Last night watching the Big 12 Championship, I heard the ESPN halftime crew say that Oklahoma should be playing Alabama in the BCS Championship game, based on the quality of their losses. I don't really think you can base anything about a team based solely on their losses, but if anything, shouldn't the inverse be true?
Suppose the teams involved in the various single loss games for Florida, Alabama, Oklahoma, Texas, Texas Tech, USC and Penn State played a rematch on a neutral field the day after the initial game. The loser's chance to win the followup game based on everything we know now might be something like this:
USC (v. Oregon State): 80%
Florida (v. Mississippi): 70%
Texas (v. Texas Tech): 70%
Penn State (v. Iowa): 70%
Oklahoma (v. Texas): 50%
Alabama (v. Florida): 30%
Texas Tech (v. Oklahoma): 20%
Arguments can be made about these percentages, but that is not really the point. The point is that a loss to a poor team is generally considered to be more of a fluke, and not a real representation of the team's abilities, and a loss to a top team is often seen as a real evaluation of the quality between the two teams. It is much more likely that the better team (at the time the games were played) won the games featuring the top matchups than the shocking upsets.
This does not mean that rankings should be based on the inverse of 'quality' losses at all. Each loss has to be evaluated on a case-by-case basis in the context of the team's entire season. Just don't try to tell me that Oklahoma should be playing Alabama in Miami.
