The running back position in the NFL continues to be in flux.
There's a pretty clear decline in the willingness to overpay for running backs, and the position has largely slipped down the collective NFL draft boards in recent seasons. Of course, that still doesn't prevent teams from taking potential stars such as Ezekiel Elliott fourth overall in the draft.
And then there's this pre-draft buzz around LSU running back Leonard Fournette.
Quick text from front office exec: "Fournette just made this draft a lot more interesting. More teams will want to make move into top 5."
— mike freeman (@mikefreemanNFL) March 3, 2017
It's probably safe to say that a top-five pick on Fournette would be a bet on a similar season to that of Elliott's breakout rookie campaign.
But was even Elliott worth the cost in 2016? Let's break it down mathematically by combining our Net Expected Points (NEP) metric and salary cap hits (via Spotrac) to see which running backs produced the most expected points per dollar this past season.
2016 Running Backs
There's a bit of a problem when it comes to comparing running backs to other positions in the NFL, and there's also a problem when comparing run-first backs to receiving backs, so we'll have to cover our bases here.
In 2016, the average quarterback drop back yielded 0.12 expected points, by our NEP metric. The average quarterback carry led to 0.26 points per rush. A reception by a receiver led to 1.10 Reception NEP per catch, and that mark was 0.93 for tight ends.
A running back catch, on average, led to a 0.45-point boost, but a running back carry? Whew. The average running back carry resulted in -0.02 Rushing NEP per play.
So we're going to account for Total NEP, which includes rushing and receiving production (as well as passing) when examining running backs. And we'll hone in on the 56 backs with at least 100 combined touches.
But remember how the average carry actually results in negative expected points? Yeah, so guys like Doug Martin wound up losing more than 24 expected points, all things considered (his Total NEP was -24.43 to be exact). In order to adjust, we're going to make Martin the baseline, meaning Martin added no points above expectation for our salary-focused purposes.
With Martin as our measuring stick, how did the rest of the backs stack up? Here's the bottom 28 in the group in terms of Total Net Expected Points per $10,000 in salary from 2016.
| Player | Team | Cap Hit | NEP+ per $10,000 |
|---|---|---|---|
| Doug Martin | TB | $7,764,706 | 0.00 |
| Todd Gurley | LA | $3,141,632 | 0.00 |
| Jonathan Stewart | CAR | $9,550,000 | 0.02 |
| Lamar Miller | HOU | $5,500,000 | 0.03 |
| Rashad Jennings | NYG | $2,812,500 | 0.03 |
| Chris Ivory | JAC | $3,468,750 | 0.03 |
| C.J. Anderson | DEN | $6,000,000 | 0.05 |
| Ryan Mathews | PHI | $4,000,000 | 0.08 |
| Giovani Bernard | CIN | $4,621,517 | 0.08 |
| Darren Sproles | PHI | $5,100,000 | 0.09 |
| DeMarco Murray | TEN | $6,000,000 | 0.09 |
| Frank Gore | IND | $4,000,000 | 0.09 |
| LeSean McCoy | BUF | $7,675,000 | 0.10 |
| Matt Forte | NYJ | $3,000,000 | 0.11 |
| DeAngelo Williams | PIT | $2,565,000 | 0.12 |
| Jerick McKinnon | MIN | $738,750 | 0.14 |
| T.J. Yeldon | JAC | $1,344,040 | 0.16 |
| Ezekiel Elliott | DAL | $4,537,516 | 0.18 |
| Melvin Gordon | SD | $2,424,948 | 0.20 |
| Matt Asiata | MIN | $840,000 | 0.21 |
| Theo Riddick | DET | $1,785,250 | 0.23 |
| Devontae Booker | DEN | $545,848 | 0.23 |
| Carlos Hyde | SF | $1,004,094 | 0.24 |
| LeGarrette Blount | NE | $1,025,000 | 0.25 |
| Thomas Rawls | SEA | $530,000 | 0.26 |
| Mark Ingram | NO | $2,510,000 | 0.26 |
| Dwayne Washington | DET | $468,941 | 0.26 |
| Alfred Blue | HOU | $629,483 | 0.28 |
For added context, 25 of the 56 backs had cap hits of at least $1 million. Of those 25, 22 are listed among the 28 least-cost-effective performers at the position, even when forgiving contracts that led to fewer than 100 touches this season. Put another way, even the expensive backs who were on the field plenty didn't really provide top-tier production.
The primary reason is that only seven running backs recorded at least 40 Total NEP this season (James White hit that mark on 99 touches, so he narrowly missed our cutoff). By contrast, 23 quarterbacks, 77 receivers, and 24 tight ends produced that many points.
The three highest cap hits at the position (Adrian Peterson at $12 million, Jonathan Stewart at $9.6 million, and Doug Martin at $7.8 million) resulted in -15.14, -5.77, and -24.43 Total NEP. That's about $29.3 million for -45.33 expected points.
Of course, LeSean McCoy ($7.7 million), DeMarco Murray ($6 million), and C.J. Anderson ($6 million) rounded out the top six and combined for 91.38 Total NEP.
Still, no running back secured more than 37.00 Rushing NEP, either. For that reason, overpaying for the position from a production standpoint is hard to justify.
And Elliott, despite a great season, produced 0.18 expected points per $10,000 in salary. In all, he produced 57.97 Total NEP, including the second-best Rushing NEP of 35.79. So, even with the 2nd-best rushing campaign of the year by a running back and a solid receiving total (22.18 Reception NEP, 13th-most), he wasn't worth the contract from an expected points standpoint.
That becomes more apparent when examining the most cost-effective backs from 2016 (and we'll flip the order to show the most cost-effective options on top).
| Player | Team | Cap Hit | NEP+ per $10,000 |
|---|---|---|---|
| David Johnson | ARI | $708,843 | 1.32 |
| Jordan Howard | CHI | $512,005 | 1.25 |
| Devonta Freeman | ATL | $721,106 | 1.03 |
| Jalen Richard | OAK | $450,000 | 1.00 |
| Mike Gillislee | BUF | $600,000 | 0.95 |
| Jay Ajayi | MIA | $580,203 | 0.84 |
| Tevin Coleman | ATL | $753,811 | 0.83 |
| Chris Thompson | WAS | $675,000 | 0.81 |
| Kenneth Dixon | BAL | $545,848 | 0.76 |
| Ty Montgomery | GB | $693,386 | 0.74 |
| Le'Veon Bell | PIT | $1,197,347 | 0.73 |
| Duke Johnson | CLE | $732,523 | 0.72 |
| Isaiah Crowell | CLE | $603,334 | 0.64 |
| Latavius Murray | OAK | $701,550 | 0.62 |
| Paul Perkins | NYG | $513,137 | 0.61 |
| DeAndre Washington | OAK | $517,298 | 0.60 |
| Zach Zenner | DET | $526,666 | 0.59 |
| Rob Kelley | WAS | $450,000 | 0.54 |
| Tim Hightower | NO | $840,000 | 0.50 |
| Terrance West | BAL | $600,000 | 0.49 |
| Jacquizz Rodgers | TB | $715,294 | 0.48 |
| Spencer Ware | KC | $958,333 | 0.41 |
| Matt Jones | WAS | $698,854 | 0.38 |
| Jeremy Hill | CIN | $1,023,710 | 0.38 |
| Derrick Henry | TEN | $983,259 | 0.37 |
| Charcandrick West | KC | $958,333 | 0.34 |
| Bilal Powell | NYJ | $1,733,333 | 0.33 |
| Shaun Draughn | SF | $860,000 | 0.31 |
David Johnson led the position with 69.12 Total NEP, but he actually posted a -0.53 Rushing NEP on the season. Let that sink in. Essentially, he produced only 0.02 points with each carry relative to the NFL's average performance by a running back (again, the average carry earned -0.02 Rushing NEP).
The Atlanta Falcons benefitted from having two of the best producing backs at bargain prices in the NFL, and the Chicago Bears got a steal of a season from Jordan Howard.
In all, just 28 backs produced at least 15.00 Total NEP this season, and 19 of them had cap hits below $1 million.
Investing heavily into running backs is going to be hard to defend in today's NFL, and sure, the threat of the run is important. But establishing the run is a recipe for losing in the NFL today.
The best approach, based on the numbers, is to reward pass-catching backs and to avoid overspending on run-first players.