Lies, Damn Lies, and Advanced Statistics

Judging by the traditional statistics used to measure gridiron performance, it's a wonder UW managed to break the .500 mark in 2010. The Huskies finished #76 in the nation in total offense, and #96 in scoring offense at only 21.85 points per game. On defense, UW was #97 against the run and #70 in total defense, surrendering 29.31 points per game—roughly a touchdown more per game than the offense was generating. Washington gave up, on average, 22.31 yards more than it gained in 2010, and even special teams were subpar—a respectable 55^th in net punting, but 83^rd in kick returns and 102^nd in punt returns.

Can Washington make the purple line go above the yellow line in 2011?

With the Huskies in the lower half of the rankings in virtually every significant statistical category, and on the wrong end of both a points and yardage differential against its schedule, one may very well wonder how UW managed to notch more wins than losses in 2010. But it happened, probably because almost all of Washington's wins were close games, while four of their six losses were blowouts (Nebraska, Arizona, Stanford, Oregon). 

Click chart to enlarge

Using advanced statistics, however, the Huskies' 2010 season makes a bit more sense. Washington ranked #41 in overall S&P+ rating, which evaluates team's performance based on the success of plays run from scrimmage, adjusting for such factors as the quality of opponent, field position, and game situation. This ranking drops the Huskies neatly into a cluster of .500-ish teams, and behind only three with losing records (and each of those teams was just one win from a .500 or better finish (Oregon State (5-7), Georgia (6-7), Clemson (6-7)).



Similarly, the 2010 Huskies finished much higher--#51 nationally—in the Fremau Efficiency Index (FEI) than their total yardage and scoring statistics would suggest. Unlike the S&P+, which evaluates teams based on individual offensive and defensive plays, the FEI rating assesses a team's performance on the basis of entire possessions. Like the S&P+, the FEI ratings are adjusted to control for such factors as strength of opposition and game situation. But while the S&P+ formula produces an abstract number that is really only valuable as a ranking or in comparison with other teams, the FEI produces an efficiency rating that expresses the rate at which the team exceeds or falls short of its expected performance in a given situation. Washington's 2010 FEI was -0.004, meaning UW was almost exactly as efficient as would be expected, given the quality of their opponents (an FEI rating of zero indicates that a team has performed exactly as expected; a positive FEI means the team exceeded expectations, and a negative FEI means the team fell short of expectations).

Click chart to enlarge

The Huskies finished with a substantially similar FEI of +0.008 in 2009. With two consecutive seasons ending in FEI ratings of practically zero, UW has essentially achieved mediocrity quantified. The Huskies have a long way to go to catch national champion* Auburn and their +0.348 FEI, but being only two years removed from an 0-12 season (and woeful -0.256 FEI), this isn't necessarily a bad thing.


Unfortunately, two years of statistics is really not enough data to reveal meaningful trends, which is what we would need to make predictions for the upcoming 2011 schedule. At this point, it might be enough to just say UW should work on getting the purple line above the yellow line.