Home / General / Regression to the Mean and the Marginal Value of Running Backs

Regression to the Mean and the Marginal Value of Running Backs


Bill Barnwell, as part of an article on the decline of several star running backs, says about Adrian Peterson’s off year:

I wrote about Peterson’s likely decline before the season, and he’s hitting a lot of those boxes this year. The big plays are down. The average play is down. He’s beginning to deal with nagging injuries, notably a groin injury that Peterson admitted was slowing him down this past week. (Peterson also had to cope with the tragic death of his child last month.) The quarterback play hasn’t helped, given that the Vikings have played musical chairs under center this year, but their passing attack wasn’t all that great during the second half of the 2012 season, either. That was with Christian Ponder throwing to a group of receivers without Percy Harvin (or 2013 arrivals Greg Jennings and Cordarrelle Patterson), and Peterson still put together the greatest half-season a running back has ever managed. Last year, Peterson ran through nine men in the box and still picked up big plays. This year, nine men in the box have slowed Peterson down. There’s no shame in that. They call them career years for a reason.

We’ve been through this with Marshawn Lynch, but (especially with Peterson’s performance having recovered a bit over the last two weeks) it’s not hard to see what’s going on with him:

Year YPC
2007 5.6
2008 4.8
2009 4.4
2010 4.6
2011 4.7
2012 6.0
2013 4.6

Peterson is having a fairly typical Peterson year, in other words — 2012, much more than 2013, is the outlier. This isn’t a knock on Peterson — a back you can count on being well above-average every year is a very rare commodity. But it also illustrates the compressed range of performance for running backs. If you win the lottery and get Peterson, you get a guy who will be nearly a yard a carry better than an easily available back in his typical year and two yards better in his peak seasons. That has real value, but it’s not a huge impact. If 2012 were typical for Lynch and Peterson — if great running backs could be counted on for 6 yards a carry and good ones for 5 — then running backs would be worth big contracts and top 5 picks with reasonable frequency. But in fact the typical performances are pegged about a yard lower than that, and when you combine that with running backs generally being more inconsistent and having shorter shelf lives than other skill position players…this is why no competent organization would draft a running back in the top 5 (let alone trade up to do so) unless they were very confident they had another Barry Sanders-level historical great on their hands. And if you think a guy who was only a starter for one year because he was playing behind…Mark Ingram as a sophomore had any serious chance to be the next Barry Sanders, well, you probably also think that a pushing-30 Brandon Weeden had potential as a franchise QB, and you’re probably also no longer employed.

All of which brings us to one of this blog’s longest standing pastimes, making fun of Gregg Easterbrook. The Browns somehow getting a first round pick for Trent Richardson might be the biggest heist pulled off by a North American sports GM since the Blue Jays somehow got Mike Napoli in exchange for letting the Angels pay Vernon Wells an elite salary for sub-replacement level performance. Getting this trade right at the time is a fairly elementary test for an NFL analyst. Easterbrook’s reaction was predictably hilarious:

In my AFC preview, I noted of the new Cleveland management team, “Rob Chudzinski at coach and Michael Lombardi at general manager traded away fourth- and fifth-round picks to bank extra selections for 2014. Strong teams bank draft choices; for a weak team to bank draft choices is a head-scratcher.” Scratch your heads anew over last week’s decision by Chudzinski and Lombardi to trade the team’s young star, Trent Richardson, to Indianapolis in order to bank another pick for April 2014. The Colts are likely to have a winning season: giving a high 2012 first-round pick for a low 2014 first-round pick is a bad deal unless Richardson is a bust. If he’s not a bust and churns out yards for Indianapolis, Cleveland faithful will be hit over the head yet again.

First of all, I don’t understand why it’s a “head scratcher” for a weak team to bank draft choices. But the real marvel is Easterbrook describing Richardson as a “young star.” Not even “promising,” but after 17 games of extremely expensive sub-replacement level performance, as a “star.” And it’s not even a question of a disagreement between advanced and conventional metrics: his conventional stats were terrible, his advanced stats were terrible, he looked like a highly unimpressive plodder — there was as of the time he was traded no evidence that he was a good running back, let alone a “star,” apart from the tautological assumption that if an already-deservedly-fired Mike Holmgren thought Richardson was worth trading up to draft at #3 he must be onto something. Would anyone have described JaMarcus Russell or Vernon Gholston as a “young star” after 17 lousy games and suggested that a team that managed to unload their contract and get a first rounder in exchange probably blundered? But certain ground-and-pound nostalgiacs evaluate running backs the way Joel Seigel and Peter Travers evaluate Hollywood moves — almost everyone’s a star!

In fairness, Easterbrook does hedge his bets. If Vernon Wells wasn’t a bust, Blue Jays fans would have been hit over the head yet again…

  • Facebook
  • Twitter
  • Google+
  • Linkedin
  • Pinterest
It is main inner container footer text