There is a lot to learn from the CFP Selection Committee’s first rankings. It tells us where teams stand as the season moves forward and gives us some sense of what to look for in the future. Of course, it’s just one set of data points. So while we can’t draw full conclusions from one set of rankings, we can learn much about their thought process.
What the CFP Selection Committee Taught Us: Week 9
I will pre-empt all of this with a major warning, though. What we learned this week, from these initial rankings, will pale in comparison to what we will learn from next week’s rankings. Next week we will see how the committee shifts teams–how much it punishes for a loss and how much it rewards for a big win. Will it be like the polls, where teams just slide up and down when teams near them lose or win big games? Or will they actually look at a full profile all over again each and every week? Will we see Auburn drop if Kansas State starts losing? Will Oregon drop if Michigan State starts losing? How will the committee react to the entire college football landscape? Or will it react to it at all?
Those questions are getting ahead of ourselves though. First and foremost, let’s discuss this week’s rankings. And I’ll start with the elephant in the room, because that’s all anyone wants to talk about anyway. The SEC West has four teams in the Top 6. Good, there. I said it. Do you know what I’ll also say? It doesn’t mean anything and it doesn’t matter at all. What do I mean by that? I mean that people are missing the most glaring factor in these rankings.
It isn’t being discussed yet, but it should be. With the exception of Duke (who has beaten exactly one team with a winning record), every single 1-loss power conference team is ahead of all 2-loss teams. Every undefeated power conference team is ahead of every 1-loss team. The committee’s message is beyond clear: unless your schedule is absurdly weak (like Marshall, Duke, ECU, and Colorado State), the number of losses will be the determining factor. This can change in future weeks, but the committee clearly decided that they aren’t going to play around with how many good wins counters a loss.
I think Jeff Long made this fairly clear in his interview on ESPN after the rankings were released. He did not tell us too much, but I found it very interesting that he discussed Ohio State in particular. He mentioned that the Buckeyes have a weak resume so far but will have chances to build on it in the future. Now, Ohio State is the power conference team (again, aside from Duke) who will have the weakest 1-loss resume if they win out. Yet the message was sent that they would have a chance to work their way into strong consideration. I think that the committee is driving home a point that a 1-loss conference champion will be in consideration over 2-loss teams, except where there is an extreme difference in SOS.
So when we get to the SEC West, it’s obvious. The division (and the SEC conference as a whole) can have a maximum of two teams end the season with one loss. Odds are there will only be one, though, if even that. It just happens that the SEC West currently has four undefeated or 1-loss teams, all of which have picked up a pretty quality win somewhere along the road. The only team whose resume can possibly be quibbled over is Alabama (their only real quality win is currently ranked #20), but the committee already put them three spots lower than the major polls. Maybe the committee will give a 2-loss SEC champion from the West a little extra benefit of the doubt, but we have nothing to indicate that yet. What will be a major indicator is when we see how far the Auburn/Ole Miss loser drops this week. If we see them fall behind all of the 1-loss teams, we’ll know that wins and losses are absolutely paramount. If they only drop 5-7 spots (like most teams who lose a Top 5 matchup do in the AP and Coaches’ polls), we’ll know that the SEC West is in a very superior position at the moment.
The second clear thing is that the committee is rewarding strength of schedule, more or less. We have no proof to clearly say either way, but it does not seem like they are using any pure SOS metric. Rather, they are looking at quality opponents played. This is fairly obvious from Notre Dame, who by most metrics has a tougher overall schedule than Kansas State, Michigan State, TCU, and sometimes even Oregon. The committee sees that Notre Dame’s opponents all have mediocre records this season. And while most metrics utilize the fact that it is much more likely for a mediocre team to pull an upset than a bad team (of which Notre Dame plays none), the committee is working with the assumption that a good team should beat all mediocre teams and isn’t giving much credit even for a very large collection of mediocre wins.
What Notre Dame also proves is that the committee is not giving a bonus for close losses to good teams. They are not using the logic of, “They played the #2 team very close therefore they are pretty close to #3.” The committee is looking for a solid win on top of that good loss to back up a ranking.
Of course, the confluence of these two opinions is the same logic that led to Alabama being selected over Oklahoma State in 2011, a decision which the BCS was roundly blasted for. It could lead to a situation where one good win and ten bad ones (along with a good loss) is worth more than a lot of good wins and a worse loss. We’ll have more information in the future regarding whether the committee thinks this way, but it is definitely something to keep in mind for now.
We do not yet quite know what the committee thought about using head-to-head matchups as a tiebreaker. Ole Miss is ahead of Alabama, but Ole Miss has a better SOS and a better overall resume so far. Oregon is ahead of Michigan State, but again, Oregon also has a superior SOS and resume. We do see that TCU is far ahead of Baylor even though the Bears won that head-to-head matchup, but Baylor’s SOS is very weak by most metrics and it’s clear they were punished for playing one of the absolute worst non-conference schedules in the country. It is clear, though, in the case of Oklahoma and West Virginia, that head-to-head mattered as a tiebreaker. The resumes are very similar and West Virginia has the better top win (and the losses are about equal), so Oklahoma got the nod. The pair were also in the same ranking pod, so it’s clear there were discussions in regards to each other.
It is tough to pin down whether or not the committee used the “eye test” at all to help in these rankings. Arizona is ranked very low for a team with a decent loss, a great win, and a solid SOS. They are right behind Georgia, who has a worse loss and no win even close to as good. Either they are being punished for their poor early-season performances (they won all of those games but did not look good at all doing so) or they are not being given full credit for their victory over an Oregon team missing most of its offensive line. I would have loved it if Long had weighed in on their ranking in particular, but hopefully future rankings will shed more light on this.
What is clear, thankfully, is that ESPN’s FPI is not playing a factor in these rankings at all. There was a chance for there to be an extreme conflict of interests involved (if the network airing the games had an input in which teams were selected) and ESPN has been using their FPI more and more on-air as the past few seasons have progressed. The committee, it seems, is ignoring them the way they seem to be ignoring most computer metrics. Or, at the very least, none of these computer metrics are weighing heavily on the decisions.
One final thing we have to remember: one of the five main criteria for the committee is “conference championships won.” No one has won a conference championship yet. We could see an extreme shift in the final week if the committee decides to give conference champions a major boost, something not at all relevant before then.
Thank you for reading. You can follow me on twitter – @Yesh222. Support LWOS by following us on Twitter – @LastWordOnSport and @LWOSworld – and “liking” our Facebook page.
For the latest in sports injury news, check out our friends at Sports Injury Alert.
Have you tuned into Sports Events Guide Radio? LWOS is pleased to bring you 24/7 sports radio to your PC, laptop, tablet or smartphone. What are you waiting for?
