Re: 2015-2016 USCHO Posters Poll
NE (4.57 against 2.13) has beaten down a number of teams including some in the mid pairwise.
Unadjusted goals for and goals against still has the problem of comparing teams with different schedules. When looking at Northeastern vs. Bemidji State, we have teams that have lost all of their games against the big three, and that have both gone almost unbeaten against everyone else. Bemidji's stats reflect that they have played more than a quarter of its games against Minnesota and Wisconsin, while Northeastern has played less than 7% of its games against BC, though this will go up to just under 10% tonight ad to 12.5% this weekend.
What happens if we look at the numbers against all teams except BC, UW, and UM?
Northeastern's GF/GA goes from 137/64 (2.14 ratio) to 134/54 (2.48)
Bemidji's GF/GA goes from 61/56 (1.09) to 56/25 (2.24)
That's just a rough look, and there's no particular reason to assume that their SoS has been the same even if we exclude those games against the top three, but it does say that one shouldn't just use goal ratio to make a case without looking more closely. Northeastern wins by scoring goals, and Bemidji wins by preventing them, but there's some evidence that they are roughly similar in overall quality.
FWIW, the change for North Dakota isn't nearly as extreme, which should be expected, since they don't in any way fit the description of a team that has "lost all of their games against the big three, and that have . . . gone almost unbeaten against everyone else." Their GF/GA goes from 60/52 (1.15) to 51/37 (1.38).
