Weird Science: Examining Adrian Peterson’s Strange Projection
After the success of How to Lose a Fantasy League in 10 Picks, I considered calling this 10 Things I Hate About Adrian Peterson. But I don’t hate Adrian Peterson. I love him just like everybody else. Plus, I could only find nine issues with him (I know, only nine problems with the greatest running back in the NFL).
If you’ve read Why Adrian Peterson Shouldn’t Be a Top 5 Pick, you know he owns the No. 8 projection according to the RotoViz Sim Scores. In that article I compared Peterson’s 2012 to the 25 best seasons of the last ten years and concluded that AP is a much better reality player than fantasy asset.
That was the appetizer. This is the main course. By the time you finish, you’ll understand why Purple Jesus owns such a low projection and why I think that projection is basically justified.
1) Peterson’s yards after contact numbers are not sustainable.
Adrian Peterson averaged 3.9 yards after contact per attempt in 2012. That’s about 0.8 YCo/A above his career average. This can be helpfully illustrated by pointing out that he averaged fewer YCo/A in 2011 than Ryan Mathews.
Here are the rest of the players who have averaged more than 3.5 YCo/A in a season since 2008: C.J. Spiller 3.6 (2012), LeGarrette Blount 3.7 (2010), Jamaal Charles 3.6 (2009), DeAngelo Williams 3.9 (2008).
I’m not suggesting Peterson’s tackle-breaking ability is fluky. It’s not. He may possess the best combination of tackle-breaking power and long speed of any back in NFL history (although he’s probably no Bo Jackson). I am suggesting that even for someone like AP the high yardage totals are a positive manifestation of randomness. Within the context of one season, it’s even possible for someone like LaGarrette Blount to bunch his runs together in such a way that the after contact numbers are extraordinary.
2) AP’s first step is overrated. Peterson averaged fewer yards before contact than Chris Johnson in 2013 despite receiving far better blocking. And last year represented a career high. He was quite a bit worse from 2008-2011. I don’t expect scouts or most fans to ever believe this, but hard core contrarians might be tempted to see his poor 11.49 Agility Score as representing a partial explanation.
3) Peterson wasn’t an efficient receiver.
Peterson averaged fewer yards per attempt on receptions than runs in 2012, a feat that requires transcendent ability in the running game and a complete lack of intuition when catching passes. An inability to perform in the passing game is also something suggested by Agility Score.
It’s possible that Peterson’s poor performance is fluky, merely statistical noise – he performed well as a receiver in 2009 – but it’s equally possible that this is a red flag. You can use the RotoViz QB-Receiver Efficiency Tool to compare Peterson with the underrated Toby Gerhart since Christian Ponder took over.
|Christian Ponder||Adrian Peterson||63||49||307||2||1||4.79|
|Christian Ponder||Toby Gerhart||48||38||284||2||1||5.81|
4) Peterson plays in a terrible offense.
This issue is concomitant to the phenomenon I explored in the original AP article. The Vikings simply do not score enough points or move the ball well enough to put him in the conversation for the top pick. In 2006, LaDainian Tomlinson rushed for 1800-plus yards, caught 56 passes, and scored 31 touchdowns. Peterson’s upside is curbed at a much lower level.
5) The Vikings will face more large deficits in 2013.
If you haven’t checked out Mike Clay’s amazing run/pass splits by game situation, it’s a must for serious fantasy owners but even more for people who are hungry for great football info. You’ll learn the Vikings ranked 26th in drop back percentage last year and third in percentage of time playing with the lead. It’s virtually certain Minnesota will be forced to throw more this season.
Before you draft Adrian Peterson in a league that awards points for receptions, make sure you run the projections for yourself. If you use AP’s career line for attempts per game, receptions per game, and touchdowns per game and project at his efficiency averages, you end up with approximately 300 points in a PPR league. That simply isn’t a strong enough line to draft him in the top five picks.
Read Jonathan Bales take on Martin and follow Jacob Myer’s lead in putting Charles into an Andy Reid offense. After doing that, I challenge you to come up with a projection for Adrian Peterson that doesn’t require an insane 6.5 yards per carry to push him above Martin and Charles.
7) Peterson’s historical comps collapse.
Is it a fluke that almost all of his comps imploded the following year? Probably.
David Deutsch, the leading pioneer in the field of quantum computing and an incredible read, advocates for a philosophy of science based upon good explanations. Now, when he’s talking about explanatory power, he’s basically talking about the exact opposite of the narrative fallacy. He’s talking about explanations that have reach and lead to further innovation.
I think this relates to the strange case of Adrian Peterson’s Similarity Scores. Simply put, they would be better if he hadn’t performed as well. You should be skeptical of such a conclusion, and we are at RotoViz. (The Fantasy Douche talks about this a little in his intro to the must read Multi-Year RB Projections.)
If you want to see how Adrian Peterson’s projection would change with slightly different stats, you can now do that with the brand new RotoViz RB Sim Score Lab. I strongly recommend playing with the tool and creating your own perfect Peterson. It may be that your version of AP does rank No. 1.
But even admitting that we don’t actually expect the relative collapse projected by the apps, most drafters aren’t giving it enough credence. And there are a couple of clear reasons why . . .
8) Peterson is injury prone.
Purple Jesus has developed a reputation as the most insanely talented healer in the history of the known world (after the actual Jesus). That may be true. It’s also true that he gets hurt. A lot.
Peterson missed part of two collegiate seasons with a high ankle sprain and broken collarbone, a factor that helps explain why the Arizona Cardinals passed on him for Levi Brown. AP has missed games in three of his six NFL seasons, including a fantasy playoff game in both 2010 and 2011.
9) Peterson is already getting old.
Running backs do not age in elf years. Is this less of a concern with Peterson than some other players? Possibly. His body’s freakish ability to rejuvenate itself after the knee injury suggests a type of physiological superiority that might carry over to age effects. If you were receiving any type of discount for his age – the type of discount James Goldstein received in the RotoViz startup draft – then his likelihood of flouting the age issue would represent a buying opportunity.
But you’re not. AP is being drafted No. 1 in almost all redraft leagues.
Keep in mind that age represents just one of a cluster of RB risk factors that pushes Peterson well down the aforementioned multi-year rankings. It may amount to nothing, but the data on RB efficiency strongly point to a steep decline almost from the first moment they arrive in the NFL. AP seems like a good bet as an exception, but the same could be said for Eric Dickerson, O.J. Simpson, and Barry Sanders.
Right up until they weren’t.