23 Sep 2008
by Brian Fremeau
The Fremeau Efficiency Index principles and methodology can be found here. Like DVOA, FEI rewards playing well against good teams, win or lose, and punishes losing to poor teams more harshly than it rewards defeating poor teams. Unlike DVOA, it is drive-based, not play-by-play based, and it is specifically engineered to measure the college game. FEI is the opponent-adjusted value of Game Efficiency, a measurement of the success rate of a team scoring and preventing opponent scoring throughout the non-garbage-time possessions of a game. Like DVOA, it represents a team's efficiency value over average. Only games between FBS teams are considered. Since limited data is available at the beginning of the season, the ratings to date are a function of both actual games played and projected outcomes based on the 2008 Projected FEI Ratings. The weight given to projected outcomes is reduced each week until mid-October, at which point the projections will be eliminated entirely.
The complete Week 4 Ratings for all 120 FBS teams can be found here. Southeastern Conference powerhouses Florida, Georgia, LSU, and Alabama each went on the road last Saturday and notched an impressive victory, leaving 320,000 agonized opposing fans in their collective wake. It is still too early to tell if any of those four games will prove to play a significant role in the national championship hunt, but all four teams appear fully poised to take on all comers, anywhere and anytime. Five games are scheduled to be played between the foursome, beginning with Alabama at Georgia this Saturday. Pencil in a likely sixth showdown in the SEC championship game, and the survivor (if there is one) of the near-round-robin will be well-tested come January.
USC, on the other hand, probably won't be tested. A week after their demolition of preseason contender Ohio State, the Trojans sat at home and watched its league-mates suffer another demoralizing weekend. UCLA and Arizona State continued their slide, and California is still picking up the pieces from their trip to Maryland in Week 3. Oregon appeared to be destined for the Pac-10 silver medal, but the Ducks flopped miserably at home against Boise State in Week 4.
In the PFP 2008 essay, I proclaimed USC's schedule to be among the toughest in the country, featuring a slate of 12 opponents ranked in the top half of college football. Five of those opponents are currently ranked in the bottom half, and none of the others appear to be actual big-time contenders. Even more so than the would-be Pac-10 challengers, the Big East has fallen off the face of the earth. West Virginia is a shell of its former Rich Rodriguez-led self and has yet to record a FBS victory in 2008. Undefeated Connecticut eked past lowly Temple in its only road game in Week 2. And the league's best team, No. 37 South Florida, struggled on Saturday to put away one of the worst teams in the country, Florida International. USF is currently ranked behind eight teams each from the Big 12 and SEC, six from the ACC, and five from the Big Ten. And a total of six non-BCS conference teams are currently rated ahead of the Bulls.
The FEI top 25 shake-ups from last week mostly reflect strong road play from teams on the rise and less impressive home play from others. Five non-BCS conference teams are now in the mix, with No. 13 Boise State making the biggest leap following their win at Oregon. No. 16 Utah boasts three road victories already this season, the most in the country. As a whole, 27 of 61 total road victories (and 20 of 31 road wins against BCS conference teams) have been recorded by the FEI top 25. No. 17 Missouri and No. 25 Colorado are the only teams in the FEI top-25 that have not yet won a true road game -- the Buffaloes get their first chance this weekend at Florida State; Missouri heads to Nebraska next weekend. No. 4 Georgia and No. 11 Wake Forest have each won two games against BCS conference teams on the road this year.
Last year, after Michigan fell dramatically to FCS-heroes Appalachian State, I was challenged to address what some considered a gaping hole in the FEI ratings: I disregard all games against FCS competition, win or lose. The masses have hollered ever since, "How could this possibly be sound?! Michigan has to be punished!" To date in 2008, a total of 77 results have been ignored by FEI, more than 30 percent of the major college football games played this year. That's a big percentage, but since most FBS vs. FCS games are now in our rearview mirror, it will shrink as the year goes on and settle at around 10 percent.
Still, can ignoring one out of every 10 games played by FBS teams in a season be justified? Wouldn't the ratings be better by including that data? They sure would -- if we could trust the data. The raw data, sure, we can use that. But how can we make opponent adjustments to those data points if we don't have any clue as to how good the opponent is? Including this season, FBS teams are 381-36 against FCS teams since 2003, with an average margin of victory of more than 30 points. And it isn't just BCS heavyweights dominating these games. Teams with a losing FBS record are 150-25 against FCS teams over the same stretch. Teams with two or fewer FBS wins in a season are 51-10 against FCS teams since 2003. The lightest of the FBS lightweights are fearsome bullies in that other subdivision. The Appalachian States of the world come along every once in a while and look to be at least as good as an average FBS team. And the worst FCS teams are getting their doors blown off by the dregs of the Sun Belt Conference. With a possible ranking of anywhere from No. 40 to No. 200-plus, it would be foolish to assign every FCS team a single value against which the FBS game results could be adjusted.
Could multiple values be assigned? Should 2007 Appalachian State be lumped in with, say, the best third of FCS teams and be assigned a value equivalent to No. 80? Either way, it's problematic. Grouping all FCS teams together severely distorts the value of the outliers, and separating them would be a completely artificial and subjective exercise. And at the end of the day, did the loss to Appalachian State really tell us anything about Michigan that we didn't learn from its other twelve games last year? The Wolverines lost to FEI No. 3 Oregon, No. 8 Ohio State and No. 35 Wisconsin. They defeated No. 5 Florida, No. 29 Illinois, No. 33 Michigan State and No. 38 Penn State. FEI ultimately ranked Michigan at No. 22. Appalachian State fell into the "Others Receiving Votes" category in the final AP poll. It doesn't appear to me that the Appalachian State loss was much of an outlier, but that's just my opinion. I don't have any real data to back it up.
2 comments, Last at 26 Sep 2008, 10:14am by blurdog