The college football playoff committee had three extremely easy calls. Despite no unbeatens from the power five conferences, there were three one-loss conference champions who played solid enough schedules to be automatic selections.
The only question left was who would be the fourth team. The Big Ten and Pac Twelve both had two-loss champions, while there were also two one-loss non-champions (Alabama and Wisconsin). Among the conference champions, Ohio State would be the obvious selection ahead of USC due to strength of schedule. Among the one-loss non-champions, Wisconsin and Alabama were comparable - but clearly Ohio State would be picked ahead of Wisconsin so let's focus on that comparison.
The published criteria for determining between "comparable" teams are as follows:
For the sake of argument, let us assume that "comparable" means "equal records", so number of losses would be an implicit fifth criterion.
- Championships won
- Strength of schedule
- Head-to-head competition
- Comparative outcomes of common opponents
Ohio State and Alabama did not play each other or any common opponents, so the last criteria are N/A. Ohio State won its conference championship while Alabama did not. And for strength of schedule, a typical ranked team would have lost 3.0 games against Ohio State's schedule but only 1.8 against Alabama's - so we can be a little generous to Alabama and consider the strength of schedule and number of losses to be a wash overall.
Adding it all up, one ends up with two fairly comparable teams, but with one having won its conference and the other not even allowed to play for its conference title. Should be clear-cut choice for Ohio State ... but Alabama was selected. What could have possibly happened?
Overall, the moral of the story seems to be that the college football playoff committee will not adequately consider strength of schedule. So, if you have a tough conference schedule, best to play as many cupcakes as possible. And, memo to the Big Ten - better to follow the SEC's example of playing only eight conference games and add another Mercer to your schedules.
- Ohio State's two losses automatically placed them behind Alabama as a "not comparable" team for which the four criteria above would not be applied. This is not credible, given that Alabama was ranked behind two-loss Auburn in the previous week's rankings. Had Auburn won the SEC title, there is no doubt they would have been ahead of Alabama in the final rankings.
- The college football playoff committee doesn't consider conference championships as much as they claim. Given that Ohio State got into the 2016-2017 playoff rather than the Big Ten champion, this seems plausible. However, even if the committee ignored conference championships, the composite compute ranking tallied by Massey shows that Ohio State's strength of schedule was more than adequate to compensate for the extra loss.
- The strength of schedule metrics used by the playoff committee are horribly flawed. Often, strength of schedule metrics are some sort of composite of opponents' and opponents' opponents' records, which is meaningless when picking from the best teams. Playing an 11-1 team instead of an 8-4 team is a huge deal, while playing a 3-8 team instead of an 0-11 team is meaningless as either one is an automatic win for any team being considered for the college football playoff. What is key is the quality opponents. Ohio State had three: #2 Oklahoma (loss), #6 Wisconsin (win), #9 Penn State (win). Alabama's only quality opponent was #7 Auburn, whom they lost to.
Return to ratings main page
Note: if you use any of the facts, equations, or mathematical principles introduced here, you must give me credit.
copyright ©2017 Andrew Dolphin