National Signing Day is complete, which means that the 2015 college football season is officially underway.
It also means that there is enough information available for ESPN’s Preseason Football Power Index (FPI) to determine the best and worst teams heading into next season.
Preseason ratings are historically flawed. Team reputations are generally overvalued, and the top historical programs often end up on top.
Preseason FPI takes a lot of the subjectivity out of the process. It uses data that has been found to be predictive of future performance to project how strong each FBS team will be in the coming season. Last year, the top three teams in preseason FPI – Florida State, Oregon and Alabama – made the playoff.
Top 5 teams in Preseason FPI 2015 season
It is important to note that this is the first version of preseason FPI. In the offseason, ESPN will gather updated information on returning starters based on injuries, suspensions, transfers, etc., increasing the accuracy of the predictions. A final version of the rankings will be released before next season along with team projections.
The projections will likely look different than the current rankings because the projections account for schedule. For example, LSU ranks third in preseason FPI, but that does not mean that the Tigers will win the third-most games or make the College Football Playoff. LSU plays a tough SEC schedule, including road games against Alabama and Ole Miss; therefore, the Tigers might not rank in the top 10 in projected win total even though they are one of the strongest teams in the country on a neutral field.
What goes into preseason FPI and why?
Four components are factored into the ratings: prior years’ efficiency, information on returning starters, recruiting and coaching tenure. Based on statistical modeling, these factors interact with one another and are assigned different weights depending on the team.
The most important features of the ratings are prior years’ performances using ESPN's efficiency ratings. The most recent year’s performance generally has the greatest impact, but previous years’ performances have also been found to be predictive. For example, Baylor and Texas A&M have ranked in the top 10 in offensive efficiency in each of the last three years, despite starting multiple quarterbacks during that time. The system and coach have an impact, which is why both of those teams are expected to have a top-five offense in 2015.
The second-most important factor is returning-starters information, which was provided by ESPN Insider Phil Steele. All teams – even bad ones - are helped by a large number of returning starters. Offensive and defensive starters are tracked separately, with special consideration given to teams returning a quarterback. In fact, a team returning a quarterback and at least three other starters results in about a three-point boost in offensive efficiency (equal to about three points per game), on average.
In 2015, Notre Dame is expected to return a Power 5-high 19 starters, including its starting quarterback. Despite finishing 35th in overall efficiency last year, Notre Dame’s unusually high number of returning starters, as well as solid recruiting and a sixth-year head coach, has the Irish ranked seventh heading into next season.
The third component of preseason FPI is recruiting. A four-year average recruiting ranking of four systems (ESPN, Scouts, Rivals and Phil Steele) is used to measure the talent on the roster. Whether teams recruit well because they are successful or teams are successful because of recruiting is a bit of a chicken-or-the-egg discussion. Nonetheless, recruiting makes a difference – though probably not as big of a difference as many would think – when measuring future performance.
LSU not only returns 15 starters (nine on offense), but also, based on its four-year average recruiting rank, has a talented pool of players. The Tigers might surprise some as the third-ranked team, but given whom they are returning, not many programs will field a more talented team than LSU.
Finally, coaching tenure is accounted for in the rankings, mainly to capture the impact of a new head coach. A first-year coach brings his own system, which likely is different from the one used by the former coach. Therefore, the year-to-year correlation with efficiency ratings will be smaller with a new coach. Depending on a team’s performance in prior years, a new head coach can have a positive or a negative impact on the projections.
With Will Muschamp as its head coach, Florida was weak on offense (ranked 77th in the last four years) and strong on defense (ranked fifth in the last four years). Jim McElwain, in addition to other factors such as returning starters, helps Florida’s offensive projections (57th) but hurts its preseason defensive ranking (11th).
For a more detailed explanation of what we learned while developing preseason FPI last year, click here. Keep an eye out for the newest version of the preseason rankings and team projections coming out before next season.