FEI Week 12 Ratings and Possession Margins
by Brian Fremeau
The push for the College Football Playoff has had very little in the way of drama over the last several weeks. The top four teams -- Alabama, Clemson, Notre Dame, and Michigan -- have held those positions for three straight weeks, and there have been few changes among the top ten in the same span. The seeming inevitability of this season has been focused on the Crimson Tide's pursuit of a sixth national championship under Nick Saban, but there have been few obstacles for Alabama to overcome, and very few remaining obstacles for the other playoff contenders as well.
Notre Dame will wrap up its regular season and likely claim one of the playoff spots this weekend; FEI projects the Irish with a 92.9 percent win likelihood over their rival (and reeling) USC Trojans. Clemson has a rivalry game this week against South Carolina followed by an ACC Championship Game against Pittsburgh; FEI projects the Tigers to have an 84.5 percent chance to survive both of those tests. After a likely win over Auburn this weekend, Alabama has a formidable challenge against Georgia coming up in the SEC Championship Game, but is still more likely than not to win out (67.6 percent according to FEI projections). Michigan has a tough test against rival Ohio State on the road this weekend, and would follow a victory up with a rematch against Northwestern in the Big Ten Championship Game; FEI projects the Wolverines with only a 52.5 percent chance of winning both, but that's still more likely than any of the teams ranked below Michigan in the playoff rankings winning out themselves.
Playoff contenders can be compared in a number of ways, and I'll throw yet another comparison tool into the mix this week. In the first year of the College Football Playoff, the committee repeatedly referenced the term "game control" in its discussions of the contenders, and perhaps not coincidentally, ESPN has produced a "Game Control" ranking in each of its playoff picture standings since 2014 as well. This ranking has drawn some criticism over the years, mostly due to the lack of clarity of its meaning. ESPN defines game control as "the chance that an average Top 25 team would control games from start to end the way this team did, given the schedule." ESPN ranks teams by this definition but does not publish the underlying math or ratings behind it, although it certainly appears to be a ranking method that credits teams for taking a strong lead in its games and not putting itself in position to have to come from behind to win.
I have been interested in the subject matter of game control, but have taken a different approach to measuring it. Instead of creating a single rating that encapsulates how successfully teams take and hold leads over their opponents, I'm more interested in evaluating their distribution of controlling leads by possession. What percentage of possessions are contested by a given team while tied, while in the lead, or while losing? What percentage of possessions are contested by a given team when in a more commanding lead of two, three, or four possessions? And how often does a team play from behind, by similar margins? I have posted a comprehensive distribution of these Possession Margins to date this season on my site.
What can be gleaned from this data? I haven't done enough work to understand if there is predictive power to be drawn from it, but I do think it helps frame (and reframe) narratives about how teams have been playing this season. Let's run through the remaining nine College Football Playoff contenders and examine them through the framework of Possession Margins. Note that all but one of the CFP top-10 are included in this rundown; LSU (No. 7 in this week's CFP rankings) is not included, as they have already been saddled with two losses and have almost no shot of overcoming them to get into the playoff mix.
CFP No. 1 Alabama Crimson Tide
Possession Margins underscore how dominant Alabama has been throughout the year. The Crimson Tide have held a lead on 90.6 percent of their non-garbage contested possessions this year, and have led by 17 or more points on a whopping 52.5 percent of possessions. They're No. 1 in nearly every Possession Margin split category, and have rarely been tested. (Note that these numbers only reflect FBS game possessions, so Alabama's less-than-impressive first half against The Citadel this past weekend is not included in the data. Alabama did go on to dominate the second half, of course, so the numbers wouldn't have changed too dramatically even if that game were included).
CFP No. 2 Clemson Tigers
Clemson ranks right behind Alabama in terms of dominance among the playoff contenders, having led by 17 or more points on 31.9 percent of possessions (third highest nationally behind Alabama and Utah State) and leading by 25 or more points on 16.4 percent of possessions (also third). The Tigers have had only one instance all year in which they were in real danger of defeat, falling behind Syracuse by nine points late in the first half and not claiming the lead again until 41 seconds were left in the game.
CFP No. 3 Notre Dame Fighting Irish
Unlike Alabama and Clemson, Notre Dame hasn't distinguished itself with eye-popping final scores, but the Irish have been in control of solid leads this season better than any other team outside of Tuscaloosa. Notre Dame has held a two-score lead (nine points or better) on 50.6 percent of possessions it has played this year, second only to Alabama. They have trailed opponents on only 6.4 percent of possessions played this year as well, again, second only to Alabama. What they haven't done as successfully is take control of a game and then turn it into a blowout, though they have been getting better at this as the season has progressed. Notre Dame has held a lead of 17 or more points on 34 contested possessions in their last four games, but only had eight such possessions in their first seven games.
CFP No. 4 Michigan Wolverines
Michigan has received a great deal of credit in national media circles for looking the part of a playoff contender ever since their Week 1 loss to Notre Dame. They have racked up a higher percentage of blowout wins than most of the other contenders, and have held a 25-point lead or better on 7.7 percent of non-garbage possessions (seventh-most nationally). But they haven't opened up such leads very quickly, and in fact, they have been tied or have trailed their opponents on 42.5 percent of possessions played this season. Their defense has often ultimately taken control and led the way to opening up second-half leads, but they have kept opponents in the game longer than all but one of the other playoff contenders.
CFP No. 5 Georgia Bulldogs
Georgia's possession margin profile has been among the best in the nation this year, though their one loss to LSU is a bit of an albatross -- the Bulldogs trailed by at least two scores for 19 out of the 24 possessions in that matchup. Still, Georgia ranks behind only Alabama, Notre Dame, and Clemson in terms of percentage of possessions in which they have held at least a two-score lead themselves.
CFP No. 6 Oklahoma Sooners
Like Georgia and Michigan, Oklahoma has only one blemish on its schedule, a 48-45 loss to Texas. That loss never got terribly out of hand, but Oklahoma held a lead on only one possession in the game (after taking a 7-0 lead). Oklahoma's explosive offense has led the way and overcome its defensive liabilities in most games this year, and the Sooners rank behind only Alabama, Clemson, and Ohio State among playoff contenders in the percentage of possessions in which they have held a lead of 25 or more points (7.8 percent of non-garbage possessions).
CFP No. 8 Washington State Cougars
The Cougars have only trailed by more than a single score on two total possessions this season, both in their 39-36 loss to USC on September 21. Washington State hasn't taken too many commanding leads, however, and rank second-worst among the playoff contenders in playing with a lead of nine points or more (30.9 percent of non-garbage possessions).
CFP No. 9 UCF Knights
I'm including UCF in the playoff contender list because they remain undefeated, but the committee hasn't shown their willingness to rank non-Power 5 teams according to the same criteria that they apply to Power 5 teams. Still, UCF is worth some discussion, as chaos over the last several weeks may provide a window in which they can move up in the standings. The Knights haven't opened up 25-point leads very often (only 3.6 percent of non-garbage possessions, worst among the contenders), but they have opened up three-score leads with high frequency (23.7 percent of non-garbage possessions, third-best among the contenders).
CFP No. 10 Ohio State Buckeyes
The Buckeyes have looked very good at times this year, but have looked very shaky often as well. They, like Michigan, have played a significant number of possessions this year while tied with or trailing their opponents (45.8 percent of non-garbage possessions, the highest percentage among the playoff contenders). But they have opened up huge leads at times as well, ranking fifth nationally (and third among playoff contenders) in playing with a dominant lead -- 9.7 percent of their non-garbage possession have been contested with the Buckeyes holding a lead of 25 points or more.
Again, this rundown and the underlying data is more descriptive of how teams have competed, but isn't necessarily indicative of how they will compete in the upcoming matchups that will ultimately decide the College Football Playoff field. And for me, it's a good reminder that there untold numbers of ways to evaluate the contenders, more narratives and counter-narratives to unearth, and more data to explore.
FEI Week 12 Ratings
The Fremeau Efficiency Index (FEI) is a college football rating system based on opponent-adjusted possession efficiency. Adjusted Possession Advantage (APA) ratings represent the per-possession scoring advantage a team would be expected to have on a neutral field against an average opponent, calculated as a function of current FEI overall, offense, defense, and special teams ratings.
Strength of Schedule ratings (PSOS) represent the average number of losses an elite team (two standard deviations better than average) would have against the team's regular season schedule to date. Offensive FEI (OFEI) is scoring value generated per drive adjusted for starting field position and opponent defenses faced. Defensive FEI (DFEI) is scoring value generated per opponent drive adjusted for starting field position and opponent offenses faced. Special Teams FEI (SFEI) is scoring value generated per possession by a team's non-offensive and non-defensive units adjusted for opponent special teams units faced. The team's record to date against opponents ranked in the FEI top 10 (v10), top 20 (v20), top 30 (v30), top 40 (v40), and top 50 (v50) are also provided.
Ratings and supporting data are calculated from the results of non-garbage possessions in FBS vs. FBS games.