Enough, wrote Scott of Harrisonburg, Va. Apparently I had referenced one too many times that accuracy rates among college quarterbacks typically translate when they reach the NFL. Prove it, Scott said. Here's what he dropped in the mailbag:
My first born if you compile a list of *all* the quarterbacks that rolled through the NFC North (as starters since 2000?) and rank them by their college completion rate.
I've got my hands full already, so I won't be collecting that bounty or any other reward for this post. But Scott's question intrigued me. Accuracy is the single-most important characteristic of any quarterback, especially when you expand the definition to include decision-making. After all, throwing to the right person should increase the chances of a completion.
If a college quarterback finishes his career with a relatively low completion percentage, is it reasonable to expect he can elevate it at the professional level? Research performed by actual football statisticians, most notably the David Lewin, has suggested college quarterbacks who complete less than 60 percent of their passes at least warrant closer inspection by NFL teams considering drafting them.
There are obvious 2011 angles to this theory as the Minnesota Vikings plan to draft a quarterback next week. Washington's Jake Locker finished his career with a 53.9 completion percentage, having never completed better than 58 percent of his passes in any season. Meanwhile, Missouri's Blaine Gabbert has raised questions among those who have pulled apart his college career to find he completed only 44.3 percent of his third-down throws. Last season, he completed only 38 percent of his throws that traveled at least 15 yards.
So here's what I decided to do from an NFC North angle. Using Scott's suggestion as a guide, I looked up the college completion percentages of the 31 quarterbacks who were either drafted by one of our teams since 1999 or were the primary quarterback for at least one season in this division over that stretch. (Hat tip to totalfootballstats.com for the information and to ESPN.com blog editor Brett Longdin for helping me compile it.)
I'm not a statistician, and I don't think we should consider the information below a representative trend for the entire game. I just think it's an enlightening illustration from a pool of players most of us are familiar with. First, here is a ranking of NFC North quarterbacks based on their NFL accuracy, organized to show how the division's most accurate quarterbacks fared in college.
Again, we should be careful about drawing any firm conclusions from this data. But here are some points worth noting:
Five quarterbacks brought their completion percentage from below 60 percent to above 60 percent, with Brett Favre forging the most dramatic path. Jay Cutler was another notable climber. I didn't include the Minnesota Vikings' Joe Webb in that count because of his small sample size. The other 11 sub-60 percent quarterbacks remained there when they reached the NFL.
Context is critical for evaluating college completion percentages. In his original paper, in fact, Lewin noted that Cutler's 57.2 college completion percentage was misleading given the long history of much lower rates on Vanderbilt's perennially undermanned teams in the decade before Cutler arrived.
The most accurate college quarterback on this list is Brian Brohm, who completed 65.8 percent at Louisville. We all know how that worked out for Brohm, who was stunningly inaccurate during his practice time with the Packers and hasn't been much better in several stints with the Buffalo Bills. Another example of college completion rates not translating is Drew Stanton, who hit 64.1 percent of his throws at Michigan State but hasn't come anywhere close to that with the Detroit Lions.
Overall, 12 of the quarterbacks currently have higher completion percentages in the NFL than they did in college. A total of 16 dropped and three have not yet thrown an NFL pass. I realize there are huge discrepancies between the number of throws made by, say, Brad Johnson and Matt Flynn, but those are the raw numbers.
It's worth nothing that the NFC North's most accurate quarterback over this stretch, Aaron Rodgers, finished his college career at 63.8 percent.
If our small sample leads us in so many directions, I think it's fair to assume that NFL teams will look well beyond the percentages. Ultimately, teams must decide what factored in to both high and low percentages. Did Locker play on an overmatched team, as Cutler did? Were Gabbert's downfield receivers substandard?
On the other side of the spectrum, TCU's Andy Dalton completed 66.7 percent of his passes last season. Was that performance the result of pinpoint throwing and smart decisions? Or is he the next Brian Brohm or even a Tim Couch, quarterbacks who benefited from a college scheme that facilitated a high completion rate?
Once NFL teams establish that context, then they can begin examining the prospect's raw throwing skills. This is where concerns have arisen about Locker, especially. Why did he have so many passes that simply failed to hit their mark during his career?
ESPN analyst Jon Gruden made clear that "accuracy can be improved" but that Locker needs to "work hard" at re-establishing his fundamentals.
"Sometimes it's because of your fundamentals," Gruden said. "Sometimes you're out of rhythm, you're in the shotgun, you're underneath the center. Sometimes you're under duress and out of rhythm. Sometimes you're hurrying, you're playing too fast. You're anticipating congestion around you when maybe there isn't."
But Gruden warned: "Accuracy sometimes can be terminal. Sometimes you can't cure that. I think that's a big concern with Jake Locker, because he does miss some throws."
We can't conclude that Locker or Gabbert are destined for NFL failure because they sometimes struggled to complete throws in college. The necessary improvement has been made over the past decade in this division, be it from a Hall of Fame quarterback like Favre or a journeyman like Brian Griese.
But it's also a reason for pause. Based on our sample size, at least, it's more likely that a relatively inaccurate college quarterback will be inaccurate at the NFL level. Scott of Harrisonburg probably didn't need me to tell him that.