The College Football Playoff selection committee appears to be off the hook for now. The top four in Tuesday’s rankings will almost certainly comprise Alabama, Michigan, Clemson and Washington, with the only debate about the those teams relating to their order.
But chances are those four teams will not be left standing in the top four by season's end. Five of the eight teams in playoff position in the initial rankings the past two years did not make the playoff. Looking at this year’s crop, there is only a 16 percent chance that all four of those teams (Alabama, Michigan, Clemson and Washington) end the season as a conference champion with fewer than two losses, according to ESPN’s Football Power Index.
As the playoff debate heats up, the discussions over which teams are best and most deserving will almost certainly come to the forefront. We already saw a hint of that debate on College GameDay on Saturday and it will only heighten as the season progresses.
Defining best and most deserving
ESPN has two metrics that can help shape that debate: FPI (best) and strength of record (most deserving). Neither FPI nor strength of record is designed to predict the playoff field, but they can help answer two separate questions:
-- FPI: If two teams met on a neutral field, who would win and by how many points?
-- SOR: Which team’s W-L record is most impressive, given its schedule?
You can read about the difference between these metrics in this explainer, but in summary, FPI looks beyond a team’s win-loss record to determine how it won its games and whom those wins came against. It’s a forward-looking metric whose ultimate goal is to measure team strength and get its game predictions and season projections right.
Strength of record is a backward-looking measure of team accomplishment. Unlike FPI, strength of record doesn’t care about how a team won its games; it simply cares about the difficulty of a team’s schedule and the result (win or loss).
In listening to many analysts around ESPN and other media outlets, there appears to be two camps -– those that favor the best teams and those that favor the most deserving ones. Or to use a recent example, those that thought an undefeated Iowa last year deserved a top-four ranking and those that didn’t.
But what about the CFP selection committee? Has it favored the best or most deserving teams?
Has the committee favored best or most deserving?
Based on the first two years of the playoff, the CFP committee appears to have sided with the most deserving teams. As can be seen in the chart below, strength of record correlates more closely to the actual CFP Rankings than FPI. That is not an indictment of FPI, which has been accurate predicting games the last three years, but rather an observation about what types of teams the committee has valued.
As the season progresses, strength of record inches even closer to the CFP top four as the metric gains more information to evaluate a team’s résumé. Seven of the eight playoff participants ranked in the top four of SOR when the final rankings were released. The only outlier was 2014 Ohio State, which ranked behind Baylor and TCU on Selection Sunday.
Again, strength of record is not designed to predict the playoff field; it is an objective measure of accomplishment that happens to correlate closely with the committee’s top four. FPI and strength of record both have value, but when ranking teams, it ultimately depends on whether you favor the best or most deserving ones.
Best and most deserving teams after Week 9
Looking ahead to Tuesday’s CFP rankings release, how do the likely top four teams stack up in ESPN’s two main metrics?
Alabama, Clemson and Michigan all rank in the top four of strength and record and FPI. That means that at this point in the season, they have the most impressive résumés (SOR) and are predicted to be the strongest teams going forward (FPI).
After its win in Tallahassee, Clemson moved up to No. 1 in strength of record, but it’s a tight race between the Tigers and Crimson Tide. An average Top 25 team would have a 1.2 percent chance to start 8-0 against Clemson’s schedule, which includes three wins (two away from home) against FPI top 10 opponents. That same team would have a 2.2 percent chance to start 8-0 against Alabama’s schedule, which includes four wins (three away from home) against teams in the FPI top 20.
While Clemson may have a slightly stronger résumé, Alabama is expected to be by far the best team going forward. The Tide, who rank No. 1 in FPI, would have at least a 57 percent chance to beat any other team in a potential national championship game.
What might be surprising is that Washington ranks fifth in strength of record and seventh in FPI. The Huskies have played the fourth-easiest schedule of any Power 5 school and do not have a win against a team ranked in the top 25 of FPI. They have been dominant but have not faced the same tests as the other undefeated teams.
According to strength of record, it is actually harder to go 7-1 against Texas A&M’s schedule than 8-0 against Washington’s. The Aggies have played FPI’s 26th-ranked schedule and have a number of marquee wins. Their loss at Alabama is also forgivable, as an an average Top 25 team would have about an 11 percent chance to win in Tuscaloosa.
Similarly, even after accounting for its closer-than-expected win on Saturday, Louisville is expected to be one of the four strongest teams going forward. With Lamar Jackson, the Cardinals have the nation’s highest average offensive projection and ninth-ranked defensive projection.
Louisville and Texas A&M likely won’t end up in the top four on Tuesday, but keep an eye on where they end up to get a sense of what side of the best versus most deserving debate the committee is leaning toward.