Over the past three weeks, the Washington Football Team has revived its season with an approach seemingly in contrast with the modern, pass-happy NFL. Washington has run the ball early and often, maintaining possession and sustaining long drives just often enough to secure one-score wins.

The recent approach is counter to the overwhelming evidence that passing is far more efficient than rushing. The worst passing offense (Jacksonville, 6.01 yards per pass) produces more than the best running offense (Indianapolis, 5.18 per rush).

Washington offensive coordinator Scott Turner, 39, rose to his position in 2020 as NFL analytics reached the mainstream, and while he understands and incorporates the numbers, they apparently don’t dictate his decisions. Still, Turner said his play-calling during his team’s winning streak — 59 percent run on early downs when the game is within one score, up from his career rate of 46 percent — does not represent a philosophical shift. Rather, it reflects him taking advantage of specific situations, even though he does harbor a soft spot for the approach.

“When the run is working, it makes you really feel like you can call any play on the sheet,” he said. “When you’re having some issues running the ball, that’s when [play-calling] gets a little bit tougher. [The run] just lets you play a little bit more free.”

Still, the numbers prompt a question: What is the place of the running game in the modern NFL?

Since the late aughts, the analytics movement, rule changes aimed at protecting quarterbacks’ health and other factors have fueled increased throwing. From 2010 to 2020, the league’s passing rate jumped from 47.3 percent on early downs in one-score games to 50.3 as some began to question why teams with elite passers weren’t throwing even more. In early 2019, a Seattle fan frustrated by his team’s underutilization of quarterback Russell Wilson coined the phrase “Let Russ Cook.”

Like many other debates in the analytical age, run-pass decisions are really about data vs. gut instinct. Analysts and analytically savvy players and coaches generally agree on a few major points — passing carries more risk, running is more efficient in select situations and is necessary to give quarterbacks breaks — but disagreements remain on what other factors, if any, figure into the equation.

When Washington plays at Las Vegas on Sunday, the numbers might dictate continued reliance on the run. The Raiders have one of the NFL’s most explosive offenses, which increases the odds Washington must throw to keep pace. But the Raiders also have a porous run defense. Meanwhile, Washington started its winning streak by upsetting Tampa Bay, a team with the best run defense in the league, and still Washington rushed 34 times that day. When asked whether Washington was now a physical, ground-and-pound offense, Turner demurred.

“Every game plays itself out differently,” he said, adding, “If you want to be a great offense . . . you got to be able to do it all.”

Yet around the NFL, it’s clear defenses are willing to make some concessions to opponents’ running games in hopes of stifling their explosive air attacks. More and more, defenses are moving players out of the tackle box and instead employing two-high safety shells to limit the deep ball, betting offenses will score less trying to sustain long drives than hitting chunk plays. But earlier this season, Los Angeles Chargers Coach Brandon Staley, an analytical, defensive-minded coach, went viral for his justification of why the run remained important.

“If you’re just a passing team, there’s a physical element to the game that the defense doesn’t have to respect — and that’s the truth,” Staley told reporters, adding: “Running forces the defense to play blocks and to tackle. . . . [Running] really challenges your physicality.”

Pro Football Focus analyst Eric Eager rejects the idea that physical run-blocking is worth sacrificing offensive efficiency. Eager, a former Division II tight end, remembers how Minnesota State Moorhead manhandled opponents in both wins and losses, and he’s skeptical the physical play outweighed skill and decision-making in determining the outcome.

Over the years, as he got his doctorate in applied mathematics and became an NFL data scientist, he recast some childhood memories of Vikings Coach Brad Childress. Eager empathized with coaches in charge of overmatched teams — they needed to find other ways to encourage players — but saw Childress urging his team to be more physical than the opponent as silly, an example of prioritizing something not tied to winning instead of trying to score more points.

“The only scoreboard is the scoreboard,” Eager said. “I don’t want to hang banners and say, ‘Oh, we were the most physical team.’ ”

Not long ago, Eager heard echoes of Staley’s argument during a meeting with a team for which he now consults. Eager went looking for other reasons coaches love running despite its overall inefficiency.

In a recent article, Eager proposed a rather human rationale. He concluded coaches overlook the rarity of “perfectly blocked” runs, the most efficient play in football, because the payoff is potentially huge. He compared the running game to a gambling parlay — one bad block dramatically reduces success in the same way one wrong wager busts the bet — and argued a play-caller should mostly rely on passes, which are steadier, and sprinkle in a few runs. The article reignited the run-pass debate on Twitter.

Mitchell Schwartz, who called himself “as analytic-minded as they come” among former NFL offensive linemen, disagreed with Eager. He echoed Staley, arguing football is based on physical, “animalistic” dominance, and added running well is the best way to breed confidence among players, which is key to winning. He argued numerous studies have shown the impact of a positive mind-set: that internal confidence can influence external results. He noted behavioral science is better established than football analytics.

“We know teams can have mental letdown games,” Schwartz said, pointing to Jacksonville’s upset of Buffalo in Week 8 and Denver’s shellacking of Dallas in Week 9.

While Schwartz and Eager are not so different — both in their 30s with playing experience — their fundamental disagreement is over the function of the running game. Eager sees it as a tendency-breaker, a way to maximize the most efficient tool an offense has, while Schwartz believes it establishes a tough, disciplined mind-set that helps a team win. He remembered how, after some particularly physical series in which the offense imposed its will, the defense ran out a little more fired up.

“I’d love to be able to show” physicality impacts performance, Eager said. But he pointed out the data community’s best efforts have yet to find any evidence that an offense establishing the run early makes it more effective later. Eager argued offenses can replicate the demoralizing effect of runs by throwing the ball more and scoring more points, but Schwartz dismisses the idea. He has been on too many sidelines to ignore the ineffable qualities of an offense imposing its will.

“We’re at a little bit of an impasse, at least on this specific topic,” Schwartz said. “One side is saying: ‘No, we know what it feels like. We understand that it has an impact.’ And the other is saying, ‘If it does, it should show up in the data, and we’re not seeing it, so that means it doesn’t have a quantifiable impact.’ ”

For now, Eager is certain that every offense in the NFL is underutilizing the pass and that there is an optimal run-pass ratio for each team. Schwartz, of course, is skeptical de-emphasizing the run further would be effective, and he wonders whether new tracking data will one day illustrate what he has always felt. If an offense runs early, he said, perhaps a defensive lineman’s average speed per play declines by a half-mile per hour later in the game, making certain types of runs more effective.

Both are open to being proved wrong by new data.

“Maybe something comes along that is able to actually cut through the noise,” Schwartz said.

Until then, play-callers around the league must live in a liminal space, with enough information to draw strong conclusions but not enough to test everything. On Sunday in Las Vegas, in the coaches’ box high above Allegiant Stadium, Turner will do what he always does, which is juggle what he studied all week with what he’s seeing in the moment, the numbers in his head with the feeling in his gut.

The challenge is to process all that information and find the right answer. Turner would like to have more right answers — even if they aren’t the most efficient ones.

“You can’t let the defense make you one-sided,” he said. “Some games, it might be run heavy. Some games, it might be pass heavy. But the best offenses in the league — they can be successful doing both.”