Suppose that the ECAC had come up with the following scheme for
accounting for missing Vermont games:
 
(1) Ignore all games with Vermont when determining ECAC final standings,
then
 
(2) If two teams are tied at the end of this 20-game regular season,
   the FIRST tiebreaker will be their records against Vermont with
   (a) a "win" against VT beating a "didn't play" VT
   (b) a "didn't play" VT beating a "loss" to VT
   (c) a "tie" vs VT beating a "didn't play" IF the two teams have a
       record below 0.500, but a "tie" loses to a "didn't play" if
records are
       above 0.500.
 
(3)  If the FIRST tiebreaker fails to break the tie (i.e. tied teams did
    the same against VT), the usual ECAC tiebreakers (e.g. head-to-head,
...)
    would be used to break the tie.
 
Sound bizarre to use the Vermont games this way?  Well, this is
essentially equivalent to what the "Rank by Pct. with Vermont games
counting" scheme the ECAC has adopted does!  Try it - you should see
that including the seven Vermont games in a percentage based ranking
system produces the system above  (with one possible exception - can you
find it?).
 
Robin Lock
St. Lawrence University
[log in to unmask]
 
HOCKEY-L is for discussion of college ice hockey;  send information to
[log in to unmask], The College Hockey Information List.