End of story, they should be ranked higher. Of course, in baseball we get many more matchups and a billion games to prove that right or wrong, that is not the point.

In CFB, it is obvious that we are not afforded that opportunity. But it is asinine to throw out other data we have and say, hey, Wazzu beat UW. End of story. They are better. They should be ranked higher. Any system that says otherwise is wrong. Instead, upsets happen. The point isn’t to make a transitive property ranking system of who beat who. The point is to use all the data we have and rank the teams. As the season goes on, we get more and more data and need fewer inputs from less reliable sources such as pre-season numbers. But until then, there would be far too much noise in a ranking system that relies on tiny samples.

In real life, they get one shot at Wazzu. Nobody has disputed this. But that is not the point of these models. At all.