Statistics in an Uptempo Era
Larry Fedora's uptempo system and others like it are the primary reason these stats—and many other traditional measurements—are now obsolete. When one team averages 75 plays per game and another 60, per-game stats are skewed to such a degree that they are nearly meaningless when comparing the teams, since one team is essentially playing an extra quarter per game.
Because of the differences in pace of play, it's no longer as simple as looking at the final score and concluding that a defense that gave up 35 points played especially poorly or that an offense that scored 40 was a juggernaut. Think about it this way: if one team scores 40 points on 20 possessions and another scores 20 points on 5 possessions, which offense was more efficient and successful? The offense that scored 20 points—despite scoring half as many points—was twice as effective as the 40-point offense. Conversely, the second defense only gave up 20 points but was significantly worse than the one that gave up 40 points. This example highlights the only thing that really matters in football: efficiency.
Essentially, every extra possession an offense gets means an extra possession that team's defense ultimately has to face, meaning what really matters in terms of winning football games—and measuring a team's success—is efficiency on a per-drive and per-play basis. This means that North Carolina fans who are still adjusting to Fedora's new system must forget per-game measures in favor of per-play statistics and other more advanced efficiency metrics.
Basically, instead of looking at game totals, what really matters is how a unit performed given the number of opportunities (plays and possessions) it had. It's also important to account for the quality of opponent, as it's a whole lot easier to put up good stats vs. Idaho or Elon than against an ACC foe. This is all especially important when looking at the defense, which is quite simply going to give up more yards and points in the equivalent of an extra quarter of play than they would at a slower pace.
A simple way to find stats better suited to the modern game is to look at per-play data limited to BCS conference (or FCS) opponents, easily accessible through sites like cfbstats.com. By these measures, neither side of the ball was quite as good as its overall per-game numbers might suggest, with the offense 13th in yards per play (YPP) vs. BCS AQ competition and the defense at 69th in YPP allowed.
More complex metrics adjusted for opponent strength and pace of play—such as those provided by Football Outsiders—can also be helpful and will be used regularly throughout this column this year. Unlike raw statistics like YPP, however, these metrics are not quite as easily understood. In brief, Football Outsiders provides three separate ratings: the FEI, S&P+, and F/+. (Yes, they could afford to give them less wonky names to make them more appealing.) One benefit of these ratings is that they eliminate "garbage time," essentially when games are out of reach, and adjust for opponent strength across the board.
To put it simply, the FEI measures efficiency on a per-possession basis, with a unit rewarded for producing higher efficiency against its opponents than they ordinarily give up and penalized for lower efficiency. S&P+ measures efficiency on a per-play basis, also accounting for the expected rate of success considering field position (e.g. a play from inside the red zone has a higher expected point value than one on the other side of the field). F/+ combines the two measures and will be the preferred metric of choice in this column the rest of the year.
Looking Forward to 2013
UNC's 2012 F/+ rankings were 20th on offense, 74th on defense, and 44th on special teams. Those numbers are a bit lower than might have been expected given the oft-cited per-game stats, but they provide a better sense of where Carolina was last year and can help set good benchmarks for this year.
One might ask at this point why, if efficiency per play or per drive is all that matters, Fedora wants to run at such a high tempo. What benefit is there to scoring more points if it only means your defense will allow more points? The answer is that Fedora believes—based on good evidence—that increasing the tempo can help improve per-play efficiency. Not only does higher tempo force defenses to play simpler looks—since they don't have time to disguise or communicate complex schemes—a team used to playing at a high tempo can theoretically exhaust a team used to a slower pace into mistakes, thus producing much better efficiency in the fourth quarter than in the first. In addition, a better team benefits from a higher tempo in the same way the casino benefits from a gambler playing more hands of blackjack—when the odds are in your favor, you want as many plays as possible.
What then does this suggest in terms of realistic expectations for the 2013 Tar Heels? First of all, as good as the offense looked last year—a big improvement from 49th-ranked offense in 2011—there is still significant room for improvement in 2013. Given greater depth at receiver and a quarterback who better understands the system, it is reasonable to expect this year's offense to be among the top 10–12 in the F/+, provided the new pieces on the offensive line develop as they should. Defensively, there's even more room for improvement, and I think a move into the top 45–50 is a reasonable expectation. To put that in perspective, those numbers are comparable to Clemson's 2012 figures on offense (7) and defense (51). Over the next few columns, we'll assess what that might mean for UNC's expected record in 2013 and why it just might be enough to beat South Carolina in the opener.
Jason Staples was a walk-on wide receiver at Florida State in the early part of the last decade and has provided in-depth football analysis on the Scout network since 2007. A member of the Foootball Writers' Association of America, he is presently finishing a Ph.D. at UNC-Chapel Hill and will be providing scheme and stats analysis for InsideCarolina in 2013.