Not all NFL holdouts are created equal. This is especially true for running backs, who have seen their value decline in recent years, even as the salary cap has jumped prodigiously. This summer, a pair of backs began training camp by withholding their labor in attempt to leverage a strong second contract. One, Ezekiel Elliott of the Cowboys, inked the richest running back contract in history just before the start of the regular season. The other, Melvin Gordon of the Chargers, slunk back to practice last week with nothing to show for it beyond a few extra weeks of rest. Gordon didn’t get a deal, and he stands to lose up to $2.771 million in both fines and lost earnings. Both holdouts offer some basic lessons in negotiating strategy. But where do things go from here for running backs in an increasingly pass-oriented NFL?
Elliott had two years remaining on his rookie deal when he and the Cowboys came to terms on a contract with an average per year of $15 million, plus $28 million in full guarantees, and up to $50 million in rolling guarantees. Gordon had been set to earn $5.605 million this season, the fifth and final season of his rookie contract. He reportedly turned down an offer from the Chargers in the $10 million-per-year range, though the details about that offer’s structure (including guarantees) were never made public. Gordon has also missed a total of nine games due to injury. It doesn’t help that he’s sustained a torn meniscus (which required microfracture surgery), a sprained MCL, a bruised knee, and another sprained MCL since entering the NFL. Which is a good deal of wear and tear.* Next year, when he will be 27 and entering his sixth season, Gordon is eligible for unrestricted free agency. It’s hard to gauge what kind of deal he can expect.
Elliott had leverage in ways Gordon didn’t, and not just because he’s been more productive throughout his career.
Joe Banner, a former executive for the Eagles and Browns who made his bones by understanding the salary cap and resource allocation, explained this further in an appearance last week on Andrew Brandt’s podcast.
“You’ve got to know who you’re picking a fight with,” Banner told Brandt. “There are a group of teams in the league, as you and I know, that they’re just going to stick to their guns. They’re not going to worry about some short-term agitation in the locker room; they’re going to worry about long-term precedence and sticking to their beliefs. And the Chargers are one of those teams.”
The Chargers, in other words, were never going to budge, no matter what. There’s an echo here of what the Steelers did last year with Le’Veon Bell. The Steelers, like the Cowboys, like to retain home-grown talent. But other than for quarterback Ben Roethlisberger, they do not extend full guarantees beyond Year 1 as a matter of policy. They reportedly offered Bell $19.5 million in the first year and $33 million total after Year 2 of his extension, but Bell wanted a guarantee beyond the first year and was determined to test his worth on the open market. He sat out 2018 (and punted on $14.54 million in salary), only to find the Jets were his lone serious suitors once he finally hit free agency. He had to settle for $13.125 million in APY and just $27 million in full guarantees. He basically gave up $19 million to secure just $7 million in guarantees. It’s not hard to see why this happened.
Running backs have been devalued for a number of reasons. There’s the rookie pay scale, new to this CBA, which robs young backs of any bargaining power for up to four years—five in the case of first-round picks, a group that includes both Elliott and Gordon. There’s the physical toll of playing running back, particularly for three-down backs who can average 250 or more touches per 16 games played—a factor that further limits running backs’ earning potential, since their youngest, freshest years are restricted by the rookie scale. (Two years ago, ESPN’s Mina Kimes noted that “the average back’s rushing yards begin to fall off a cliff as soon as he turns 28; wide receivers, on the other hand, continue to perform at the same level until they turn 32.”) There’s the fact that running backs are plentiful and tend to be replaceable. And there’s the analytics trend that emphasizes the efficiency of passing rather than rushing—except in specific, narrow circumstances. It’s a trend that’s been exacerbated by rule changes designed to protect quarterbacks and defenseless receivers combined with a greater emphasis on shorter, quicker throws from QBs.
Last year, Todd Gurley and the Rams agreed to what was then the largest RB deal ever ($14.375 million in APY, $34.5 million payable by Year 2). A few weeks later, David Johnson got $24.68 million in full guarantees and $13 million in APY from the Cardinals. Both players then sustained serious injuries. With Gurley hobbled, the Rams still managed to get to the Super Bowl; in his absence, C.J. Anderson posted three 100-yard rushing performances, including in a playoff win. James Conner replaced Bell and put up 1,470 scrimmage yards for the Steelers (even though he wasn’t nearly as efficient as Bell has been throughout his career). Phillip Lindsay, an undrafted rookie, posted a 1,000-yard rushing season for the Broncos. The Chiefs released Kareem Hunt in late November, after video of him pushing and kicking a woman surfaced; they nearly got to the Super Bowl anyway. And the Patriots won another Super Bowl by relying on an adaptable scheme fueled by the deployment of multiple, low-cost backs. It’s thus easy for teams to assume the running back supply exceeds demand.
The trend lines have become clear during this decade. In 2011, the first year of the current CBA, Adrian Peterson set the bar with a contract worth $14.2 million in average annual value that included $36 million in guarantees. But until Gurley’s deal last year (since surpassed by Elliott’s), Peterson’s remained the standard. Owing to the growth of the cap, Gurley’s APY as a percentage of the cap at signing (8.1 percent) ranks just 14th all-time, well below what Peterson’s had been (11.8 percent). The top three earners among RBs, as a percentage of the cap, all date from the 1990s. And before Gurley, Johnson, Bell, and Elliott got their big contracts in the last 15 months, the Giants’ Saquon Barkley ($31.9 million) and the Jaguars’ Leonard Furnette ($27.1 million) ranked second and fourth all-time in contracted guaranteed money for running backs—in their rookie deals, just by virtue of having been selected in the Top 5.
Teams still run the ball a lot: 25.9 attempts per game last season, according to Pro Football Reference. But that’s down from 27.6 attempts per game back in 2001. They’re also throwing it more than they used to: 34.5 pass attempts per game last season, versus 32.6 passes per game in 2001. But how teams run the ball has also changed. The analytics community has spent years banging the drum that “establishing the run” has no correlation to winning games. In the introduction to each annual Football Outsiders Almanac, Aaron Schatz makes the following points about rushing success:
- Winning teams tend to run the ball a lot because they’re running out the clock to win games;
- A back tends to put up a 100-yard game because his team is winning, rather than the other way around;
- Rushing in short yardage tends to be more successful than passing;
- Play-action success is not dependent on rushing volume, or on rushing success;
- Rushing tends to depend more on the offensive line than pass protection depends on the quarterback.
Similarly, the data shows that passes to running backs are not as efficient as throws to receivers and tight ends, and that passes on early downs tend to have a better success rate than runs. Josh Hermsmeyer of Five Thirty Eight believes that Elliott is not all that valuable in short yardage in the red zone, in short yardage in the open field, and when closing out games—the three situations in which it’s best to run the football:
Situationally, Zeke is profoundly average, but some perspective here is probably needed. Situational running, while important, is relatively rare. Around 5 percent of Elliott’s carries came in the red zone in 2018. Fifteen percent came in situations when the Cowboys were trying to close out the game, and 14 percent came on short-yardage runs in the open field. The majority of Zeke’s carries—about 65 percent—occurred in other situations. The problem is that those other situations turn out to be awful times to run the football.
Hermsmeyer’s piece was published in mid-July, just before Elliott began his training camp holdout. His headline? “Ezekiel Ellliott is not worth the money he wants.” He was not alone in this thinking. Here was Schatz’s immediate reaction to the news Elliott had agreed to a lucrative deal:
Just last week, Banner shared an article published last year by Stanford data scientist Ed Fang that showed how rush efficiency does not correlate to wins. Through the season’s first four games, Gordon’s replacement, Austin Ekeler, caught 24 passes and scored six touchdowns, even though he rushed for just 3.9 yards per carry.
The NFL’s longstanding practice of placing specific value on positions, rather than on players—a by-product of the franchise tag, a device designed to suppress players’ earning potential—has long functioned as a quiet kind of collusion. The analytics movement’s focus on maximizing efficiency lends itself to concerns about team-building under a salary cap, rather than labor dynamics, let alone the joy of watching a talented running back like Elliott or Gordon just do cool shit when he has the ball in his hands. When Gordon at long last gets to free agency, this likely won’t bode well for him.
* This post has been updated to include details about Gordon’s injury history.