As the NFL playoffs approach, teams will be competing hard down the stretch to secure home-field advantage for at least one postseason game. It’s also the reason teams like the Rams and Saints, who have comfortably locked up a playoff spot and even home field for their first playoff game, will risk injury to their starters as they chase home field for a potential conference championship game. This all makes good sense: Over 57 percent of games are won by home teams in the NFL, putting the road team at a distinct disadvantage.The benefits of playing at home in the NFL are clear. Statistician and FiveThirtyEight contributor Michael Lopez, along with Gregory J. Matthews and Benjamin S. Baumer, found that the effect of home field-advantage in American football is second only to the NBA among the major sports. But what is less clear is why.Michael Lombardi, NFL analyst and former front-office executive, believes that much of the disadvantage stems from players being unable to hear the snap count. According to Lombardi, losing the ability to hear the count takes away the offensive line’s inherent advantage of knowing when the play will start. This allows linemen and skill players to burst off the line and get a small early advantage on the defense. This loss of first-mover advantage on the offensive line — I’ll call it the Lombardi hypothesis — then manifests itself as road teams being less effective running the ball.The Lombardi hypothesis is intuitive. Most fans who have watched an NFL game in Denver or Seattle have seen the effects crowd noise can have on an offensive line’s ability to communicate. It’s not a huge leap to think that lower rushing efficiency might be the natural result of a loud crowd. But is it true? Are teams less effective running on the road? And if teams are less effective rushing on the road, how confident can we be that crowd noise is the cause?To find out, I took play-by-play data generated by Elias Sports Bureau from 2009 through Week 15 of 2018 and broke out all rushing plays by home and away team. I then plotted the distributions of yards gained per rushing play for both groups to see if there was a difference between rushing effectiveness for home and away teams.It turns out that road teams are indeed slightly less efficient rushing. Home teams average 4.37 yards per carry while road teams average 4.27, a tenth of a yard less. The distributions of yardage gained on rushing plays are extremely similar, however, and the disadvantage road teams face when rushing the ball is quite small. Assuming both teams ran 30 times in a game, we would expect the road team to rush for just 3 less yards than the home team.The relative benefits to the home team are magnified if we look at rushing expected points added, which account for game context like down, distance and field position. EPA per rush play is negative for both home and away teams, -.073 away vs. -.058 for home teams, but it is slightly less negative of a proposition for the home team. Over those same 30 rushing plays, we would expect the home team to lose 1.7 points, while the away team would be expected to lose 2.2 points, good for a half-point differential.So the first part of the Lombardi hypothesis appears to be correct: It is slightly harder to run on the road than at home. But is crowd noise the most likely cause? As a first approximation for a loud crowd, I broke out the rate at which both home and road teams were penalized for false starts1Using the same Elias data from 2009-2018.. We might expect more false start penalties to be called on the away team in a hostile, loud environment than the home team. Yet this not what we find. From 2009 through Week 15 of 2018, false start penalties were called on 1.4 percent of all home team plays and 1.34 percent of road team plays.This is a strange result if crowd noise is the driver of road teams’ lower rushing efficiency. To validate the finding, I drilled down into situations where teams were either backed up inside their own 10 yard line or in their opponent’s red zone. We’d expect the home crowd to be especially boisterous in those high-leverage situations, leading to more false starts for the road teams. But again that isn’t the case. Road teams were penalized for a false start on 1.44 percent of such plays, while home teams were penalized at a nearly identical — but still higher — rate of 1.47 percent. There’s other research that suggests crowd noise is not a factor in NFL team performance as well. Economist Tobias Moskowitz and Sports Illustrated writer L. Jon Wertheim were also unable to find evidence that crowd noise affects player performance. In the NFL specifically, they found that kickers and punters appear to be unaffected by crowd noise — undermining another common perception.But if crowd noise can’t explain home-field advantage, then what does? Unfortunately, the answer to this rather fundamental question remains elusive. Probably the best evidence for a true home-field advantage comes in the form of the sports teams in Denver. It may be the case that Denver’s high win percentage at home, especially early in the season, is dragging up the home league-wide home-field advantage. Each mile-high team in the major sports enjoys a home-field advantage far above those of their peers. The reason for it can be traced back to temperature and altitude. Playing at a high altitude without properly acclimating to it, especially in warmer temperatures, is a legitimate physiological disadvantage.I asked Lopez, who is now director of analytics for the NFL, his opinion on what drives home-field advantage, and he was circumspect. “It really is unclear,” he said. Lopez identified a few areas that might account for at least some of advantage, including referee bias in high-leverage situations.“If you look at the 15 most impactful, controversial calls in games over the past few years I think you’d find that maybe 14 of the calls went for the home team,” Lopez said. Next Gen Stats data might hold promise in this regard, however. “Using the ball-tracking data we have available, there are probably incremental ways we can help make official’s lives easier while increasing fairness in the game,” Lopez said.Lopez also noted that some recent studies have shown a drop in the size of the effect of home-field advantage. The drop could perhaps be explained by more comfortable travel, or better institutional controls on referee bias. How much these factors explain the 57 percent win rate home teams enjoy is difficult to say. Blaming crowd noise and its hypothesized effect on home-team rushing efficiency, however, appears to be unfounded.