FEI Week 16 Ratings
Now that the college football regular season has wrapped up and bowls are underway, I've begun to reflect on what a strange year it has been. Recalling the season tentatively starting in early September with only a handful of programs taking the field, we ended up with far more games played than I would have wagered on at that time, 508 games in total between FBS opponents through last Saturday. When we published the Football Outsiders Almanac in July, we posted projections anticipating a full regular season of 762 FBS games. When I posted my preseason projections two months later, only 382 games were on the books and four conferences hadn't yet opted into the season.
It's fun to check our work and see how right or wrong our initial projections had been in comparison to where things stand now. I published FEI ratings at the start of the season for all 130 FBS teams, thinking there was still an opportunity for the Big Ten, Pac-12, MAC, and Mountain West to reverse course and decide to have a season after all. But I also ran College Football Playoff projections assuming teams from those conferences were out of the playoff mix. The teams with the best chance to make the playoff in my numbers at the start of the season were, in order, Clemson, Alabama, LSU, Oklahoma, Notre Dame, Florida, and Georgia. Five of those six teams ranked among the top seven in the final CFP selection committee rankings, and three are in the field. (Ohio State ranked No. 2 in our preliminary FEI projections and would have been a clear playoff favorite as well).
Those were solid projections (LSU notwithstanding), but as has been the case in many seasons over the years, they weren't particularly remarkable or bold. The best teams in college football year in and year out tend to be the same teams year in and year out. Alabama and Clemson have each made the playoff field in six of seven seasons, Ohio State will be making its fourth appearance, and Notre Dame is in the field for the second time. The best of the best teams have separated themselves even further -- 14 of the 18 playoff games played to date have been won by either the Crimson Tide, Tigers, or Buckeyes, and given the odds against Notre Dame in the Rose Bowl semifinal, those numbers are highly likely to be 17 of 21 won by three teams when everything is said and done in January.
College football has always been dominated by a select number of programs. This isn't a unique circumstance ushered in by the playoff era, and it's not one that would magically vanish if the playoff format were to change. More access to a six-team, eight-team, 12-team, or 16-team field would certainly open up more potential pathways to a national championship for more teams. But even though Alabama, Clemson, and Ohio State aren't infallible, I think we would still expect two of those three teams to advance to the championship game no matter how we structured the tournament to get there.
As a college football fan, I'd like to see more teams given a shot. As a college football data geek, I'd like to collect more reliable information about the relative strength of all teams. As the season has progressed and I've reduced the weight given to preseason projections, a handful of non-Power 5 teams have risen in my FEI ratings. BYU, Coastal Carolina, and Cincinnati are all ranked in my top 10 this week, and none of the three were given as much consideration by the playoff selection committee as their performance metrics suggest they should. That's in part because we didn't get to see those non-Power 5 teams against Power 5 competition, and the committee is wired to hold that against them. The only Power 5 opponent faced by any of those three teams was Kansas (119th in FEI this week).
It's all of the rich non-conference data that's missing from the puzzle this year. From 2007 to 2019, an average of 209 non-conference regular season games were played each year between FBS teams. We only had 59 such games this season. An average of 35 non-conference games between Power 5 opponents were played in the regular season over the last 13 years, and we had zero such games in 2020. We quite literally do not know anything about the relative strength of Alabama, Clemson, and Ohio State heading into the playoff that isn't based in part on preseason priors. Alabama played 11 games against an exclusive SEC-only schedule. Ohio State played only six games against an exclusively Big Ten schedule. And Clemson's only non-conference opponent was The Citadel (FCS).
Maybe our priors are good and the relative strength of the SEC, Big Ten, and ACC conferences in previous seasons is an indicator of the relative strength of these conferences this year. Or maybe the bowl season is going to be a shock to the system. Most likely, as is usually the case in bowl season, there will be a healthy mix of both expected and unexpected results . As of this writing, only 27 bowl games plus the national championship final are scheduled to be played, and with COVID disruptions still possible, there's a good chance not every one of those will actually happen. As we said back in September, every game we get is precious, so let's enjoy them while we can.
2020 FEI Ratings (through Week 16)
FEI ratings (FEI) represent the per-possession scoring advantage a team would be expected to have on a neutral field against an average opponent. Offense ratings (OFEI) and defense ratings (DFEI) represent the per-possession scoring advantages for each team unit against an average opponent unit. FEI, OFEI, and DFEI ratings are based on a combination of opponent-adjusted results to date and preseason projections.
Net points per drive (NPD) is the difference between points scored per offensive drive and points allowed per opponent offensive drive. Net available yards percentage (NAY) is the difference between offensive available yards percentage and opponent offensive available yards percentage. Net yards per play (NPP) is the difference between drive yards per offensive play and drive yards allowed per opponent offensive play. Three different schedule strength ratings for games played to date are provided, based on current FEI ratings, representing the expected number of losses an elite team two standard deviations better than average would have against the given team's schedule (ELS), the expected number of losses a good team one standard deviation above average would have against the schedule (GLS), and the expected number of losses an average team would have against the schedule (ALS).
Ratings and supporting data are calculated from the results of non-garbage possessions in FBS vs. FBS games.