Jahmyr Gibbs, Bijan Robinson and Jonathan Taylor signed inside a week, each briefly topping the previous new-money average at the position. But all three are two- and three-year deals, not the five- and six-year contracts that made the previous generation of running back money disastrous. Front offices did not decide the analytics were wrong; they decided the term, not the rate, was what made the old deals bad.
For roughly a decade, the settled analytics position on running backs was that you do not pay them. Not that they lack value — the argument was always narrower and better than the slogan it collapsed into — but that the surplus value over a replacement-level alternative is small, the athletic peak is short, and every dollar committed to the position buys less than the same dollar spent almost anywhere else on the roster.
In early August 2026, three teams reset the top of that market three times in under a week.
Robinson signed first. Gibbs cleared him by $250,000 a year two days later. Taylor landed just underneath both. Read as a list of headline numbers, this looks like a market repudiating its own analytics. Read as three contract structures, it looks like something considerably more interesting: front offices conceding most of the analytics argument and pricing around it rather than against it.
The steelman version is worth stating properly, because the popular version is a caricature that is easy to knock down.
The claim was about surplus value — production delivered minus cap dollars consumed — and it rested on three empirical observations. First, the drop-off from an elite running back to a competent one is smaller than at almost any other position, because rushing production is heavily mediated by offensive line play and scheme. Second, running backs decline earlier and more sharply than other skill players, with the aggregate curve bending down somewhere around age 27. Third, the position carries high injury variance and a workload that compounds it.
Stack those together and the conclusion follows: a five-year deal signed at age 25 pays premium money for two good years and three years of decline you cannot cut without eating dead cap. The canonical failures of the last decade — long deals signed at peak, carrying guarantees deep into the aging curve — were failures of that exact shape. The analytics case was never 'running backs don't matter.' It was 'this contract structure destroys value.'
The distinction matters because it determines what would count as the market being wrong. If the argument was about the position, these deals refute it. If the argument was about the structure, these deals are a response to it.
Gibbs and Robinson signed three-year deals. Taylor signed two. Not one of the three is the five- or six-year commitment that characterised the previous generation of top-of-market running back money.
This is the whole story, and it is a genuinely different bet. A three-year deal signed by a back in his mid-twenties expires before the steepest part of the aging curve. The team is buying the years the research says are worth buying and declining to buy the years it says are not. The high annual value is the price of that option: the player accepts a shorter term and, in exchange, gets a rate that would have been unthinkable on a six-year structure, plus another trip through free agency while still young enough to be paid again.
Put differently, the front offices did not conclude that running backs are worth more than the models said. They concluded that the exposure — the long tail on a position that ages fast — was what made the old contracts bad, and that a short deal at a high rate prices the aging risk instead of swallowing it.
The guarantee structures point the same way. Gibbs's $51.5M guaranteed against a $67.5M base is a high guarantee percentage, but on a three-year term the guarantee runs out roughly when the production is expected to. Taylor's deal is the cleanest example: $39M guaranteed on a two-year, $44M extension, which keeps him in Indianapolis through his age-29 season and puts him back at the table before 30.
There is a second observation that most of the reaction has skipped. Gibbs and Robinson are not the workhorse ground-game backs the positional-value research was built on. They are dual-threat players whose receiving usage is a substantial and, in some weeks, dominant share of their value.
That matters for two reasons. Replacement level for a back who can run routes from the slot, beat a linebacker in space, and be split out to force a coverage decision is meaningfully thinner than replacement level for a back who takes twenty-two inside handoffs. And the receiving skill set appears to age somewhat more gracefully than pure early-down rushing, because it depends less on the specific physical qualities — short-area burst through contact, tolerance for accumulated hits — that decline first.
If you believe that, then these players are partially being paid as a hybrid position that the old positional-value framework does not cleanly capture. It is a real argument. It is also, conveniently, an argument that would justify paying almost any expensive back, so it should be held to evidence rather than accepted because it is flattering.
Short term does not make the bet small. $22.5M per year is roughly what a very good starting cornerback or a second receiver costs, and the injury variance at running back has not changed at all. A back who tears an ACL in year one of a three-year deal has consumed a third of the term and a large share of the guarantee before producing anything. The shorter structure limits the downside but does not remove it, and on a three-year deal there is less time for a healthy season to average out a lost one.
There is also a market-dynamics reading that is less flattering to the front offices. Three deals in four days, each one clearing the last by a small margin, is exactly what an agent-driven anchoring cascade looks like. Robinson's number set a reference point; Gibbs's camp needed to beat it by a visible amount; Taylor's slotted in just below. Whether the third team would have arrived at $22M APY absent the first two is unknowable, and the compressed timing is at least suggestive.
And the surplus-value argument survives the term change on its own terms. If the elite back is only modestly better than the competent one, paying $22.5M rather than $6M for the difference is still a poor trade, whether the commitment lasts three years or six. Shorter deals reduce the cost of being wrong. They do not make the underlying valuation right.
This is the useful part, because both sides of this argument have a habit of declaring victory on ambiguous evidence.
The honest read as of today is that the market did not reject the analytics. It absorbed them, and then priced the risk they identified rather than avoiding the position entirely. Whether $22.5M a year for three years is the correct price for that risk is a question three seasons will answer and no amount of August commentary can.
Jahmyr Gibbs of the Detroit Lions, by new-money average. His three-year extension carries a $67.5M base value worth up to $75.75M with $51.5M guaranteed, which comes to $22.5M per year. Bijan Robinson is just behind at $22.25M APY, and Jonathan Taylor at $22M.
DrafterDaily covers contracts, media rights and league economics by reading the structure rather than the headline number.
Subscribe to DrafterDailyThe NFL wants a 50-60% raise from its broadcast partners. Fox has said no on the merits. CBS cannot say anything at all, because an antitrust dispute in an unrelated deal has frozen its owner.
Sportradar is now supplying Polymarket with official data, live streaming and integrity monitoring across 20-plus competitions and roughly 300,000 matches a year. The counterparty argument survives. The stack argument is new.
A $245.3 million cap and a $171.2 million floor. Twelve clubs would have to add $617 million; eight would have to cut $578 million. Read those two numbers together and the proposal's real shape appears.