Well, I did not really intend to crunch numbers, but I was curious...
Using the assumption that the lower ranked team has a 30% chance of winning
one game I came up with the following interesting numbers (which may not be
correct either):
2 out of 3:
P[Low rank team loses first 2 games] = (0.7)(0.7)= 0.49 = 49%
  thus there is a 51% chance the low ranked team turns it into a one game
 seriP[Low rank team wins] = .216 = 21.6% (not positive on this one)
 
What does this mean??  I don't know, but I have two points which basically make
any such statistical analysis mean much less:
(1)  It is economically and educationally unfeasible, IMO, to have a 5 game
College Hockey series unless the teams were less than 50 miles appart, which
could never be guaranteed.
(2)  The statistical analysis above implicitly ASSUMES that any one game between
the two teams is completely INDEPENDENT of another with the same probability of
p=0.3.  Clearly if one team gets blown out on the first night, it will have
some effect on how they will play the following night, not only psychologically
but alos the Coach will most likely change the game plan.  Thus, the 3 game
series can not be modeled as three independent random events.
 
Statistics and Probabilities are fun but be careful when drawing conclusions...
 
-Ryan Stone
 Lawrence University (Appleton, WI) '93
 Rensselaer Polytechnic Institute '93
 
 ----------------------------------   --------------------------------------
| RYAN G. STONE                    | | "In the Beginning, there was a Great |
| 218 North Hall, R.P.I.           | |  Void.  Then God said 'Let there be  |
| Troy, NY   12180-3590            | |  Light!'  There was still a Great    |
| (518) 276-7230                   | |  Void, but at least He could see it."|
| e-mail: [log in to unmask]          | |                   -Source Unknown    |
 ----------------------------------   --------------------------------------