by Brian Fremeau
The college football season is nearly a month old, and preseason projection data is steadily being removed from my FEI ratings with more weight given to the results on the field each week. This week's ratings mark the point in the year when the influence of preseason projection data falls below 50 percent in my calculations. In three weeks' time, all preseason projection data will be entirely removed from the FEI ratings, and we'll have a clearer picture of the teams that have done the most to position themselves to meet, exceed, or fall short of those preseason projected expectations.
The top of this week's FEI ratings appears very similar to the top of the ratings published four weeks ago before the season began. Each of the projected top 10 teams remains ranked 11th or better this week, though the order of teams has shuffled somewhat. In addition to the preseason influence, those teams have generally played to expectations thus far. The only team projected outside of the top 10 that has crashed the party is the No. 7 LSU Tigers (preseason FEI No. 20). LSU is undefeated despite having played the nation's third-toughest schedule to date according to FEI. Among undefeated teams, the Georgia Bulldogs check in with the second-toughest schedule faced to date, 50th in the nation to date overall.
LSU still has a tough schedule ahead (seventh-toughest nationally, per FEI), with games against top-ranked Georgia and Alabama and a handful of other SEC challengers in the regular season yet to come. Still, the Tigers have already improved on their preseason projections with their strong start and have performed more like a conference contender than the team initially projected to be in the back of the SEC West pecking order. I doubled down on that projection by calling out LSU as the team most likely to fall short of our Football Outsiders Almanac projections, and I'm now very likely to miss wildly on that forecast.
I'm also very likely to miss wildly on my forecast of the team most likely to exceed expectations. I nominated the Florida State Seminoles in that category. Few teams have been less impressive than Florida State out of the gates, with embarrassingly non-competitive losses to Virginia Tech and Syracuse already on the books. Much of the rest of the ACC conference hasn't impressed either, and the Seminoles still have an opportunity to right the ship, but bowl eligibility isn't likely at this point, jeopardizing the nation's longest bowl appearance streak (36 straight bowls).
FEI preseason projections were off the mark on LSU and FSU, but my personal instincts on those two teams were even more off the mark. (Lesson here for me is to trust the data more than my instincts, even though the data missed as well.) The Tigers and Seminoles aren't the only teams that have deviated significantly from their preseason projections. In fact, 72 teams are currently projected to finish within one win of their preseason projected win totals, and 91 teams are projected to finish within 1.5 wins. That leaves 39 teams currently projected to finish about two or more games better or worse than initially projected.
On average, teams have shifted 13 spots in the FEI ratings over the first four weeks of the season. This week's top 10 has only shifted an average of 2.5 ranking spots in that time. This week's top 20 has only shifted an average of 4.1 ranking spots. But the rest of the rankings have been shaken up more significantly. The biggest positive ranking changes from the preseason belong to the Buffalo Bulls (preseason No. 84, current No. 39) and the North Texas Mean Green (preseason No. 104, current No. 59), each up 45 ranking spots in four weeks. The biggest negative ranking changes from the preseason belong to the Louisville Cardinals (preseason No. 30, current No. 81) and the Florida Atlantic Owls (preseason No. 46, current No. 98), both down more than 50 ranking spots in four weeks.
The tables below indicate the teams that have changed their projected win fortunes most positively and most negatively through the first four weeks of the season. These win likelihoods are a function of both the ratings change for the team itself, but also ratings changes for its respective opponents. In the tables, Pre MW indicates the average wins projected for each team against FBS opponents prior to the start of the season, RMW indicates the average remaining wins projected for each team against FBS opponents, and TMW indicates the total mean wins currently projected (the sum of the team's current FBS wins and its remaining mean wins).
|Teams With Most Positive Mean Wins Change From Preseason Projections to Current Projections|
|Teams With Most Negative Mean Wins Change From Preseason Projections to Current Projections|
There are a number of new coaches leading teams on the negative ledger here, and the respective fan bases at these schools in particular will need to be patient. Some of the teams in the positive ledger could fall back a bit as well. And it's still early, of course, with more than two-thirds of the FBS games this season yet to play.
FEI Week 4 Ratings
The Fremeau Efficiency Index (FEI) is a college football rating system based on opponent-adjusted possession efficiency. Preseason projections (43 percent weight in this week's ratings) are based on five-year results, recruiting success, and returning offensive and defensive production. Strength of Schedule ratings (SOS) represent the average number of losses an elite team (two standard deviations better than average) would have against the team's regular season schedule. Strength of Schedule ratings against opponents played to date (PSOS) and remaining regular season opponents (RSOS) are also calculated from the perspective of an elite team.Net points per drive (NPD) is the difference between each team's points scored per offensive drive and its points allowed per opponent offensive drive. Net starting field position (NFP) is the difference between the average starting field position for each team's offensive drives and its opponent's offensive drives. Possession success rate (PSR) indicates each team's percentage of possessions played that result in a score when in possession or a stop when the opponent is in possession.
|20||North Carolina State||2-0||.121||.79||72||.07||111||.72||58||2.20||8||2.3||39||.725||5|