.png)
August 27, 2006
KEEPING SCORE
Numbers Often Lie When It Comes to Football
By MARTIN B. SCHMIDT
In 2001, after four seasons in the N.F.L. and with little success, linebacker Mike Vrabel was frustrated. Drafted in the third round in 1997, it seemed that the Steelers could not find a use for him. Maybe it was time for law school, he figured.
But thinking about law school is about as close as he came to leaving the N.F.L. Someone did find a use for him: Patriots Coach Bill Belichick, who, as Sports Illustrated reported, was impressed with the way he played the power block and with his understanding of the game. It was not Vrabel’s time in the 40-yard dash, or his vertical leap, or his total number of tackles.
Since then, Vrabel has been one of the Patriots’ most productive players, helping them win three Super Bowls, and scoring touchdowns in Super Bowl XXXVIII and Super Bowl XXXIX. Not once has he made the Pro Bowl.
He is a good example of how little an individual’s statistics mean in football. Numbers like the quarterback rating, yards per rush and total tackles are so closely tied to the performance of others that they become almost meaningless.
If Vrabel does not break through the offensive team’s blockers, the running back may gain 15 yards instead of being thrown for a loss of 2. Yet the running back gets the statistics. If Vrabel fails to break through the pass defense, the quarterback completes the pass 30 yards downfield rather than taking a sack. In this case, the quarterback gets the statistics.
In the N.F.L., if your teammate fails to perform, if the defense happens to be in the wrong place or if the coaching staff makes a bad call, your statistics will be affected.
Take the quarterback ratings for Joe Montana. In each of the 13 seasons he played regularly, his rating was above average. But in eight of those seasons, his rating averaged 86, not much better than the league’s average. In the five others, it was significantly higher, near 100, among the league’s best. Why such variance?
Obviously, some was due to variability in his performance. Athletes can have bad days. But one would expect more of this variability from game to game rather than from season to season.
The rest is due to factors beyond his control. A quarterback can control the direction and speed of the ball, but he cannot block his blind side, nor can he compel a receiver or a running back to run the correct route. A quarterback’s statistics are an outgrowth of the his actions, as well as the actions of his 10 teammates, the quality of his opponent’s defense and the quality of plays called by his coaching staff.
How much of a quarterback’s rating is predictable? Not much. If we look at six seasons of quarterbacks who threw at least 224 passes in successive seasons — the minimum to be ranked in the N.F.L.’s quarterback rating — a past rating has a poor predictive value.
Take Peyton Manning. He produced a quarterback rating of 104.1 last season. This season, based on the six seasons of ratings, there is a 95 percent chance that his rating will be between 73 and 111. In other words, he is going to be either pretty bad or pretty good.
Maybe the problem is with the quarterback rating itself. After all, it captures only what a quarterback does with his arm. The rating is complicated, but in the end, it does not provide a full picture of a quarterback’s production. Few would doubt Michael Vick’s impact on a game, but his rating has hovered around the low 80’s for much of his career. In 2002, when he was so dazzling in passing for 2,936 yards and running for 777, his rating was a moderate 81.6.
A better measure for a quarterback would incorporate passing attempts, sacks, rushing attempts and turnovers. But even with these improvements, there are too many variables to create a reliable rating.
So beware, fantasy football players. Last season, three former Pro Bowl quarterbacks changed teams. Daunte Culpepper moved to the Dolphins, Steve McNair to the Ravens and Drew Brees to the Saints.
What kind of seasons will they have? My guess is that one will be great, one will be average and the third will struggle.
Who will be great? I have no idea. On second thought, I’ll take McNair or Brees; neither plays Belichick’s team.
Martin B. Schmidt is an associate professor at the College of William & Mary and the co-author of “The Wages of Wins.”
