Do Running Back-By-Committees Work in the NFL?

DeMarco Murray had 392 carries in 2014, but more teams are using the committee approach at running back. Is it smart?

Acronyms have become a huge component of modern life, as we constantly are looking for ways to become more efficient and succinct with our forms of communication.

Can’t stay any longer to look at the string of cat videos your friend sends you on social media? A little “GTG BRB” (got to go, be right back) will get you home free. Indignant that your friend didn’t invite you to the concert, but then posted pictures of it? A sarcastic “KTHX” (okay, thanks) will show them. These even work in real life, when your boss asks you when you can get that project in that’s past deadline. “ASAP” (as soon as possible) is just vague enough to serve your needs. These are helpful acronyms.

But RBBC (running back-by-committee) is nearly universally reviled by fantasy football players and NFL running backs alike.

RBBC operates based on the theory that multiple running backs divvying up the backfield responsibilities will allow the running game to be more efficient and the players to stay healthier -- something that makes sense for NFL players’ health and safety. I’m curious, though -- with how prevalent the strategy has become, does it actually work?

How successful has the RBBC in the NFL been?

The Death of Workhorse Running Backs

First of all, how prevalent is the use of running back-by-committee in the NFL? We hear all the time that the “workhorse” running back is dying in the professional game of football, and it’s true. In the year 2000, there were nine running backs with at least 300 carries and 11 with at least 350 opportunities (carries plus targets). In 2015, however, there was one running back with 300 carries and just two with 350 opportunities.

The graph below shows the steady downward trend of these numbers since the new millennium, the blue line indicating running backs with 300 or more carries and the red line indicating those with 350 or more opportunities.

It’s interesting to note that there are significant drop-offs between certain years, plateaus in the usage of high volume running backs. The data for workhorse runners can be divided into eras from 2000 to 2006 (average 10.0 with 300 or more carries and 10.6 with 350 or more opportunities), 2007 to 2010 (6.3 and 5.8, respectively), and 2011 to present (2.4 and 3.2, respectively).

This is almost too simplistic, though. We know the league has gone towards a high-volume passing game at the expense of running back opportunities, but the RBBC has done a ton of work at making the “workhorse” or “bell cow” running back a remnant of NFL antiquity.

The chart below shows the Opportunity Market Share (or percentage of total running back opportunities) for lead running backs each year since 2000. By adding up the opportunities that the lead running backs on each team had in each year and dividing it by the total opportunities afforded to NFL running backs that year, we arrive at this metric.

So, not only are lead backs losing pure volume to the passing game, they’re losing tons of touches to other backs. This is a brutal development for the lead back role in the NFL.

But have these factors actually helped or hurt the game on the field?

Opportunity and Efficiency

The easiest way to check if a trend is or isn’t working is to run a correlation between that trend and a success factor. R correlation is measured on a scale of 0.00 to 1.00 (either positive or negative). The closer the value is to 1.00, the more direct of a relationship there is between two variables; the closer it is to 0.00, the more random the association is.

For us, the trend we want to examine is the usage of RBBC. The success factor we’ll use to evaluate it is numberFire’s signature analytic, Net Expected Points (NEP).

NEP is an analytic helps us take the numbers we get from the box score and shows how that player or team did versus expectation. By adding down-and-distance value to standard box score information, we can see just how much each play and each team as a whole influence the outcome of games. If a running back rushes for five-yards on 3rd-and-2, it means more to the game than it does on 3rd-and-10, and those plays should be valued accordingly. For more info on NEP, check out our glossary.

We have a few different ways to measure how the running back-by-committee has been used and whether or not it’s been successful.

The table below looks specifically at the Opportunity Market Share of each running back in the last three years, and compares it to their Total NEP (Rushing NEP plus Reception NEP) and Total NEP per opportunity. Is there a correlation?

RBBC Total NEP Total NEP/Opp
RB Opp.% 0.31 0.01

There’s a moderately significant correlation between the amount of a team’s work a running back gets and the amount of Total NEP they accrue, but this is sort of obvious; the more chances a player gets to touch the ball, the more total value they will earn. In terms of the per-opportunity Total NEP -- more of a quality metric than raw Total NEP -- there is no correlation at all.

In other words, a running back getting a ton of work in his backfield isn't any more likely to be more effective than one who's seeing little work.

What if we look solely at the primary running back’s Opportunity Market Share, though? The table below compares the lead running back’s Opportunity Market Share to three success factors: team Adjusted NEP, team Adjusted Rushing NEP, and team running back Total NEP (the sum of all Total NEP accrued by running backs on that team).

RBBC Adj. NEP Adj. Rush NEP RB Total NEP
RB1 Opp.% 0.12 0.29 0.23

There is a compelling correlation here for a team’s full Adjusted Rushing NEP production, but -- again -- this evaporates when comparing the Opportunity Market Share to the per-play version of the metric. Even when we look at the Total NEP accrued by all running backs, there’s nothing strong enough to say that one back getting more of the work leads to better production for all of the backs (or vice-versa).

Finally, I simply counted how many running backs each team used over the last three years and compared that to the team running back Total NEP over that same period -- how many running backs were used versus how effective those running backs were. The correlation: a measly -0.07.

Does It Work?

All of this is to say that there is no conclusive data that proves that RBBC is either a good strategy or a bad one.

For some teams, especially those with limited players in either workload or skill-set, this is probably a great way to game plan each running back into their best possible situation. For other teams, it may be a horrible idea, because they have a talented all-around player that doesn’t need spelling. Fantasy football players will continue (rightfully so) to see the RBBC as an unnecessary burden on our valuation of this volatile position, but for the NFL, it’s become a sort of shorthand for how to plug in their players into good matchups for their tools.

Whatever the effectiveness or efficiency of the RBBC, the trends of the league over the past 16 years (and longer) show it is here and it’s here to stay. That much we know for certain.