Named after the last selection in the annual NFL Draft, Pick 256 is a place to take a look at statistical trends and oddities in the NFL. The idea for this blog came after the 2009 NFL Draft. At the time, Patriots coach/GM Bill Belichick had made a number of draft-day trades amassing a number of second and third round picks (a strategy he has continued to use in subsequent drafts). After the dust had settled, pundits and draftniks alike questioned Belichick’s strategy. Had he missed on on “impact players” by trading out? Was he, in NFL terms, buying a lot of lotto tickets in the hopes of increasing his chances of a big score? Discussions went endlessly back and forth in cyberspace and in the media, but almost all of debates were short on evidence and heavy on faulty logic.
So, I decided I wanted to settle things once and for all – if only for my own personal satisfaction. So the first (of many) questions I asked myself was “how do we measure draft success”? If you’ve ever read an NFL messageboard you know that the words “bust”, “reach” and “steal” get thrown around so much that those terms have lost nearly all meaning. A 7th round pick who gets cut is a “bust”. A 2nd round pick who is merely a serviceable backup is similarly considered a “bust”. The problem lies in the nature of draft-day expectations. Any player drafted in the top 3 rounds is expected to be a high-quality player. Rounds 4-5 are generally thought to produce “good situational players” while the 6th and 7th rounders are “backups” or “projects”. Also, anyone drafted outside of the second round has “special teams value”. Anyone wh0 has followed the NFL Draft for awhile should know how crazy these expectations (and thus, post-draft evaluations) are.
In order to prove these expectations as being unrealistic, I first needed to find a way to evaluate large sample sizes of players. Unfortunately, football is, in many ways, the ultimate team game. Quantifying success in the NFL is very difficult to do using statistics, even for stat-accumulating players like QB’s, WR’s and RB’s. It’s even more challenging for offensive lineman and defensive players. The site pro-football-reference.com has done some great work coming up with a metric to evaluate players which they call AV (approximate value). While it’s a useful and mostly logically sound metric, it does have some serious shortcomings. For example, it doesn’t value kickers or punters at all and kick/punt returners are significantly undervalued. In light of these flaws, I decided to try to come up with my own player rating system so that I can answer nagging questions such as:
What should we expect from a 2nd round pick?
Is it true that WR’s bust at a higher rate than other positions? Are they just late bloomers?
QB’s drafted after round 1 – a good idea? Or almost always duds?
Is it really a waste to draft a RB in the 1st round?
Pick 256 is an effort to use statistics to improve our understanding of historical trends, performance and personnel decisions. It’s not realistic to think that numbers can definitively answer all of our questions, but it should serve as a good starting point. To that end, this blog is probably best thought of not as a place to end a debate but rather a place to start one.