How Home Field Advantage is Like Mike Trout

Home-field advantage is a strange concept, or should I say, a strange reality. It doesn’t really matter, for our purposes, why it exists — it just matters that it exists. It’s there, all of the time, in every single baseball game, and while I wouldn’t say it’s an unspoken thing, it’s seldom thought of in depth. A team playing at home has an advantage it wouldn’t have in a neutral site. A team playing on the road is at a corresponding disadvantage. We accept that it is, and we don’t talk much about it, and when we talk about potential edges, it’s usually ignored in favor of pointing at match-ups. It’s almost too boring to point out Team X stands better odds because they’re playing in their ballpark. Someone’s always playing at their ballpark.

But home-field advantage is exactly what the Reds and Pirates have to play for this weekend. Very fleeting home-field advantage — home-field advantage in the one-game wild-card playoff between the two rivals. The teams will play three before they play one, and the Pirates are 50-31 at home, while the Reds are 49-28. Each would prefer to play before its own partisan audience. It’s obvious that it matters who gets to play at home. But how much does it matter? What’s a way that we can think about this?

Fortunately, it’s simple to figure out how home field will be decided. The Pirates, right now, are a game ahead of the Reds in the standings. In head-to-head match-ups, they’re even at 8-8. If the Pirates win this series, they’ll play at home. If the Reds sweep the series, they’ll play at home. If they win the series 2-1, they’ll play at home, still, because they’d have the head-to-head advantage, 10-9. The winner of this series gets home field, all uncomplicated-like. Let’s just ignore, for the sake of simplicity, that the Pirates are three back of the Cardinals.

So what’s the history of home-field advantage? This, too, is pretty simple. All-time, in the playoffs, the home team has won about 54% of games. Just since 1995, the home team has won about 54% of games. In the regular season, the home team has won about 54% of games. We’re getting a pattern. We’re going to stay simple and use that 54%. The Reds and the Pirates are pretty even teams, in record and in talent. We’re going to say that the home team in the one-game playoff would win about 54% of the time. So what’s at stake this weekend is a 54% chance of advancing to a real playoff round. Lose, and it’s more like 46%. It’s simultaneously a big and small difference.

The Reds and Pirates might argue that they have particular, unique home-field advantages, that are stronger. That they’re tailored to their ballparks, that they’re exceptionally familiar with their ballparks. Maybe, but probably not. Single-season home records have to be really strongly regressed, and it’s best to just use the average advantage. You’re free to disagree, but if you disagree, the following analysis probably won’t suit you.

So what’s another way of looking at what’s at stake? For this part, we’re going to make use of the Log5 method, which I will link instead of explain. As noted earlier, the Reds and Pirates seem pretty even. Maybe you think they’re true-talent 90-win teams. For a one-game playoff, they could optimize, maybe briefly making them 95-win teams. These numbers don’t really matter. Let’s just accept that there isn’t a huge difference between the ballclubs.

All things being equal, if you have even ballclubs, you’d expect each to win 50% of the time. So then: what does it take to turn 50% into 54%? What kind of boost do we have to give one of the teams to turn the winning odds into 54/46, to mirror home-field advantage?

The answer: 6-7 wins. A 97-win team would beat a 90-win team 54.4% of the time. A 102-win team would beat a 95-win team 54.5% of the time. At least, that’s what Log5 says, and Log5 is smarter than I am. That’s one way to think of the strength of home-field advantage. It makes a team act like it’s several wins better.

If you figure the Reds and Pirates are even, home-field advantage gives one a winning-percentage edge of about four percentage points. A way more fun way to imagine this is that home field would allow the Reds to exchange Jay Bruce for Mike Trout. It would allow the Pirates to exchange Pedro Alvarez for, say, healthy Miguel Cabrera. What’s at stake this weekend is home field in one game, and that’s the equivalent of one team being able to add a superstar before playing at a neutral site. So home-field advantage = added superstar + neither team at home.

At the same time, that makes it seem big and small. There’s no more impactful player than Mike Trout. But, this isn’t basketball. One position player in baseball can mean only so much, and in one game, on average, you’d expect Trout to be worth 0.3 – 0.4 more runs than Bruce. You’re talking a fraction of a hit, a fraction of a total base. Incrementally better defense. There’s a lot that takes place that doesn’t involve one of many starting position players, and so a team at home can still lose quite often. In last year’s wild-card playoffs, the Rangers lost to the Orioles in Texas, and the Braves lost to the Cardinals in Georgia. Home-field advantage is real, even if we don’t quite know how and where it shows up, but like all advantages, it’s limited in strength.

If you want to downplay the importance, you can focus on that 54%-46% separation. That’s a gap of just eight percentage points. Play at home, and you have a good chance of moving on. Play on the road, and your chance is just a little less good. Ehh. If you want to emphasize the importance, you can focus on swapping Jay Bruce for Mike Trout, or Pedro Alvarez for peak Miguel Cabrera. Ultimately, everything gets you to the same place, but our brains can see the same things so differently. Who wouldn’t want to rent a superstar for a day? The winner of this Reds/Pirates series, effectively, gets to do just that.



Print This Post



Jeff made Lookout Landing a thing, but he does not still write there about the Mariners. He does write here, sometimes about the Mariners, but usually not.


Sort by:   newest | oldest | most voted
my jays are red
Guest
my jays are red
3 years 5 days ago

it’s okay to say that they lost in atlanta and not georgia. it’s not the georgia braves!

brig_pudding
Guest
brig_pudding
3 years 5 days ago

A better way to do this would be to look to the sports betting market. In 2013, home favorites win at a rate of 58.6%. Away favorites are at 56.6%. Note that home field advantage is already adjusted into the prices.

section223
Guest
section223
3 years 5 days ago

If the home field advantage is already adjusted into it, those win rates should be exactly even. If anything, those numbers just tell you to bet the home favorite.

Cliff
Guest
Cliff
3 years 5 days ago

No, because the price (odds) is different for the home/away favorites…

Justin
Guest
Justin
3 years 5 days ago

No because home favorites might have a larger line than away favorites on average. Those numbers don’t really tell us anything.

some.guy
Guest
some.guy
3 years 4 days ago

Betting lines don’t measure reality, they measure the betting market’s perception of reality, which is similar but different.

So looking at betting lines is a worse way to do this.

Hurtlockertwo
Guest
Hurtlockertwo
3 years 5 days ago

I would say that a team with one dramatically good starting pitcher has the advantage, at home or one the road. Have King Felix or Yu darvish or Kershaw as your starter and your team will have an advantage in a one game scenario.

Hurtlockertwo
Guest
Hurtlockertwo
3 years 5 days ago

I would also think that the great pitcher would have more impact in one game than Trout.

cass
Guest
cass
3 years 5 days ago

Also true over a whole season. That’s why position players should never win the MVP. They only bat during 1/9 of a team’s play appearances, while a pitcher faces every opposing batter.

(If pitcher’s pitched all nine innings every time out, this would be true, but it’s pretty even since starters only go about six. 1/5 * 6/9 = 1/.7.5. When you factor in fielding, starting pitchers and position players have about equal effect on a game, which is why their WAR totals are about even.)

cass
Guest
cass
3 years 5 days ago

*pitchers, not pitcher’s

josh
Guest
josh
3 years 5 days ago

“That’s why position players should never win the MVP. They only bat during 1/9 of a team’s play appearances, while a pitcher faces every opposing batter.”

what

cass
Guest
cass
3 years 5 days ago

I was making a joke based on the silly “Pitchers shouldn’t win the MVP because they only pitch once every five days.” argument. Tongue-in-cheek.

I seriously have no idea why no one ever says “A hitter only bats in one ninth of his team’s plate appearances”, though when they make such a big deal about the once every five games thing.

Just riffing on the idea that a pitcher is vastly more important to a single game than any individual hitter because they’re involved in every play.

Jeff
Guest
Jeff
3 years 4 days ago

It’s quite obviously because they also play 9 innings of defense. But defense doesn’t matter for MVP…ahhh cognitive dissonance…becoming too much…*brain explodes*

Richard LaPointe
Guest
Richard LaPointe
3 years 4 days ago

I totally agree that pitcher is more impactful than a field player!

Daniel
Guest
Daniel
3 years 4 days ago

this reminds me too much of how the Rangers lost the wild card game last year at home with a red-hot Darvish vs. Joe Saunders

DJG
Guest
DJG
3 years 5 days ago

I was curious about how the starting pitchers would change the odds. And would you sit your ace and possibly lose home field advantage to save him for the playoff? Unfortunately, it’s probably a moot question, both Liriano and Latos threw last night, so they can’t go this weekend and will be fresh for the Tuesday playoff.

Hawk Harrelson
Guest
Hawk Harrelson
3 years 5 days ago

100%. That’s why we give them wins.

jessef
Guest
3 years 4 days ago

that’s a great question

i think it has to come down to scheduling and the team’s pitching staff

on the one hand, you should reserve your ace for the playoff because the difference between an ace can make in one game far exceeds homefield advantage (which according to the article is roughly equivalent to the difference a position player makes in one game)

on the other hand, if you use your ace to try and get homefield advantage, if you win the one-game playoff, you should be able to use him twice in the five game series

matt w
Guest
matt w
3 years 4 days ago

You should absolutely save your ace for the play-in game. In fact, if you’re tied for the division lead (with the loser getting a WC) and you’re in a position to use your best pitcher for the last regular season game or save him for the next game, you should save him; the play-in game is much higher leverage than any other game he could pitch.

Jeff
Guest
Jeff
3 years 5 days ago

I know this isn’t the point of the article (an enjoyable one for sure) but the source of home field advantage certainly does matter. If, as a popular hypothesis suggests, human umpires are subconsciously influenced by the “roar of the crowd” then a movement toward “robotic” umpires should fundamentally reduce home field advantage. It provides an interesting natural experiment – as baseball moves toward more replay and more “automated” calls, does home field advantage go down?

Anyhow, good post.

B N
Guest
B N
3 years 4 days ago

Not sure if I buy that. I think that homefield advantage has a lot to do with building your team for your park. There’s been findings that indicate that homefield advantage has decreased over the years, primarily due to changes in how parks are designed (e.g., more uniform dimensions and avoiding ridiculous things like Candlestick wind patterns).

See: http://baseballanalysts.com/archives/2009/04/a_study_in_home.php

With that said, there is an equal possibility that certain park features “increase” your “home advantage” by actually making you worse on the road. Extreme hitters and pitchers parks may actually create bigger splits by forcing your players to adopt home habits that are not good on the road (e.g., pre-humidor Coors was horrible for throwing breaking balls, which can’t be good for your pitching habits when you leave home).

Mookie
Guest
Mookie
3 years 4 days ago

Homefield advantage is also built into the game. Going second is a nice advantage.

BaseballSplits (Twitter)
Guest
BaseballSplits (Twitter)
3 years 4 days ago

Batting second isn’t as much of an advantage as people think it is (if it’s even an advantage at all). In order to win, you have to get 27 outs regardless.

Just because you KNOW that a base hit will end the game in walkoff doesn’t make it any more likely you are going to get that hit…

BaseballSplits (Twitter)
Guest
BaseballSplits (Twitter)
3 years 4 days ago

Sports Illustrated had an interesting article on home field advantage a couple years ago, with excerpts from the book “Scorecasting” — basically what Jeff said is correct.

Research across various team sports showed significant evidence that umpires and refs have by far the largest impact on HFA… not because they are evil, it’s because they are human and want to “conform” to the opinion of the crowd. A telling fact that discounts the “different types of ballparks” theory is that baseball has (by far) the lowest HFA effect out of the major team sports, despite the individual playing fields being much more unique (as you mentioned). This is because umps/refs have less of an impact on the game result in baseball compared with other sports (with no penalties/fouls in baseball, calls are less subject to personal interpretation).

Anyway, it’s a good read if you ever get the chance to look it up.

evo34
Guest
evo34
3 years 3 days ago

No, that’s not why baseball has such a small HFA. It’s because road teams only have to travel once for a 3-4 game series, not 3-4 separate trips, as it is in other sports. More significantly, the HFA is smaller because luck is such a huge component in baseball, vs., say, basketball or football. If the very, very best teams in MLB only win 60% of their games due to the effects of variance, then this same high variance will make HFA less influential in determining game outcomes.

Ari Indik
Guest
Ari Indik
3 years 5 days ago

Not really relevant to this weekend, but I wonder if we can separate the home field advantage into a) the advantage of being at home, and b) the advantage of hitting in the bottoms of innings.

TKDC
Guest
TKDC
3 years 5 days ago

This was exactly what I was thinking. Even if you played at a neutral cite, one team gets the advantage of batting in the bottom half. I wouldn’t be surprised if this was the greater advantage.

DJG
Guest
DJG
3 years 5 days ago

I’ve always wondered: Is batting last an actual advantage? I get the argument, that knowing how many runs you need to score in the bottom of the ninth is an advantage. But whether this advantage actually manifests itself isn’t obvious to me, especially given that managers frequently muck up the endgame (see, sac bunts, closer usage).

I’m not sure how you would even test this. Tally up past all-star game records?

TKDC
Guest
TKDC
3 years 5 days ago

In theory, batting last is of course an advantage, but your point about the tendency to much this up is well taken. As far as how to test, how about Yankees/Rays games? Don’t the Yankees have home field advantage in both cities?

Jason B
Guest
Jason B
3 years 5 days ago

“In theory, batting last is of course an advantage”

Granted, I may be dense, but why? I guess you can do a few more situational things?

DJG
Guest
DJG
3 years 4 days ago

I’m not ready to say it’s “of course an advantage”, even in theory, but I see how it *could* be an advantage.

For example, in a tie game in the bottom of the ninth, the home team can play to maximize their chances of scoring a single run instead of maximizing their expected total runs scored.

But the thing is, the fielding team has the same knowledge and knows to play against one run. So without seeing a detailed game-theory study, I’m not convinced there is any advantage, even a theoretical one.

And then when it comes to how teams actually play in practice, that’s a whole other issue. Looking at All-Star Games would be kinda interesting. I might do that.

TKDC
Guest
TKDC
3 years 4 days ago

Maybe I should have left out “of course” but I doubt seriously if you did a full game theory analysis, it would not come to the conclusion that batting last was an advantage. After all, you also have the ability as the home team to know exactly what the team batting in the top of the ninth needs to tie the game/take the lead and can defend accordingly. I wonder how much advantage is even possible from defensive positioning based on different score/base runner environments? It is clear that you could strategize differently based on how many runs you know you need to score.

But yes, I guess it is possible, and “of course” was too strong.

One other advantage (in the NL) is that since you are pitching in the top of the inning, it is easier/more likely to get an extra inning out of your starter before pinch hitting if you are the home team.

Ari
Guest
Ari
3 years 4 days ago

There’s also the advantage of the walk off. Once you score the winning run, you win, that’s it. No lead to protect. It also changes how you use your bullpen.

BN38
Guest
BN38
3 years 4 days ago

The best way to test the “batting last” being an advantage theory at a neutral site would be to look at WBC or Olympic games I would think.

I’m not sure if the sample size would be big enough, but in those environments you have both teams playing at a neutral site.

The all-star game has too many hijinks and other anomalies (including a bench full of all-stars, not 2nd class players) to be able to extrapolate onto a regular game.

DodgersKingsoftheGalaxy
Guest
DodgersKingsoftheGalaxy
3 years 4 days ago

Disadvantage, after Donnie bunt****s us because we only “need 1 run”………

chuckb
Guest
chuckb
3 years 5 days ago

Really interesting, Jeff. Thanks. Adding a 6-7 win player gives us a good breakdown of what home field advantage is worth in MLB.

D
Guest
D
3 years 5 days ago

Does removing interleague records affect the home-team advantage %? I wonder what the data shows for home team benefits via DH (AL) or pitchers who regularly bat (NL).

DD
Guest
DD
3 years 5 days ago

I think you mean “home-field advantage = added superstar + neither team at home – ABOVE AVG PLAYER. You are netting 10 Win Trount with 4 win Bruce to get your 6 wins, right?

jessef
Guest
3 years 4 days ago

trout isn’t a superstar, he’s a superhuman

Catoblepas
Guest
Catoblepas
3 years 5 days ago

I’d be interested to see what teams historically over- or under-perform their expected home winning percentage (based on the home-field advantage and their number of wins). What teams have parks+fans that are several wins above/below replacement? My picks are the Rangers and Marlins, respectively.

evo34
Guest
evo34
3 years 3 days ago

I have done research to project future HFA based on past HFA. My current top projections are:
Col 6.1%
Det 5.1%
Oak 5.1%
Pit 5.0%

Bottom:
Mia: 3.7%
LAA: 4.0%
Phi: 4.1%
NYM: 4.1%
SF: 4.1%
Atl: 4.1%

bowie
Member
bowie
3 years 5 days ago

I can’t believe Jeff used the word “impactful.”

Andy
Guest
Andy
3 years 4 days ago

You say that the home team wins 54% of the time. But that is an average, which means that to say it has a 54% chance of winning a particular game, everything about that game has to be average. In particular, in an average game, each team will start its best pitcher, or its second best pitcher, 3d best, etc.

But what if that’s not the case? Suppose the home team has its ace starting, and the visiting team has its 2d or 3d best pitcher? That 54% goes up. And if the pitching is reversed, it goes down. And though I don’t have stats to back this up, I’d bet that having your ace on the mound on the road vs. a 2d-3d-4th choice starting for the home team is better than ace vs. ace at home. Of course how much of an advantage depends on exactly how good each member of each rotation is, but presumably there is a significant advantage whenever one team has its ace going and the other team does not.

That being the case, surely the first priority of both teams is to make sure their best pitcher is rested, and can start the one game playoff. Even if that means missing a turn in the final games of the season, and worsening the chances of getting HFA.

Andy
Guest
Andy
3 years 4 days ago

You say that the home team wins 54% of the time. But that is an average, which means that to say it has a 54% chance of winning a particular game, everything about that game has to be average. In particular, in an average game, each team will start its best pitcher, or its second best pitcher, 3d best, etc.

But what if that’s not the case? Suppose the home team has its ace starting, and the visiting team has its 2d or 3d best pitcher? That 54% goes up. And if the pitching is reversed, it goes down. And though I don’t have stats to back this up, I’d bet that having your ace on the mound on the road vs. a 2d-3d-4th choice starting for the home team is better than ace vs. ace at home. Of course how much of an advantage depends on exactly how good each member of each rotation is, but presumably there is a significant advantage whenever one team has its ace going and the other team does not.

That being the case, surely the first priority of both teams is to make sure their best pitcher is rested, and can start the one game playoff. Even if that means missing a turn in the final games of the season, and worsening the chances of getting HFA.

Edit: Someone posted above that the aces of both teams will be ready for the playoff game, as they took their turn just before this series. Still, if that had not been the case, it would raise an interesting issue, in which a manager has to balance the value of HFA vs. the value of having the ace starting. Since starting the ace during the weekend would definitely mean he wouldn’t be available for the playoff game, whereas not starting him would not definitely mean the team would lose, and even if it did might not definitely mean they lost HFA, I would guess in almost all situations it would be better to rest the ace.

Jim Bouldin
Guest
3 years 4 days ago

Yes, right on the money with that.

Furthermore, you don’t use a long term average to predict a one-time event without first doing tests of autocorrelation to make sure that a better predictor, one based on more recent performance, does not in fact exist. And very often, it does indeed.

This post points up the problems with much of sabermetric analysis–it often assumes that some long term value is the best predictor of something, without checking other evidence.

olethros
Guest
olethros
3 years 4 days ago

This post points up the problems with most critics of sabermetric analysis – namely that they ascribe statements to sabermetric analysis that it didn’t actually make. In this case, no prediction was made, and no claim that HFA is the “best predictor” of anything was made.

evo34
Guest
evo34
3 years 3 days ago

No one ever said that all home teams win 54% of all of the games. They said that in an average game (teams equally matched), the home team has an extra 4%-5% edge. This edge is meant to applied to the already-calculated odds of a team winning at a neutral site. Really not that difficult..

pft
Guest
pft
3 years 4 days ago

Having 1 more game at home means little. 54% is significant over a large number of games, not so in 1 game. Pitching matchups are a far greater factor in the outcome, as well as luck.

Josephine
Guest
Josephine
3 years 4 days ago

HFA is ~4.5%. So it is a swing of 9 percentage points. Nothing to sneeze at. It would take a pretty big difference in pitcher ability to lose that 9% WP edge.

evo34
Guest
evo34
3 years 3 days ago

If 54% means little for one game, then why not just bench your 2 top hitters? It will have a similar impact in win odds.

wpDiscuz