> Take BU, for example. Their SOS schedule is dropped (significantly) by
> playing Merrimack three times in the HEA playoffs. Boston College then
has
> its SOS dropped by playing Merrimack and Maine to win the title, rather
> than UNH, BU, or Northeastern. Likewise, in playing 3 games versus BU
> (ranked 3) and BC (now ranked 4 or 5), Merrimack's SOS skyrockets.
 
Is this a strong effect?  Let's take the difference between say Princeton
and Clarkson.  In the PS, Princeton plays #4 three times, #8 (thanks to an
upset), #1, and #2.  Clarkson plays #10 twice, #5, and #7.
 
Princeton: 4,4,4,8,1,2
Clarkson: 10,10,5,7
 
Princeton played *5* games against the top third of the league and one
against the middle; Clarkson played 2 against the bottom third and 2
against the middle.  That sure seems like a strong effect.  And if, for
example, RPI had been a bubble team, their path:
 
RPI: 8,8,8 would have been even more deleterious for SOS.
 
Now take an abstract case in which there are no tourny upsets.  Compare the
paths of all teams involved in the Final Five:
 
#1: 10,10,4,2
#2: 9,9,3,1
#3: 8,8,2
#4: 7,7,5,1
#5: 6,6,4
 
And presumably the bottom 5 teams have no chance at the NCAA anyway.  In
this case, there's really not much of an effect pushing contenders' SOS
around relative to one another.
 
Finally, as one observes that a better team will tend to have a somewhat
lower SOS because of drawing worse opponents in the playoffs, doesn't that
mean they will tend to have a somewhat *higher* opponents' SOS by the same
logic?  This should dampen what effect there is.
 
HOCKEY-L is for discussion of college ice hockey;  send information to
[log in to unmask], The College Hockey Information List.