Despite Loss, Bills Lead League in DVOA

By
Updated: October 5, 2011
DVOA100411-1

As expected, the Bills have fallen in all of the power rankings this week, which was to be expected considering power rankings are fairly simple function:

Power Ranking=Last Weeks Ranking +/- (Win/Loss)*Ranking of Team Played 

… with probably some sort of coefficient added for the writer’s biases. They are useful for determining the football media’s perception of the team and for someone who wants a quick gauge of the NFL who hasn’t been following closely. After the previous week’s shower of media attention I could live without the former, and if you’re reading this site I’m assuming you pay more than passing attention to the NFL and will subsequently find little value in the latter.

It’s probably not news that power rankings are fairly useless, but that doesn’t mean there’s still not a desire among critically thinking fans to gauge where the team stands relative to the rest of the league. That’s where Football Outsiders DVOA rankings can come in handy.

DVOA stands for Defense-Adjusted Value Over Average, and is essentially football’s best answer to the SABERmetric movement’s VORP rating that has turned Baseball analysis upside down. If you’ve read/seen the movie of Moneyball, you are at least familiar with this concept. Basically, the idea was to boil down the game of baseball to outs, measure how well a player contributed to avoiding making outs for his team, and measure it against a ‘replacement player’, roughly .8 of the league average. Since outs are what control the game of baseball, the idea is you can boil down the game to its element and quickly determine how much or little any hitter or pitcher is contributing their team’s success.

As you probably could anticipate, there are a number of issues that make such a concise rating impossible for football players . Football plays are much more complex than any particular one on one match up, and are comprised of 11 specialists with very different roles. How a weak side tackle blocks on a run play can’t be compared to a fly route a receiver runs, and furthermore it’s very difficult if not impossible to measure it’s direct impact on a given play so concisely as a hitter and pitcher can always be measured. There’s simply too much chaos.

While this may not be feasible to do w/r/t individual players (despite attempts that should be taken with a Rock of Gibraltar sized grain of salt), it is much more workable when describing entire offenses/defenses. If a baseball makes progress by avoiding outs with at bats (or making outs), a football team makes progress by moving towards first downs (or preventing the offense from doing s0). In every down, there is a set number of yards an offense must make in order to be considered successful. For instance, gaining 2 yards on a 1st and 10 would not be a very successful play for the offense, but on a 2nd and 1, it is a very successful play, as the team makes little progress towards a first down in the first situation and great progress in the second.

From there you can essentially begin to determine how effective teams are managing whatever situations they face. Now, the payoff. Buffalo is currently ranked #1 in the aggregate rankings for DVOA, despite losing last week. Essentially, Buffalo has a higher success rate on individual plays than any other team in the league.

It may seem like  just another complex stat that doesn’t mean anything, but consider this: last year, through week 4, the top 12 in FO’s DVOA rankings went 9/12 in making the playoffs, with the 13 and 14 both making the playoffs. So by this time last year, DVOA was doing a very good job determining who was a playoff contender, and who wasn’t. If you look at this year’s playoff odds, which use DVOA and DAVE (essentially DVOA adjusted for preseason expectations), Buffalo is given a 65% chance at locking up a playoff spot based on the data we have so far. It’s the first year FO has had this feature, so we can’t go back and check how they did last year.

While this all may seem very nerdy, look at it this way. Some very smart people, with a reliable, unbiased methodology for measuring a team’s performance are projecting Buffalo to continue to do well this year, even if they lost a tough game to seemingly below average opponent. I don’t know about you, but I certainly find that more substantial than a knee-jerk power ranking.