clock menu more-arrow no yes

Filed under:

Removing Fans Didn’t Remove MLB’s Home-Field Advantage

Home teams won at a higher percentage in 2020 than they had in years past. That could make a big difference in the condensed wild-card round. But what about the following rounds and World Series, which will be played at neutral sites?

Getty Images/Ringer illustration

In June, when Major League Baseball and the MLB Players Association grudgingly agreed to play ball, I listed several known unknowns about the 2020 season. Some of the mysteries I mentioned—such as how this season would affect player development; Hall of Fame inductions; and future changes to the sport’s season length, rules, playoff format, and more—remain unsolved. But we can close the book on one of my questions, which is relevant in October too: “Will home-field advantage still be in effect without fans?”

For years, analysts have argued about the origins of home-field advantage in sports. Baseball’s home-field advantage is slightly smaller than those of other major American sports, possibly because luck plays a larger role in determining the winner of any given game in baseball than in basketball and football (for instance). Proposed contributing causes have included the psychological effect on players of encouragement or opposition from fans; the influence of fans on umpires’ ball and strike calls; home players’ greater familiarity with their own parks; teams’ tendencies to tailor their rosters to the layouts of their home parks; the impact of travel and living conditions; and the tactical advantage of batting second. Past studies on the subject have collectively suggested that the answer is some mixture of all of the above, without precisely or conclusively assigning credit to each home-field factor.

What we can say with certainty is that home-field advantage has been incredibly consistent from deadball to moneyball and beyond, despite dramatic alterations to the sport. Home-team winning percentages have hovered in the .530s and .540s in every decade since the start of the 20th century. The chart below lists cumulative home-team winning percentage by decade going back to the beginning of the live ball era in 1920, in addition to each decade’s home-field advantage in OPS and home-field advantage in run differential per game.

This season, fans were banned from ballparks, which gave home-field advantage scholars the natural experiment they were waiting for. At first, teams’ home-field edge seemed to have dissipated or disappeared with flesh-and-blood fans replaced by cardboard cutouts, which seemed consistent with previous research on basketball and soccer games played in empty stadiums or between teams with the same home facility. On August 17, MLB.com’s Mike Petriello noted that home teams had won only 50.5 percent of their games, which was lower than any full-season figure.

However, FanGraphs’ Ben Clemens observed days later that while home teams weren’t yet winning at their typical clip, they were still outproducing visiting teams at the plate, which seemed to augur good outcomes ahead. Sure enough, home teams finished the season strong. And now we know the answer to that common question from June: Not only was home-field advantage still in effect, but it was bigger than usual. Historically, MLB home teams have won 54 percent of their games. (Home teams have gone 95,205-80,954 in the regular season since 1920, a .540 winning percentage.) This year, home teams went 500-398, a .557 winning percentage. That’s the highest home winning percentage in a season since 2010. Multiple teams boasted historic home/road splits in the small-sample campaign: The Astros and Twins recorded the two biggest gaps ever between home winning percentage and road winning percentage, and the Yankees showed the greatest gulf between average home and road run differentials.

That surface success was backed up by underlying production. Home teams outscored their opponents by about .24 runs per game, which was also the highest figure since 2010.

As a final indicator of dominance, home teams outhit their opponents by 44 points of OPS, the highest mark since 1978.

The absence of fans in the stands wasn’t the only development working against home teams in 2020. Thanks to the proliferation of seven-inning doubleheaders and the institution of the extra-innings automatic-runner rule—which lowered home teams’ winning percentage in extra-inning games after it was put in place in the minors—the average number of innings per game fell to 8.61, an all-time low. The shorter the game, the less time for the factors behind home-field advantage to take effect. The 60-game, extra-unbalanced schedule dramatically reduced distance traveled, which in theory would also have a leveling effect: Travel costs players sleep and inhibits recovery. Beyond that, the Blue Jays played most of their home games in a hastily upgraded Triple-A park. And 29 times, a visiting team technically played a home game by batting second in an opponent’s park. (“Home” teams went 11-18 in those games, which are included in that 500-398 overall record.)

Yet home-field advantage stayed strong. In light of the sample size—only 898 artificially shortened games—we have to temper our conclusions. According to Phil Birnbaum, editor of SABR’s By the Numbers newsletter and author of the Sabermetric Research blog, a .557 home winning percentage is only a little more than one standard deviation higher than expected. We can’t say with great confidence that the true home-field advantage was higher than usual this season, but we can say that it’s unlikely that it was significantly lower than usual, which is surprising considering the circumstances.

As I wrote in July, some studies have hinted that as much as one-third of home-field advantage may stem from umpires’ tendency to make more favorable ball and strike calls on home hitters, which some analysts have attributed to pressure applied by a vocal crowd. Unless umps are susceptible to fake crowd noise too, that shouldn’t have been an issue in 2020. Yet the home-field advantage on ball and strike calls persisted, at least to some extent. The graph below shows the percentage of pitches outside the strike zone that were called strikes on batters at home and away in each season of the Statcast era (2015 to present). Although the umps have gotten better across the board at avoiding bogus strike calls, they’re still assessing erroneous strikes more often to hitters who are on the road.

However, for the first time on record, umpires this season called balls on pitches taken inside the strike zone more often on hitters who were on the road, a reversal of the typical trend.

That supports what Jonathan Judge of Baseball Prospectus found last week when he compared this year’s strike zone to those of previous seasons: Home-field advantage on umpire calls was still apparent, but more muted than usual. According to data provided by Judge, taken pitches were a little more than 2 percent more likely to be called strikes on visiting batters than on home batters from 2015 to 2019, after accounting for location, count, pitch type, catcher receiving skills, and umpire tendencies, among other factors. This year, though, the umpire penalty on visiting hitters dropped to a little more than 1 percent.

It’s not clear why umps continued to give visiting hitters a tougher time. Maybe the fake crowd noise reinforced some subconscious conditioning to call fewer strikes on home hitters, or maybe home pitchers are more likely to hit their spots, which could lead to additional strikes (although Judge’s model should theoretically control for that already). Regardless of the reason, though, the reduced strike-calling imbalance should have weakened home-field advantage. But it didn’t, which seems to bolster the belief that home-field advantage could be primarily a product of familiarity with the ballpark, not noise, moral support, or occasional direct intervention from fans.

Historically, home teams have outperformed road teams most starkly in the early innings (especially the first), which suggests that visiting players may need time to adjust to their surroundings. Along the same lines, BP’s Russell Carleton calculated that home-field advantage exists even when the visitors are based in the same city as the home team, which suggests that there’s more to the effect than a lack of travel, home cooking, and a comfortable bed (although whatever comfort comes from being at home could have taken on a heightened importance amid an anxiety-inducing pandemic, when visiting players were largely confined to their hotels). He also found that players with extensive prior experience in a certain park don’t suffer a decline when they return to that park as a visiting player in the season after switching teams, and that home teams don’t have an advantage in the first month after opening a new park. That jibes with Matt Swartz’s discovery that home-field advantage is slightly more pronounced in interleague games than in intraleague games, and also more pronounced in interdivision games within the same league than in intradivision games.

There’s also something to be said for the value of last licks. On Tuesday, FanGraphs’ Dan Szymborski tried to estimate the portion of home-field advantage that’s attributable to batting second by backing out the impact of the disparity in players’ home and road performance. He arrived at an approximation of 12 points of winning percentage, or a little less than a third of the standard home-field advantage, but he acknowledges that the figure is little better than “a wild-ass guess,” because the upside of batting second may already be reflected in players’ stats.

The upshot of all of this, as far as the present postseason is concerned, is that during this week’s wild-card round, in which games take place at the parks of the higher-seeded teams, home-field advantage likely matters as much as ever. In the past, home-field advantage has been just as strong in October as during the regular season: Entering Tuesday, home teams had won 879 of 1,617 all-time postseason games (a .544 winning percentage), with a .21 run differential per game and a 30-point OPS advantage. (Granted, in recent years better teams have been more likely to have home field.)

A second October takeaway is that MLB’s COVID-prompted plan to play subsequent rounds of the playoffs (which may feature fans) in neutral parks will hurt the “home” teams, even though it appeared early in the season that home-field advantage might not matter this year. There’s not enough precedent to say for sure what being displaced from their digs will signify for the favorites, but in 68 MLB games dating back to the ’90s that were played in neutral parks, the “home” teams went 31-37. For what it’s worth, “home” teams have also posted losing records in both the NBA and NHL postseason bubbles. “It’s probably fair to say that home advantage requires a team to actually be at home,” Birnbaum says. All of which may mean that the expanded playoff field isn’t the only problem facing baseball’s best teams; the literal playoff fields, and the fact that they won’t be familiar, are another nontrivial ingredient in this season’s recipe for upsets.

Thanks to Lucas Apostoleris of Baseball Prospectus, Kenny Jackelen of Baseball-Reference, and Zach Kram of The Ringer for research assistance.