A Statistical Look at Drafting QBs: Who Should the 49ers Take?

Thanks to mikev, Fooch's post about Hill being named the starter turned Niners Nation into Niners Draft Day War Room. mikev's question was, "So who are the top QBs coming out of college this year?" Now, although I'm a Gator and follow them closely, I don't know my ass from my elbow when it comes to the rest of college football. What I do know, though, is statistics; particularly NFL stats and stat analysis.

Football Outsiders' yearly alternative stats bible, Pro Football Prospectus, had a seminal contribution a few years back by David Lewin on predicting the NFL performance of college QBs. Without boring you with the details, he made two major findings that, when applied mathematically to "QBs coming out of college," have come to be known as the Lewin Career Forecast (LCF):

After the jump, I'll tell you what the two most important college passing stats are for predicting NFL performance, apply them to the ill-fated 2005 49ers draft, and tell you which QB I think the 49ers should  take in the 2009 draft...

(1) Lewin found that there are only two college QB statistics that reliably, and independent of all others, predict NFL QB performance: completion percentage and games started. Comletion percentage is obvious: QBs kind of have to be able to hit their target. In (2) below, I explain why games started is also important. Generally, the magic numbers are a 60% completion percentage and 37 games started in college. As Football Outsiders points out, from 1996-2005, the worst QB (in terms of NFL performance) drafted in the first 2 rounds who achieved these standards in college was Eli Manning, and uh, he just won a Super Bowl.

(2) The LCF only applies to QBs drafted in the first two rounds because, as Lewin showed, pro scouts are actually nostradamus-like at separating the wheat from the chaff when given enough game film. The guys who go after the first 2 rounds are generally there for good reason (Tom Brady excluded). In contrast, when a team badly swings and misses on a high QB pick, it's almost always because there isn't enough relevant game film (aka not enough games started), which leads the team to focus instead on unbelievably unreliable predictors like height, arm strength, Wonderlic score, and performance at the combine/pro day.

Thankfully, several people posted lists of college QBs in response to mikev's question. What I'm going to do is look at how each one stacks up according to the LCF. But before I do, I want to apply it to the 49ers current QB situation, or more accurately, whether the choice of Alex Smith over Aaron Rodgers was the correct one given the LCF. Here's how they stacked up:

Alex Smith -- 66.3% COMP PCT, 23 GS

Aaron Rodgers -- 63.8% COMP PCT, 22 GS

Clearly, you can see that the correct answer to the question, "Alex Smith or Aaron Rodgers?" was, "Neither." Rodgers is having a nice season this year in Green Bay, but what the above stats suggest is that Smith and Rodgers projected to be the same type of QB in the NFL: an average one. So not only did the 49ers screw up the pick, they compounded the screw up by turning an average QB into an epic bust the likes of which hasn't been seen since Dolly Parton in "9 to 5." Nice job, guys.

Anyway, below is the list of QBs coming out this year according to those that responded to mikev's question. If I'm missing any, let me know, and I'll update the list. The games started values are a projection based on the QB starting the remaining games for his team this season (including conference championships and bowl games):

Graham Harrell (Texas Tech) -- 70% COMP PCT, 40 GS

Colt McCoy (Texas) -- 70% COMP PCT, 38 GS

Sam Bradford (Oklahoma) -- 70% COMP PCT, 27 GS

Chase Daniel (Missouri) -- 68% COMP PCT, 41 GS

Tim Tebow (Go Gators) -- 65% COMP PCT, 27 GS

Mark Sanchez (Southern California) -- 62% COMP PCT, 16 GS

Curtis Painter (Purdue) -- 60% COMP PCT, 44 GS

Matt Stafford (Georgia) -- 56% COMP PCT, 35 GS

Sorting the list by completion percentage, as I did above, you see that 7 of 8 QBs meet the 60% standard. However, sorting by games started, only 4 of 8 meet the 37-start standard, with another coming pretty close. Remember...what we want in a QB, according to LCF, is high completion percentage proven over a large number of games started. Both matter. Therefore, it's pretty clear there are 3 tiers of QB on the list:

Tier 1 -- Harrell, McCoy, and Daniel

Tier 2 -- Painter, Bradford, and Tebow

Tier 3 -- Sanchez and Stafford

Bottom line for the 49ers' 2009 draft: If the scouts project Harrell, McCoy, or Daniel as 1st or 2nd round picks, and one of them is sitting on the board when the 49ers pick in the 1st or 2nd round, they better not last past the 49ers pick.

Log In Sign Up

Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

By becoming a registered user, you are also agreeing to our Terms and confirming that you have read our Privacy Policy.

Join Niners Nation

You must be a member of Niners Nation to participate.

We have our own Community Guidelines at Niners Nation. You should read them.

Join Niners Nation

You must be a member of Niners Nation to participate.

We have our own Community Guidelines at Niners Nation. You should read them.




Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.