With the college football season just slightly past the halfway point, it's time to discuss some of the storylines flying a bit under the radar.
Florida State is more likely to lose than not
Following the Noles' controversial win over the Irish, the common theme running through the college football media is that Florida State has passed its toughest test, and will ride out the rest of its schedule easily. However, much like when we were all worried about the top four teams all winning out in 2012, this seems like a bit of an overestimation of the superiority of the top teams. While FSU does appear to have the best chance of winning out, it should be noted that ESPN's odds of this happening are less than one in three.
Why this difference between popular opinion and reality? For one, Florida State hasn't been able to live up to the standard of last year's team, ranking 10th in F/+ and going just 1-6 against the spread. Their 3-0 record in one score games* suggests that they've been a touch lucky to still be undefeated.
*On one hand, those three games came against two great teams (Clemson and ND), and one decent team (Oklahoma State with JW Walsh healthy). On the other hand, none of them were road games.
The other factor is that their remaining schedule is surprisingly decent. While the ACC rightfully gets maligned compared with other power conferences, Florida State's remaining conference opponents represent the better part of the conference. Road games against Louisville (15th in F/+) and Miami (26th and climbing after a big Thursday night win) represent major tests, while BC (32nd) and Virginia (30th) shouldn't be taken lightly themselves. While F/+ is a bit of an outlier in Massey's ratings compilation, the composite ratings still place FSU below the SEC West stronghold, and all five of the Noles' remaining opponents make the top 50. There is certainly a decent chance that Florida State wins its second straight title with another remarkable undefeated season, but there is a far better chance that that does not happen. You've been warned.
Auburn as a test case
As it is the first year of the playoff, there has been unending speculation concerning exactly what items will be discussed, what criteria will be most heavily weighted, and what the committee will do in certain situations. The chatter from committee members concerning simple heuristics like conference championships and schedule intent has lessened my faith in them. All that said, if the committee makes the correct choice with one potentially very interesting team, then it will restore much of my trust. That team is the Auburn Tigers.
Auburn currently finds itself positioned in the top four of the aforementioned Massey composite (#4), and is also in that same range in the metrics I most commonly use (3rd in Sagarin, 5th in F/+). Based on how they've played thus far, they are clearly in the running to be one of the four best teams.
Unfortunately for them, their November is going to be one of the most difficult months for any team in college football history. Having already suffered a road loss to the current #1 team (Mississippi State), Auburn must still travel to top ten teams Ole Miss, Georgia, and Alabama. Even though Auburn is a great team, this insane road schedule makes it quite unlikely they will be playing for a conference championship. It does not, however, make it unlikely that they will finish as a top four quality team.
Consider the following possibility: Let's say Auburn loses next week at Ole Miss (thus eliminating a realistic chance at a league crown), and then goes to town on the rest of their schedule. This scenario would put Auburn at 10-2, with a road record of 3-2 against 5 teams currently ranked in the top 11 of the AP rankings. Additionally, decisive wins over Bama and Georgia would likely mean that their computer numbers (which in theory represent the true quality of a team) would actually go up, placing them soundly in the top four.
If all this goes to plan,* I am all too prepared for the committee to overlook the Tigers in favor of a less-qualified conference champion. I've heard the argument many times, most recently from The Solid Verbal, from which I paraphrase: Every team has a chance to win their division or conference and thus if they can't do that, they probably shouldn't be considered for the playoff. The fundamental problem with this is that not every team gets the same chance to win their division. While Ole Miss gets all of their most difficult opponents at home (and misses out on having to play Georgia), Auburn will have had to play the four best SEC teams not named Auburn, all on the road. Even for a team as good as the Tigers, we shouldn't expect anything better than a 2-2 record from that set of games. When and if the Rebels clinch the division title, we won't be able to say they didn't earn it, but we can definitely say that they had a bit of an easier road.** While I have my doubts about the committee (see below), I hope they are able to appreciate a unique case like the version of the 2014 Auburn Tigers presented here.
*And if it doesn't, there will probably be another SEC West team with a similar, if less extreme case.
**If you are a top-five caliber team, you have a better chance of a successful season (going 12-0 or 11-1) playing a bunch of somewhat difficult games (road games against average teams, home games against better teams) than you do with a boom or bust schedule like Auburn's. That's why some sort of degree of difficulty metric would have use in a situation like this. Unfortunately...well, just read the next section.
Stats and the committee
If you follow any of the most prominent college football writers, then you've likely heard that the playoff committee recently held a mock selection exercise, much like the NCAA does yearly for the basketball tournament. As a fan, the concept of openness and transparency is much appreciated, but it also shines a light on some of the troublesome areas to team evaluation. Most notably, it appears that there is no firm guidance on determining strength of schedule, nor is the committee allowed to use margin of victory or any outside statistics.
Obviously, this is troubling. We live in an era where the number of people spending their free time analyzing and dissecting the game is orders of magnitude greater than at any time in the past. Conducting perhaps the most exciting development in the sport with minimal, if any, influence from that portion of the devoted is unwise and quite frankly insulting.
What's worse is that the more you think about this, the more you realize how problematic the whole idea of not discussing outside statistics is. Even if discussion of these metrics won't be allowed in the room, committee members that have done their work will have almost certainly consulted some sort of outside numbers to learn more about teams. For example, Barry Alvarez has said that he is working with a statistician at Wisconsin. This will no doubt influence his perception of which teams are best, which will most likely factor into his contribution towards the committee's discussion. While I admire his commitment to learning as much as he can, I also don't necessarily trust his numerical-related conclusions without examining them. And if the committee isn't allowed to discuss outside stats, then that won't happen. This leaves us with a set of statistics provided to the committee that isn't particularly great, the inability of the best minds to influence the discussion, and the unfettered ability of unproven and unchallenged stats to influence the proceedings of the committee. While the "true" top four* and the committee's top four will likely overlap (perhaps a great deal), I am still a little worried about the process to which we've trusted our favorite sport.**
*Sure, any halfway stat-literate person would say that a 12-13 game schedule provides no way for us to determine a definitive top four, but we can at least weed out enough teams to make a pretty good guess.
**Still better than the later versions of the BCS "formula," probably.
No comments:
Post a Comment