Head-to-head, Harvard won 8-3 at Minnesota on Nov. 7. After the game, Halldorson acknowledged that Minnesota was completely unable to keep up with the speed of Harvard's top line.

Clearly this game was not weighted as much as Harvard's three head-to-head losses to Dartmouth because it came so early in the season. This is a pretty typical practice of all college selection committees.

So the selection committee does have a statistical argument for taking Minnesota over Harvard. Now I'll show why the statistics lie.

First of all, the RPI is a stupid system to use in college hockey, men's or women's. It works pretty well in basketball, where there are many conferences and teams play almost half their games out of conference. But it isn't like that in women's hockey. There are only two conferences, and few games out of conference. Harvard, for instance, played just two games against non-ECAC opponents.

The selection committee likely used something like the RPI to justify the belief that the WCHA was as strong as the ECAC and worthy of just as many berths in the national tournament. This is a terrible justification, because there simply weren't enough non-conference games for the RPI to accurately reflect the weakness of the Western teams.

The record against common opponents seems to favor Minnesota. But Minnesota played five games against Wisconsin, which finished third in the WCHA. Harvard played only one. That warps the statistics. Wins against Wisconsin were pretty easy to come by. The Badgers are a team that lost 15-0 to Northeastern and 13-2 to Dartmouth earlier in the year.

Against common ECAC opponents, Harvard was 6-4-3, and Minnesota was 7-3-0. I would hope that the selection committee at least looked at this statistic, which still slightly favors Minnesota.