Is Home-Field Advantage Becoming Endangered?
Home-field advantage isn’t always considered a matter of great importance in baseball. Crowds aren’t as close to the action as they are in basketball. There’s nothing comparable in the sport to something like raucous Cameron Indoor Stadium. There are no 100,000-seat, canyon-like stadiums cascading noise to the playing surface like in college football.
But home-field advantage is a real thing in baseball, and significant, and has remained constant for better than a century.
The road winning percentage of visiting teams was .461 in the 1910s. Road winning percentage stands at .464 to date in the 2010s. Road winning percentage has remained consistent over the decades.
Since the 1970s…
| Decade | Road winning % |
|---|---|
| 2010s | 0.464 |
| 2000s | 0.456 |
| 1990s | 0.464 |
| 1980s | 0.460 |
| 1970s | 0.463 |
Conventional wisdom has it that home-field advantage is derived from some combination of hostile crowds, fatigue from travel, and the familiarity of the playing surface. And, in a way that’s unique to baseball, teams can tailor their roster to their home park’s dimensions. Having carried on that way for better than a century, the home-field edge seems to be something of a scientific law.
Home-field advantage is important. It’s a carrot to chase in the regular season, and is often a crucial advantage in the postseason. Sports on Earth found that there have been 154 best-of-seven series in MLB history following “the current 2-3-2 home-field pattern.” The team with home-field advantage has won 85 of those series, good for a .552 winning percentage.
But what if some other force is driving home-field advantage and it’s in danger? What if some savvy, innovative teams can close the margin that exists between home and road teams? What if playing fields tilt closer to neutral?
***
University of Chicago behavioral economist Tobias Moskowitz and Sports Illustrated writer L. Jon Wertheim argued convincingly in their book Scorecasting – published in 2011 – that home-field advantage is largely tied to umpire bias, and referee bias in other sports.
In what was innovative research at the time, the authors analyzed several years of PITCHf/x and QuesTec pitch-location data to show that home teams typically received the benefit of borderline pitch calls, particularly in higher-leverage situations.
Wrote Moskowitz and Wertheim:
“In baseball it turns out that the most significant difference between home and away teams is that the home teams strike out less and walk more — a lot more — per plate appearance than road teams.”
And in 2016, pitchers continued to strike out more, and walk fewer, when at home.
| Road/Home | K/9 | BB/9 | FIP |
|---|---|---|---|
| NL Road | 8.1 | 3.4 | 4.27 |
| AL Road | 7.9 | 3.2 | 4.35 |
| NL Home | 8.3 | 3.1 | 4.00 |
| AL Home | 8.1 | 2.9 | 4.15 |
The authors calculated the advantage equated to 7.3 runs per season for each home team, while noting home teams on average outscore visitors by 10.5 runs per season. If accurate, that bias accounts for roughly 70 percent of home-field advantage.
Anecdotally, more borderline calls seem to go against visiting hitters than home batters.
To see if this bias continues to be a significant factor and to see if it has weakened or strengthened since the Scorecasting findings, I asked FanGraphs’ own Jonah Pemstein to research the home/road splits of all balls and called strikes from 2008 to -16.
Here is a look at all called strikes for home teams 2008-15:
All called strikes for road teams 2008-15:
Compare that to called strikes for home teams 2016:
And road teams in 2016:
The force still appears to be quite strong, and it could tied to a conscious or subconscious desire to comform to the crowd. The Scorecasting authors noted several psychological studies that concluded humans conform because they want to fit into the group and because they believe the group is better informed.
Whatever the reason, the bias appears to exist after nine seasons of PITCHf/x tracking every pitch in every major-league park. The system was designed in large part to improve umpire accuracy.
After each game called behind the plate, umpires are given a report with their calls and the corresponding pitch-location data, along with accompanying video. While the strike zone is evolving — upon which subject The Hardball Times has published some excellent research — and while there’s some evidence of improvement among umpires in the PITCHf/x era, the bias in favor of the home team appears to remain remarkably consistent despite year-to-year changes in strike-zone size and umpire performance.
The following graphs are the most powerful data we identified.
Consider the rates of out-of-zone strikes for home versus road teams from 2008 to -16:
And then the rates of in-zone balls from 2008 to -16:
So that leads us to this interesting possibility: Major League Baseball could perhaps eliminate much of home-field advantage by simply handing over ball/strike duties to robot umps. (And it seems there is a greater call for that with each passing postseason.)
If 70% of home-field advantage is tied to umpire bias, then much of home-field advantage could be eliminated fairly easily.
MLB commissioner Rob Manfred told USA Today in October that pitch-tracking technology is impressive but he does not believe the game is ready for robot umps… yet.
From that piece:
“The technology of calling balls and strikes without a human being involved has continued to improve,” Manfred said. “The principal reason we’ve always done it after the fact is that unlike the box you see on a broadcast, [for] our system that we use to grade our umpires, someone goes in and manually adjusts the strike zone for the batter. As technology continues to improve and those sorts of adjustments can be made [in] real time, that technology will become more feasible for use on the field. I don’t believe we are there yet.”
If the technology isn’t there yet, it’s close. Perhaps all that’s required is for a personalized height of zone to be established for all players from Jose Altuve to Giancarlo Stanton, for each player to be attached to sensors each spring and have their strike zone measured.
Yet, even if you believe bias explains much of home-field advantage, it doesn’t explain the totality of it. And perhaps advances are being made in other areas relevant to the phenomenon.
***
Will Ahmed might be on to something.
As a squash player at Harvard he was around athletes, he said, “who didn’t really know what they were doing to their bodies.”
Ahmed felt he and his teammates “misinterpreted fitness peaks.” He felt there wasn’t enough understanding of the importance of recovery and rest.
That was the spark for his startup company, Whoop, which is part of the wearable technology wave that has made its way to professional sports and Major League Baseball.
Ahmed says that, last year, he tested his product with 230 minor-league players in a partnership with MLB. Whoop is worn around the wrist like your FitBit but it’s different. Ahmed says it does not measure “steps” but rather “strain.” The device monitors heart rate, heart-rate variability, ambient temperature and perspiration.
While the technology is designed to better enhance an athlete’s efficiency, it was studying the rest and recovery of minor-league players on the road led Ahmed to a “eureka” moment.
“We found that athletes, when traveling, on average got an hour less sleep than athletes that did not travel,” Ahmed said. “An hour less in bed, 45 minutes less sleep. That’s about 10% lower recovery. Conventional wisdom says the home team wins because they have their fans supporting them or they are playing in a familiar park. The study suggests the home team may win because they are better recovered and better well rested.”
Ahmed found the minor leaguers he was studying were slightly less effective on the the road after traveling, and in returning home, because they were less likely to be at full recovery.
“There was a direct correlation between the higher your recovery and the faster your exit [batted ball] velocity. We found the same correlation with pitchers and fastballs,” Ahmed said. “I didn’t know for sure we were going to see such strong correlations with travel, recovery and performance. Those were a bit of a eureka [moment] with us and for people in Major League Baseball who saw this study.”
Ahmed said that among the 28 teams that experimented with Whoop, 70% of players participated. Whoop gives a metric before 0% and 100% in regard to recovery. He said habits changed.
“The athletes themselves are self-managing,” Ahmed said. “One thing we found is athletes would talk about drinking less alcohol or going to bed earlier. They are starting to think about their recovery. The higher their recovery the better they play. You go from thinking of yourself as an athlete three hours a day, to thinking of yourself as an athlete 24/7.”
In Scorecasting, the authors discounted the travel fatigue as a reason behind home-field advantage.
“[Home—field advantage] was virtually the same win MLB from 1903 to 1909 as it was from 2003 to 2009. This suggests that teams jetting on chartered flights with catered meals, high-thread count linens and flat-screen televisions have no more success than did teams that traveled to games in Pullmans and buses. (Either travel isn’t causing the home advantage or teams need to rethink their jet purchases and on-flight catering).”
To their point, no team travels more miles than the Seattle Mariners in a given season given their northwest location. While the Mariners have a lowly .435 road winning percentage since their inception – below the MLB rate of 46.2% since 1970 – the Mariners’ home-road winning percentage differential of 6.8 points is below that of the MLB average 7.7 since 1970.
But as more and more teams and players become more conscious of rest and recovery, as players learn how to better travel, perhaps the old data becomes less relevant.
Teams and players know more about performance and rest today than they ever have. Many NBA, NFL, NHL and MLB teams are trying to better understand how to improve player performance through rest and recovery.
Some major-league teams installed recovery rooms in their clubhouses, players have met with sleep experts and have been advised to travel with eye masks and their own pillows. In the NBA, home teams are perhaps losing some of their home-court edge. NBA teams are more and more curious about how best to manage travel and deal with a demanding schedule. In baseball, perhaps these changes and adoptions will also erode some of the home team’s century-plus edge.
***
There is another question to be answered: do we want to rid the game of home-field advantage?
While baseball’s home-field isn’t as powerful as those in basketball and football, it is significant. It creates regular-season incentive and reward. So perhaps those pining for robot umps should pause and think about all of those implications. Do we want all fields to play neutral or closer to neutral?
Perhaps no matter how many advances and adoptions in technology there are, the home team will always have a significant edge. It’s been something of a constant for more than 100 years. Or perhaps the current gap between home and road teams is more fragile than we think.
A Cleveland native, FanGraphs writer Travis Sawchik is the author of the New York Times bestselling book, Big Data Baseball. He also contributes to The Athletic Cleveland, and has written for the Pittsburgh Tribune-Review, among other outlets. Follow him on Twitter @Travis_Sawchik.
Why has home-field advantage stayed nearly constant for more than 100 years? The home team bats last. They can win with a walkoff run. The visiting team cannot. That’s just enough of an advantage to alter the percentages in their favor.
This was my first thought as well, however it is almost impossible to study since the home team always bats last. That means that all the other possible advantages that home teams might enjoy are 100% correlated with batting last. We need the control situations of the home team batting first, or teams playing in neutral environments to test this.
Think of baseball as “reverse chess.” It’s been proven over more than 150 years of play that white should an advantage of between 52 and 56 percent because playing first, it dictates play. Baseball is different than most other sports because every half inning is an independent variable. The home team can win by making less than 27 outs. The visiting team cannot. And 54 percent is basically four walk-off wins every 100 games.
I agree it’s a good hypothesis, but I can’t think of any way to test it given the available data. Maybe if the play enough games in London (or elsewhere) on neutral fields we can start to get an idea.
Every chess move affects the rest of the board and is persistent. Other than maybe pitcher fatigue and subs, baseball moves affect only the score and outs. You are comparing apples to oranges.
Except that’s not actually an advantage: it’s even a detriment when the home team can make good use of defensive substitutes.
The home team did not always bat last. Early in baseball they could choose, and home teams often chose to bat first. You can look it up.
I’ve always thought the away team has had an advantage in the fact they (sometimes) only have to pitch 8 innings. Over the course of a season, that’s roughly 40 IP (~80 losses, roughly half on the road) of rest. Of course the drawback is that you lost the game (generally not good).
Homestands allow the home team to rest and that certainly has value. As comfortable as flying has become, traveling still takes something from you.
impressive article, thanks!
Very interesting article. I hope teams are taking this seriously for their players, but also the MLB schedule-makers. Too often players finish a game in one city, go to the airport, fly far away, arrive in the wee hours of the morning, and play another game, all in less than 24 hours. Who is that good for? Building in a few more off days can reduce how frequently these kinds of “throw-away” games have to happen.
One thing which wasn’t mentioned is that some road teams, instead of leaving for the airport right after the game on get-away day, are opting to stay overnight at their hotel and flying the next day. Otherwise, excellent post. Nice hire, Dave!
With this in mind, an interesting analysis would study:
Road winning percentage:
Game 1 of series
Game 2 of series
Game 3…
And then cut that by “preceded by off day” and “not preceded”.
I thought I saw a study of this somewhere, and it mentioned that for game 1 of a series, the road team was definitely at a disadvantage–I don’t remember the winning percentage, nor the exact details, so I could be wrong about such a study existing.
On the home team in 2-3-2 playoff series, the home team is the team with the better record, in most cases, so some effect is that home team is a better team.
Except for the WS where it was alternating or AS game deciding.
“Do we want all fields to play neutral or closer to neutral?”
If the lack of neutrality is the result (at least partially) of arbitrary decisions by non-players then the obvious answer is “yes.”
Im not sure I agree, when fans go to the game I think they deserve a better than even chance to see their team win, even that means stacking the odds a bit, could just be me though.
Agreed. Any part of HFA that is structural (e.g. Home team bats last) then that’s just the way it is. I mean, unless you wanted to flip a coin to choose the batting order each game. Any part of HFA that is about player rest, then that’s up to each team or even each player to figure out and it’s not really the purview of MLB itself. But if any part of HFA is about umpires subconsciously trying to please the home crowd, then that’s got to go.
Agreed: I don’t want the rules changed; home field bats last, and each park is unique (and thus the team can be built to best match its home park).
MLB tried “cookie-cutter” stadiums in the 80s and they were awful. Let Boston have Fenway, the Mariners have their big park, and let Baltimore have its great throwback stadium, and so on.
But when it comes to balls and strikes, please get us to Robot Umps ASAP. Having plays reviewed has gotten us a long way towards removing human error: robot umps would eliminate most of the remaining human error in the game. I’ve had enough of pitch-framing and badly called balls and strikes–as with the replays, let’s just get the call correct.
I wonder what the “true” strike zone height should be for players. It’s generally accepted that a tall guy who crouches down a bit doesn’t deserve to have an upper strike zone that corresponds to his standing height. At the same time, if you had a dynamic strike zone based on something like sensors on the player’s body, it would encourage hijinks like extreme crouching to shrink the zone.
Would it be difficult to give the home plate ump a clicker that could measure the top and bottom of the strike zone at the appropriate time?
Given the extremely fickle state of most players’ swing mechanics, I would imagine that selective extreme crouching would do more harm than good.
And regardless, if the rulebook says knees to numbers but doesn’t require the same stance from one pitch to the next then I don’t even see how you can call that “hijinks”.
Would be interesting to see how technology might disrupt pitch framing. I’m all for more tools being available for use, but I’d hope the human element is still present.
With that said, there’s nothing I hate more than retaliation balls and strikes called.
It’s already happened. Pitch framing is becoming less impactful.
Great article. Still in favor of robo-umps even if that diminishes home field advantage a tad. Get the calls right (and stop the annoying groaning from home fans when a strike is called against their batter!)
Obviously this wouldn’t completely take out all of your competing variables but couldn’t comparing win loss record the day after changing cities vs the rest of road games give some input to the effect of the actual travel on teams?
Very interesting piece, though I’m still really surprised that the crowd doesn’t have more of an effect on results. Same with travelling for road games, it seems really odd if it isn’t.
Anyway can someone explain to me as a UK baseball fan, how the strike zone is determined based on height. So is it just based on hitters height which the ump has to adjust for every hitter? That sounds like a difficult enough task in itself let alone then calling that zone correctly. And can I ask why do taller players have bigger zones, rather than just higher zones? Surely tall players are at unfair disadvantages because of this?
Thanks! The idea is the crowd is having an influence … mostly on the officials.
Yes true, but I expected there to be a crowd effect on the players two-fold. When I play sports home and away I always feel more comfortable and play better at home with the crowd in our favour. Either baseball players are just machines or crowds aren’t loud enough in MLB, get louder fans!
As I said above, would love to hear how the strike zone is determined based on height. As a UK baseball fan I’ve never quite understood that aspect.
Two points on HFA: 1) You can test the “batting last/walk-off” theory mentioned in the comments as the reason for the edge in HFA by checking scores at the end of the 8th inning. The same, roughly, 8% edge holds so batting last wouldn’t appear to explain things.
2) When the authors of Scorecasting attributed the difference in pitchers’ home and away bb and k rates to biased umpiring, they ignored the forgotten fourth true outcome of plate appearances, the HBP. Very consistently over decades, pitchers hit more batters on the road than at home. You can’t attribute that to biased umpiring. It’s more indicative of performance gaps between home and away that they, the authors, seemed to largely wave away.
Have Scorecasting’s results held up? At the time Phil Birnbaum and Dave Studes suggested that the authors were rather overstating the importance of umpire bias.
This is something I’ve thought about. Could part of the HFA be due to the fact that in extra innings, the typical management style of the road manager is way less effective than that of the home manager? It feels like if every road manager used closers first, that number could scoot closer to 50%
While the difference in K and BB rates may account for 70% of the overall HFA obviously umpires are not responsible for all or even most of the K and BB differences. Even if we had robot umps home players would still walk more and K less because they play better and road pitchers pitch worse. When batters play better they walk more and K less and when pitchers play worse they walk more and K less.
I think that umpire bias as measured by called strikes and balls relative to pitch location account for about 1/3 of the HFA at most.
As Joe Peta (and others) pointed out above batting last has nothing to do with HFA as we see a consistent HFA in all innings, other than the first. The home team has a huge advantage in the first inning. We don’t know why. Could have something to do with familiarity of the mound or field (for some reason it is magnified in the first inning) or something to do with how the pitcher warms up and then goes to the mound (road pitcher warms up then goes to the bench whereas home pitcher warms up and then goes right to the mound.
HFA has almost nothing to do with the crowd in baseball. As I said around a 1/3 is umpire bias, 1/3 maybe the first inning extra HFA, and the rest park familiarity and tailoring the home team to the park, as well as sleep and other lifestyle habits (eating) and just the general psychological effects of being at home.
Travel fatigue also has little to do with HFA in baseball.
Kind of wish the author had realized the difference between K and BB rate and umpire bias. Also kind of wish the author had done a little more research on what makes for HFA in baseball. I don’t think HFA is endangered.
“HFA has almost nothing to do with the crowd in baseball. As I said around a 1/3 is umpire bias…”
What causes the umpire to be biased for the home team? Why not be biased for the “better” team? Isn’t that the home crowd affecting the game?
Has anyone looked at HFA in leagues that are more closely located, like spring training or high school, where you expect less of a home crowd bias? Is HFA the same across all of the minor leagues?
Regarding ball and strike calls – why aren’t robots and humans working in tandem?
Humans to call the vertical plane, robots to call the horizontal plane. A great deal of missed calls would be immediately addressed this way, and the job gets easier for the humans, presumably, since they can focus solely on the height of the pitch.
“And, in a way that’s unique to baseball, teams can tailor their roster to their home park’s dimensions.”
Probably true as specifically and narrowly stated, but football teams can analogously tailor their rosters and systems to home field conditions. For example, a football team can put in place an offensive system and offensive personnel that are optimized for playing on artificial turf in a dome but aren’t as well adapted for playing outdoors on grass in windy, rainy weather.
I guess we’d need tests or some sort of data to determine if this is true. But from just my own hunch/thoughts, I think that a baseball team can do a lot more. You can change the height of the grass to speed up or slow down batted balls, depending on the speed/range of your players. You can roster left-handed pull hitters with medium power to hit the ball over a short right-field porch. You can get a bunch of right-handed pull hitters that like to hit line drives to bang a bunch of their batted balls into the Green Monster.
It just seems to me that there are a lot more ways to maximize your roster to your park in baseball–many of which are based on the design of the park, which, by definition are unavailable to the other major sports.
“[Home—field advantage] was virtually the same win MLB from 1903 to 1909 as it was from 2003 to 2009. This suggests that teams jetting on chartered flights with catered meals, high-thread count linens and flat-screen televisions have no more success than did teams that traveled to games in Pullmans and buses. (Either travel isn’t causing the home advantage or teams need to rethink their jet purchases and on-flight catering).”
That’s an interesting finding, but MLB teams in 1903 to 1909 also weren’t located farther west or south than St. Louis. So one factor pushing against the advantages of modern travel is that teams travel greater distances and across more time zones – e.g., East Coast teams playing West Coast night games that start after 10 pm Eastern time.
I remember reading, like 20 or 30 years ago, that someone studied aspects of jet lag by analyzing baseball game results. The conclusion, as I recall, was that teams traveling west coast to east coast were as much as an entire run worse than teams who flew in the opposite direction. Obviously the study normalized results somehow for underlying team performance abilities. It might be interesting to increase the resolution of the data here to see whether jet lag is a factor or whether HFA is constant across distances and time zones.
Nothing has a “1 run effect” certainly not jet lag. There is a study that concludes that traveling cross country in one direction significantly affects HFA due to circadian rhythms. I tried but was unable to duplicate their results. As far as I know no one takes that study seriously but I can’t speak to it precisely.
Its not just the act of traveling. Humans sleep worse in unfamiliar environments.
Think of it as the “oh shit there might be a Sabre toothed tiger in this room” effect. Its well documented.
By the third night in the new location you’re resting as if it were your home.
But for baseball players they leave that night.