Dallas Cowboys: Does Ezekiel Elliott need weekly 20+ carry benchmarks?

By all accounts, the Dallas Cowboys are back, and back with a vengeance. After a horrifically disappointing 2020 season in which everything that could possibly go wrong did, in fact, go wrong, 2021 is shaping up to be a bit of a redemption tour.

Dak Prescott is back. The once impressive offensive line is back. The old defensive system is back (albeit with some happy changes). And even the old Zeke appears to be back. After dedicating himself this offseason, Ezekiel Elliott is arguably in the best shape of his career.

While “best shape of my career” is a laughably cliché thing to say this time of year, in this case it actually appears to be true. Zeke is noticeably slimmed down and defined. The svelte five-year veteran has been releasing clips, touting his new-found quickness and agility. He looks ready to bounce-back from an undeniably poor season in which his back-up, Tony Pollard, outperformed him in most qualitative categories.

To win more, must the Dallas Cowboys pursue benchmarks like 20+ carries/game for Ezekiel Elliott?

With Zeke back, to what appears to be his rookie self, the question now is how this potent offense should craft their attack. Should they air it out and lean on their All-Star WR corps or should they go back to basics and pound the rock the ol’ fashioned way?

Anyone who’s been around the game has undoubtedly seen those gaudy running game numbers that “prove” the merits of a run-heavy approach. Something like – teams that run the ball 20+ times per game are 70% more likely to win than teams who run less than 20 times per game. Sound familiar?

But this is a case of numbers being misrepresented to draw inaccurate conclusions. Because if it was as simple as that, wouldn’t teams just run the ball exclusively in order to reach that magical benchmark? Wouldn’t we see running backs increasing in value while quarterbacks decrease? Instead we’re seeing the polar opposite. Instead we see teams getting more pass-happy by the day and run-heavy teams are the exception to the rule rather than the standard in which to strive.

So what gives?

You've heard arguments that go something like this: "When Ezekiel Elliott gets 25+ carries, his team is 19-4."

This is explained by the incentives teams face late in games. Teams run more when they're already likely to win.

Light blue = winning teams
Dark blue = losing teams pic.twitter.com/IduMoCeBoe
— Cowboys Stats & Graphics (@CowboysStats) March 22, 2021

As shown in the graphic above, teams that are leading on the scoreboard are more likely to run the ball. Running the ball DOES NOT lead a team to lead on the scoreboard. It’s an important distinction to make because failure to differentiate between correlation and causation can lead one down a dangerously misleading path.

The whole “teams that run the ball x number of times per game are more likely to win” is a myth that was busted long ago, yet we still see respected sports figures regularly trip over the shoelaces of correlation and lead fans astray.

Running the ball does not make you more likely to win. When teams are already winning, they tend to run the ball more. And teams that are losing tend to throw the ball more. That’s all.

teams are already winning tend to run the ball more. And teams that are losing tend to throw the ball more. That’s all.

What leads to winning isn’t running or passing a set number of times, but rather doing what works best for your specific offense (whatever is producing the highest EPA).

Having a balanced approach is even overrated and often insignificant. If your passing game is twice as effective as your running game, it doesn’t make much sense to employ a balanced attack on offense.

Especially since the entire “establish the run” thing has been largely debunked and teams don’t even need a balanced running game in order to make play-action work. Play-action just inherently works and all a team needs is the threat of a running game to cash in.

So to say the Dallas Cowboys must fully commit to giving Zeke (and/or Pollard for that matter) a set number of runs per game this season, is missing the whole point. Winning on the scoreboard leads to high rushing totals, not the other way around.

This tells us, yes indeed, we want to see Zeke get 20+ carries per game in 2021. Not because giving him the rock leads to winning, but rather because if the Dallas Cowboys are already winning, so they’ll probably be giving Zeke the ball. It’s not arguing over semantics, it’s a critical distinction.

Time of possession?

In much the same way, time of possession (TOP) numbers have become chronically misleading. Just like the bogus 20+ carries/game threshold, we’ve seen people attempt to sell us on the merits of winning the TOP battle and how winning football teams often have high TOP numbers.

For the nerds: Time of possession explains point diff in a game with an R^2 of 0.22 (not even much for in-sample) and when you account for things to strive for (being more efficient on a per-play basis and moving the sticks more often), time of possession becomes insignificant. pic.twitter.com/xcSwgBSlCE
— Timo Riske (@PFF_Moo) November 22, 2019

Again, leading in the game is what often leads to high TOP numbers. Running a plodding ball-control offense doesn’t equal winning. Just like before – doing what you do best is what matters. A team like the Ravens can run the ball 80% of the time because they are absolutely brilliant at it. That’s not the Dallas Cowboys.