If the Bills utilize analytics, which QB will they draft?

Upon being named the man in charge of the Buffalo Bills, Russ Brandon discussed the importance of adding analytics to the Bills' evaluation process:

"Number one, I can promise you that this will be a forward-thinking, progressive and attacking organization heading into the future. One of the items that I am going to spend a lot of time on with Buddy, with Doug Whaley and with Jim Overdorf is we are going to create and establish a very robust football analytics operation that we layer into our entire operation moving forward. That is something that is very important to me and very important to the future of this franchise."

Taking Brandon at his word, that suggests that the Bills will incorporate some level of analytical analysis as they decide which QB to select in the upcoming NFL draft.

Some skeptics may believe that the team based nature of the NFL diminishes the impact of statistical analysis in player evaluation, compared to sports such as baseball or basketball where individual skills are more easily isolated. As a starting point to counter this argument, how successful have NFL teams, using traditional scouting methods, been at evaluating NFL caliber quarterbacks?

Since 2000, 36 quarterbacks have been selected in the 1st round of the draft. Of those 36, 19 have been able to establish and maintain their status as starting QBs. Essentially, selecting a 1st round QB is a 50-50 proposition- while the list includes Aaron Rodgers, Eli Manning, Matt Ryan, Joe Flacco and Andrew Luck, it also contains such memorable picks as Blaine Gabbert, Jamarcus Russell, Matt Leinart, Kyle Boller and David Carr.

The success rate plummets in the 2nd and 3rd rounds. Of the 15 QBs selected in the 2nd round since 2000, only 3 (or 20%) have become established starters- Andy Dalton, Colin Kaepernick and Drew Brees. This rate of three successful 2nd rounders per decade holds true in the 90's as well, with Kordell Stewart, Jake Plummer and Brett Favre the only quality 2nd round selections in that decade.

The 3rd round is even worse- while many Bills fans still mourn the selection of T.J. Graham over Russell Wilson, the success rate of 3rd round QBs is only 12% (2 out of 17). Before Wilson, the last successful 3rd rounder was Matt Schaub in 2004, and before that, Neil O'Donnell all the way back in 1990.

We don't know what analytical system the Bills might use, but for comparison purposes, we can use the Lewin Career Forecast 2.0 from the Football Outsiders website.

The Lewin Forecast heavily weights career starts and completion %, with smaller weight given to body type (BMI), career development, level of conference play, and how often the QB runs with the ball (passers who have a tendency to abandon the pocket and run with the ball are penalized in their NFL forecast).

The highest possible scores in the forecast are 2,000 points and up, with any QB with a 1500 score or better generating a very promising NFL profile. QBs who score between 1000-1500 are considered mid-range prospects, while scores below 1,000 points are considered red flags. Combining the top 9 QB draft prospects, along with 42 historical ratings for comparison purposes, gives us the following buckets:

QBs with outstanding analytical forecasts:

Russell Wilson, 2650

Robert Griffin, 2530

Philip Rivers, 2476

Landry Jones, 2276

Colt McCoy, 2092

Geno Smith, 2064

Carson Palmer, 1973

Matt Barkley, 1812

Andrew Luck, 1749

Andy Dalton, 1616

Brady Quinn, 1518

Ryan Nassib, 1506

Jason Campbell, 1506

Football Outsiders freely admits that the model has trouble with the Colt McCoy-Brady Quinn type of player, one who had significant success over a long period of time, highly accurate, a pocket passer, and played in a major conference. Nonetheless, 7 of the 9 QBs given analytic scores of 1500 or more points became regular starters in the NFL- a significantly better rate than the NFL success rate of drafting 1R QBs.

In addition, the model should be given a lot of credit for successfully predicting Russell Wilson's breakout, as well as predicting that Dalton- not Locker, Gabbert or Ponder- would be the best of the 2011 draft class (more on Cam Newton in a minute).

One of the most striking aspects to this top tier list is that analytics likes this particular draft class far more than the draft gurus- Mayock, Kiper, Cosell, etc.

Despite the model rating above, the FO team is not predicting that Landry Jones will be the best QB in this year's draft- this is where a blend of scouting and analytics comes in. Still a team that finds Nassib in the 2nd round, or Landry in the 3rd or 4th round, would be making a savvy choice per this analytical model.

Quarterbacks with mid-range forecasts:

Jay Cutler, 1444

Chad Henne, 1411

Matt Ryan, 1403

Nick Foles, 1391

Josh Freeman, 1367

Sam Bradford, 1345

Eli Manning, 1292

Brian Brohm, 1290

EJ Manuel, 1270

Kellen Clemens, 1248

Zac Dysert, 1131

Jimmy Claussen, 1062

Vince Young, 1059

Colin Kaepernick, 1044

Brandon Weeden, 1011

This mid-range is such a mixed bag of hits and misses, that I would suggest that when analytics is indecisive, let scouting take over as the primary decision lever. Zac Dysert may be a solid developmental pick in the mid rounds, even if he only turns into a good #2 QB. Just as the scouts can't make their mind about Manuel, neither can the model.

The Bottom Tier, the "red flags":

Tim Tebow, 925

Joe Flacco, 732

Ryan Tannehill, 730

Matthew Stafford, 714

Blaine Gabbert, 656

Jake Locker, 569

Ryan Mallett, 471

Tyler Wilson, 425

Christian Ponder, 413

Brock Osweiler, 248

Cam Newton, 175

Mark Sanchez, 151

Charlie Frye, 117

Rex Grossman, -124

JP Losman, -195

Tyler Bray, -201

Patrick Ramsey, -223

David Carr, -299

Josh McCown, -311

Mike Glennon, -379

Trent Edwards, -611

Brodie Croyle, -736

Alex Smith, -782

Yes, the model missed on Flacco and Stafford, and didn't know how to evaluate a one-year wonder and running QB in Cam Newton, but here's where the analytical model really shines. How many teams would have greatly benefited from avoiding the mistakes that the list above represents?

For Glennon, below average accuracy, leading the NCAA in interceptions, a decline in his senior year, and an unusual body type have me hoping that the Bills stay far away. Wilson shares the senior year decline- the FO website has an extensive article on their website at the abysmal success rate in the NFL of college QBs who had a down season in their final year- this seems to be one case where analytics and scouts agree, as Wilson appears to be sliding down the projected draft boards.

Bray's coming out early, combined with poor accuracy and completion rates has him looking like the second coming of JP Losman, not the next Jim Kelly.

Conclusion: if the Bills are using an analytical model similiar to the Lewin forecast, and if the Bills scouting department likes Smith or Barkley in the 1st, Nassib or Manuel in the 2nd, or Jones or Dysert in round 3 or later, they should probably take the plunge.

If nothing else, if only the Bills had bookmarked and occasionally checked out the Football Outsiders web site over the last decade, they would have avoided drafting either Losman or Edwards, and very well could now be sitting with Russell Wilson, Kaepernick or Dalton. I'd call that progress.

Just another great fan opinion shared on the pages of

Log In Sign Up

Log In Sign Up

Please choose a new SB Nation username and password

As part of the new SB Nation launch, prior users will need to choose a permanent username, along with a new password.

Your username will be used to login to SB Nation going forward.

I already have a Vox Media account!

Verify Vox Media account

Please login to your Vox Media account. This account will be linked to your previously existing Eater account.

Please choose a new SB Nation username and password

As part of the new SB Nation launch, prior MT authors will need to choose a new username and password.

Your username will be used to login to SB Nation going forward.

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 Buffalo Rumblings

You must be a member of Buffalo Rumblings to participate.

We have our own Community Guidelines at Buffalo Rumblings. You should read them.

Join Buffalo Rumblings

You must be a member of Buffalo Rumblings to participate.

We have our own Community Guidelines at Buffalo Rumblings. 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.