The NFL has eliminated in-person offseason practices but released a full regular season schedule last week, allowing for the possibility that the league might shorten its season, if needed, from the normal 16-games slate. But whenever it starts there is the distinct possibility the season begins without fans in attendance due to risks associated with the novel coronavirus. That reality has prompted a question over how much home-field advantage means in football. Data, coaches and oddsmakers all tout the benefits of playing at home, forging an idea that many thought was unassailable. Yet the effects of home-field advantage carry far less impact in reality, no matter where and how fans are watching.
For example, NFL teams, collectively, sported a 132-123-1 (.518 win rate) record at home last year, the lowest home win rate since the league expanded to 32 teams in 2002 and a mark that significantly eclipsed the previous low set in 2006 (136-120-0, .531). You could argue that much of that decline was due to the flailing attempts by the NFC East and NFC South, two divisions that were below .500 at home last season. But there have been divisions below .500 at home in the past. In 2008 and 2009 two divisions were below .500 at home, yet teams won 57 percent of their home games as a league in each of those years. In 2006 there were three divisions that finished with a sub-.500 record at home but the overall win rate at home was .531.
On the surface this looks like a one-year blip but the truth is the better team drives the results far more than the location of the game. Since 2002 there have been 271 games played where the home team was assigned a point spread between minus-1 and plus-1, essentially telling us the game was between two evenly matched opponents in the view of the oddsmakers. As you would expect, the end results justified that belief, with the home team posting a record of 141-130 (.520 win rate) in those contests.
So why is the belief in home-field advantage so pervasive? And does the perceived advantage derive from the location of the game, or the fans filling the stands?
Michael Lombardi, a former NFL scout and front-office executive, explained being on the road makes hearing the snap count that starts every play difficult. “The inability to listen to the snap then allows the offense to lose its most significant advantage and gives the defense a chance to control the line of scrimmage, which is why so many teams struggle to run the ball on the road,” he wrote for The Athletic.
It’s true, teams have been less efficient rushing the ball on the road than they have at home, but the difference is slight. From 2002 to 2019 teams averaged 4.2 yards per carry at home compared to 4.1 on the road with similar first down, touchdown, fumble and third-down conversion rates as well. These results would also seem to rule out crowd noise as a significant disadvantage for the road team. In fact, there were more false start penalties called on the home team than there was on the road team last year (298 to 281).
|2002 to 2019||Yards per carry||First down rate||TD rate||Fumble rate||3rd down conversion rate|
|Home||4.2||23 percent||3 percent||1 percent||49 percent|
|Road||4.1||22 percent||3 percent||1 percent||48 percent|
Anecdotally, stadiums like CenturyLink Field (Seattle Seahawks), Mercedes-Benz Superdome (New Orleans Saints), Lucas Oil Stadium (Indianapolis Colts), U.S. Bank Stadium (Minnesota Vikings) and Arrowhead Stadium (Chiefs) are known for their ability to generate crowd noise, yet only CenturyLink Field (ranking 5th in the league for opponent false starts per game) and U.S. Bank Stadium (tied for 10th) saw their teams’ opponents register an above-average rate of false starts per game last year. Arrowhead Stadium, listed in the Guinness Book of World Records for registering a record 142.2 decibels on Sep. 29, 2014, ranked last in 2019 (0.1 false starts per game by Chiefs’ opponents). So, for the most part, even the stadiums with the most well-known reputations for being noisy aren’t providing much of an advantage in terms of crowd noise.
In addition, the size of the home crowd was not a big factor in the results of the aforementioned coin-flip games. Since 2002, three of the 271 games played where the home team was assigned a point spread between minus-1 and plus-1 also had between 25,000 and 29,999 fans in attendance. The home team went 2-1 in these games while those seating 90,000 or more went 2-4. Those are admittedly small sample sizes at the extremes, however, there is almost no correlation between the reported attendance and win rate of the home team in these coin-flip games over the last 18 seasons.
If attendance and crowd noise don’t reliably move the needle then we need to look at other factors such as distance traveled, time zone changes and, in the case of Empower Field at Mile High in Denver, altitude. And it’s here that we see the one place where there may be a distinct home field advantage.
Since 1970, the Broncos are 47-26-4 at home (.610) in games where the point spread is between minus-1 and plus-1. Since the point spread implies these opponents are evenly matched we can infer the advantage is due to playing 5,280 feet above sea level. It takes the human body weeks of exposure to adjust to such a high altitude, putting opponents at an immediate disadvantage. And while there are only nine games that fit this criteria (games in Denver with a spread between minus-1 and plus-1) since 2002, we’re still seeing the Broncos win at home more often than not. Denver has the eighth-highest win rate at home since the last expansion (.646), but it doesn’t appear to be due to its fans.
Aside from Denver, travel appears to have an impact in certain cases.
To a degree, the advances made in transportation has made traveling easier, limiting the disadvantages associated with distance traveled. According to research done by Rob Greer, the average distance traveled by an opponent for a divisional game (650 miles) is worth 2.3 points to the home team per his predictive model (which incorporates travel) while the average distance for other games (1,000 miles) is worth 2.5 points. However, it’s time zone changes that appear to provide a tangible home-field advantage due to the change in a person’s circadian rhythm or internal biological clock.
From 2003 to 2018, East Coast teams hosting a West Coast team went 133-80 in those contests, a 62 percent win rate. Football Outsiders found Central or Eastern teams hosting a team from the Pacific or Mountain time zones won 64 percent of their home games from 2001 to 2015. In our earlier sample of matchups featuring teams with a point spread between minus-1 and plus-1 those same class of teams went 28-19, a win rate of nearly 60 percent. In other words, teams that travel from west to east suffer enough jet lag as to impact their performance.
There will be 38 games in 2020 (with a projected win rate of 57 percent based on preseason point spreads) with a similar time zone advantage, perhaps fewer if West Coast teams playing back-to-back weeks on the East Coast decide to stay over rather than fly back and forth. For example, the San Francisco 49ers play the New York Jets and Giants, who share a stadium, in Weeks 2 and 3, leading General Manager John Lynch to lean toward less travel.
“Those 10 a.m. [Pacific time] starts, if you allow it to be, can be rough on your team. We don’t make any excuses, but we do try to prepare, and a great way to counteract some of those negative effects is to stay out there,” he told the NFL Network. “So, believe me, as soon [the schedule] came out, we started looking at spots. We’re looking at some different venues that will host us, and pretty sure we’re going to stay out east. That really does help our team because you eliminate travel, and you also eliminate that 10 a.m. start, that early start that can be rough, if you allow it to be."
Given the uniqueness of the situations, we won’t know for sure what effects, if any, empty stadiums will have on home-field advantage in the NFL this year, but we do know that unless the team is traveling from the West Coast, or playing in Denver, it’s unlikely to be much of a factor.
Read more from The Post: