267   5-5  .500   0- 4 .000   3-5   .350   0-8  .000
 
Note that no MAAC team has beaten or tied Mankato or Niagara, while no
"big four" team has lost to or tied Army or Air Force.
 
Mitchal Hawker made the excellent point that the KRACH rating (which I
single out because it embodies the strength-of-schedule idea of RPI
and I know it doesn't include margin of victory) does put all of the
MAAC teams near the bottom of the barrel.  This includes MAAC and
non-MAAC games in exactly the same way, and the conclusion is that
Quinnipiac doesn't have to be that good to amass such a winning
percentage against the seven teams that appear on their schedule.
 
I could go on at greater length, but I'm trying to avoid having this
consume my entire week.
                                          John Whelan, Cornell '91
                                                  [log in to unmask]
                                     http://www.amurgsval.org/joe/
 
HOCKEY-L is for discussion of college ice hockey;  send information to
[log in to unmask], The College Hockey Information List.