We have had two seasons of the College Football Playoff and its selection committee. There have been over a dozen sets of rankings to analyze and discuss for clues at to what the committee wants and prefers from its contenders. We have a general idea of what the committee has liked historically. We have a general idea of how the committee treated Group of 5 teams. We also know that there’s really no way to know anything. That’s not quite a bad thing, but it’s not a good thing either.
Preseason CFP Implications
Why do I say that we don’t really know anything? First of all, it’s because precedent isn’t inherently meaningful. The committee has done things in the past, and we do have the list on the website of what the members have to consider (strength of schedule, head-to-head, conference championships won), but there is nothing mandating how those factors are weighed.
Also, we have seen a different main talking points from committee chair Jeff Long in each of the first two seasons, neither one of which is one of the specific criteria that the committee is supposed to use. In 2014, we heard a lot about “game control,” which was essentially a way of saying the “eye test” without using those words. Long spoke at length about how strong or dominant teams looked, focusing much less on things like strength of schedule. In 2015, we finally got a solid strength of schedule metric. Unfortunately, it was also not a particularly useful one. We heard many comments about “wins over teams with records of .500 or better.” That wasn’t really discussing how good any of the teams were; it is an objective metric that unfortunately misses how good the team is or how it got to that record.
What will we hear from the committee this year? There is no way to guess. The fact remains that ranking these teams has always been more of an art that a science. The BCS brought a measure of objectivity in that it had numbers that could be seen and understood. Of course, a large percentage of those numbers came from entirely subjective human rankings. Now, though, we have no numbers to look at. We have a list of teams and some far-from-exhaustive explanations from the committee chair about what those numbers could mean.
There is a new chair of the committee this year. We won’t be hearing from Jeff Long on Tuesday nights anymore. Kirby Hocutt is the new chair, so he will take over the thankless job of giving impossibly concise explanations as to how a group of 12 people reached its rankings. Yes, the committee will only have 12 members this year. Lloyd Carr dropped out for personal reasons, so for the second time in three years there will only be 12 members of the selection committee, not 13. Because of the convoluted voting system, Lloyd’s absence will have no real effect on the committee.
It is interesting to note, though, that Carr would have had to recuse himself from any vote involving Michigan. While it sounds like the policy would hurt teams who have recused committee members, it really has no effect. Recused members cannot vote in any ballot that their team is on. Thus, if Carr was voting and Michigan has the potential to be in the top six, Carr would not vote in any round involving the top six. (The committee selects teams three at a time out of groups of six teams. Thus, if Michigan was eligible to be in the top six, Carr could not vote in that first ballot. If Michigan was not selected in the top three and therefore is again eligible for spots 4-6, Carr could not vote in that second ballot, either. If Michigan was selected in that 4-6 group, Carr would have then been eligible to vote in all subsequent ballots.)
The Conference Champions Fallacy
One potential takeaway from the first two seasons of the CFP is the importance of being a conference champion. After all, all eight Playoff participants so far have been champions of Power 5 conferences. This certainly is important. However, we have to be very careful to need read too much into it.
First of all, if you remember back in 2013 when the Playoff was announced, Bill Hancock made it very clear that it would be about the “best four teams,” regardless of conference. Many saw that as a hint to Alabama being a second SEC team that would have made the Playoff that year. Moreover, it’s pretty clear on the CFP website that the committee will put in a team that is not a conference champion–so long as it is one of the best four teams in the country.
Also, and this is really the most important reason to not get carried away with what happened the last two years. 2014 and 2015 really never provided an opportunity for anyone but a conference champion. Think about it. There were six teams in the mix in 2014. All six were conference champions (co-champions in the case of TCU and Baylor). Michigan State and Mississippi State were clearly outside of the conversation compared to the top six. If Mississippi State had not been upset by Ole Miss to end the 2014 season, we might have learned something important. As it shook out, though, the only teams possibly in the conversation all happened to be conference champions.
In 2015, things weren’t quite as clear-cut, but we actually had a standout top four. There were those like Ohio State and Iowa that had some sort of claim, but at the end of the day there was no real way, other than a very subjective “eye test,” to truly argue that either was better than Oklahoma. Thus, while we have only had conference champions thus far, it might be way too soon to take that as any sort of precedent.
Bear in mind that we are just two upsets in conference championship games in any given year from being almost guaranteed to have a non-champion make the field. It hasn’t happened yet, but two years is a tiny sample size. Any statistician knows the be careful when extrapolating from small sample sizes, and this case is an obvious example of it.
Two Meaningless Meaningful Gestures
The selection committee has tried to sneak something very interesting into the final rankings each year, though neither one slipped past my watchful eye. In 2014, Baylor finally jumped over TCU in the final ranking, showing that head-to-head will finally come into play when the resumes are similar. Last year, Iowa remained ahead of Stanford in the final ranking, even though the Hawkeyes were not a conference champion. The obvious implication of that decision was that had, say, Alabama lost its conference championship game, the Hawkeyes would have made the Playoff over the Cardinal.
Of course, these two precedents are both meaningless. First of all, they literally didn’t mean anything. There is zero difference between No. 5 and No.6. Sending a message with any team outside the top four doesn’t get that message across because nothing meaningful comes of it. Second of all, the committee changes each year. Just because there was a decision or a consensus in a previous year does not mean that it will remain the same in any other year. Again, it’s really easy for the committee to send a message that doesn’t have any real effect and can be ignored by the next committee with zero consequence.
Week 1 Implications
As has been discussed in many places, Week 1 this season is absolutely stacked. How much will it affect the rest of the season and the Playoff picture? It is way, way too soon to guess that. If previous years taught us anything, at the top of the list should be the fact that early-season troubles can be overcome if a team shows real improvement. We haven’t yet seen a team make the Playoff after two early-season losses, but the eventual champion each of the last two years suffered a loss within the first three weeks of the season.
The one Playoff hopeful who cannot afford an early-season loss is Houston. There are a lot of things we can’t really say with certainly about who will make the Playoff, but we can be pretty sure that, barring mayhem well beyond 2007 levels, a Group of 5 team needs to be undefeated to make the Playoff. Other teams in high-profile Week 1 matchups can afford a loss if they improve throughout the season. For Houston, it’s do-or-die from day one.
