Wednesday, March 11, 2009
Following up on Stafford
By Kevin Seifert
Posted by ESPN.com's Kevin Seifert
I'll say this much about Tuesday's post on evaluating Georgia quarterback Matthew Stafford's aptitude for the NFL: It drew your interest.
Reasonable people can debate the merits of the formula developed by Ryan McCrystal of ESPN Research. The point was always to provide some context for the ongoing public discussion about whether Detroit should take Stafford with the No. 1 overall pick of next month's draft. Why are there so many mixed opinions? This formula helps identify some of Stafford's perceived shortcomings.
I did my best to sift through the comments and mailbag. So here are a few follow-up notes:
- I should have included an important caveat that was originally part of Ryan's study. The specific scores for each quarterback proved relevant only to put them into broader categories. Never did we suggest, for example, that Matt Leinart was deemed most ready for NFL success because he rated the highest of any first-round pick over the past 12 years. There was no intent to rank the quarterbacks 1-31. Ryan only used the scores to place them in groups: Ratings of 20 or higher, 1-19 and 0 or below.
- The titles of each category were important. The first was "Strong likelihood of success." The second was "Hit or Miss." The third was "Busts." You can argue whether Atlanta's Michael Vick was a bust, but otherwise I think the titles adequately account for the variances within each group.
- I alluded to the fact that Stafford's score was dragged down by his struggles as a freshman, when he completed 52.7 percent of his passes while throwing seven touchdowns against 13 interceptions. Many readers thought this unfairly dragged him down when compared to a quarterback like USC's Mark Sanchez, who didn't play as a freshman. So here you go: If you only use Stafford's sophomore and junior numbers, he comes out with a 4.5, which places him in the "Hit or Miss" category.
- With that said, I don't think you can simply dismiss Stafford's freshman year. Doing so would negate whatever benefit he derived from playing as a freshman rather than debuting as a sophomore. Did his freshman experience help make him a better player as a sophomore? Or would he have produced the exact same performance if he had never played as a freshman? I don't think you can automatically come to the latter conclusion. Even if we concede on that point, is "Hit or Miss" the kind of label you want on the No. 1 overall pick of the draft?
- Many people asked why the only draft-eligible quarterbacks included were Stafford, Sanchez and Kansas State's Josh Freeman. That's an easy one. The study focused only on quarterbacks taken in the first round. At this point, that trio are the only projected first-rounders for 2009.
- I should note that the guys at Football Outsiders pioneered this type of statistical analysis. David Lewin, in fact, evaluates quarterbacks based primarily on completion percentage and career starts.