Foul Ground Home Field Advantage
By no means am I a veteran of the press box. But already I’ve tapped into a strange phenomenon that may be universal to all baseball writers — I’m not rooting for a team, I’m rooting for my story.
For example. Going into the American League Division Series, I decided I would do some research on the home field advantage offered by the extensive foul ground in Oakland. After gathering some quotes — easy, considering the optics of Josh Donaldson, running forever for a foul ball — I was ready for a Game-Changing Moment. A foul ball over the bullpen mound in the late innings. A missed foul ball by the Tigers. I was rooting for a foul ball.
I didn’t get my big foul ball, or you’d have gotten this story, with a much different lede, weeks ago instead of now. Looking back through my numbers, though, there was a surprise. More the kind of surprise you get when you turn on the television in the fall and get a random Arizona Fall League game. More that than the sort of surprise that Nate McLouth got when Donaldson did this:
The stage was set for this sort of surprise during the ALDS. Bob Melvin said he chose Daric Barton for a few of those games because the first baseman “covers a lot of ground” and would help specifically because of the foul ground in Oakland. “You have a hundred-something feet out there,” Barton said. Barton added “there’s a lot of outs out there,” but admitted that “they say it’s worth about ten-to-fifteen points on your batting average.” He laughed about that one, though — that cuts both ways and “it’s so frustrating.” But Barton felt that he knew the ground well and that there were plenty of times when he got to balls that people felt were automatically in the seats.
Watch that highlight of Donaldson, and listen to Barton talk about how he can avoid the mound, and about plays like the one where Trevor Crowe went after a foul ball and fell on the bullpen mound, and it’s easy to think: Oakland must get home field advantage from their intimate knowledge of the foul ground.
Turns out, not so much. Check out the home and away foul ball outs by stadium, 2012-2013:
| Home Team | TFO | RTFO | HTFO | HFA |
|---|---|---|---|---|
| Phillies | 204 | 119 | 85 | 34 |
| Mets | 215 | 123 | 92 | 31 |
| Twins | 191 | 110 | 81 | 29 |
| Padres | 158 | 92 | 66 | 26 |
| Dbacks | 207 | 112 | 95 | 17 |
| Rangers | 188 | 99 | 89 | 10 |
| Giants | 218 | 112 | 106 | 6 |
| Angels | 188 | 96 | 92 | 4 |
| Brewers | 210 | 107 | 103 | 4 |
| Athletics | 398 | 200 | 198 | 2 |
| Cubs | 149 | 75 | 74 | 1 |
| Indians | 173 | 87 | 86 | 1 |
| Red Sox | 165 | 83 | 82 | 1 |
| Yankees | 209 | 105 | 104 | 1 |
| Royals | 238 | 119 | 119 | |
| Astros | 194 | 96 | 98 | -2 |
| Reds | 182 | 90 | 92 | -2 |
| White Sox | 222 | 109 | 113 | -4 |
| Rockies | 139 | 67 | 72 | -5 |
| Braves | 214 | 104 | 110 | -6 |
| Tigers | 228 | 110 | 118 | -8 |
| Dodgers | 165 | 78 | 87 | -9 |
| Pirates | 157 | 70 | 87 | -17 |
| Mariners | 270 | 126 | 144 | -18 |
| Nationals | 223 | 102 | 121 | -19 |
| Rays | 242 | 111 | 131 | -20 |
| Orioles | 218 | 98 | 120 | -22 |
| Cardinals | 204 | 89 | 115 | -26 |
| Marlins | 198 | 86 | 112 | -26 |
| Blue Jays | 254 | 103 | 151 | -48 |
TFO = total foul outs in the stadium
RTFO = foul outs that occurred while the road team is batting
HTFO = foul outs that occurred while the home team was batting
HFA = RTFO – HTFO
Yeah, so Oakland, over the last two years, has played almost exactly neutral in foul grounds, despite having the largest foul ground in baseball.
Thanks to Bill Petti’s Interactive Spray Chart Tool, we can see, graphically, how different the foul grounds are compared to, say, the number one team in foul ground home field advantage: the Philadelphia Phillies.
You can’t quite blame this on personnel. According to Josh Reddick, the outfield portion of the foul ground is about the same as most parks, and he figures that it’s mostly on the first baseman and third baseman to get to these balls. The combined UZR over the last two years for the first and third baseman on the Phillies (-10) and Athletics (+31.3) has the signs reversed if you want to blame this phenomenon on the defenders on the corners. Shortstops and second basemen can get to some of those balls, too, though — the range components for the infield in Oakland (+22.4) and Philly (+24.9) tell a bit of a different story.
So let’s put the foul ground home field advantage up against the UZR infield range factors for the past two years. It doesn’t seem to match up.
| Home Team | HFA | IF RngR |
|---|---|---|
| Phillies | 34 | -1.5 |
| Mets | 31 | -1.4 |
| Twins | 29 | -33.3 |
| Padres | 26 | -6.9 |
| Dbacks | 17 | -33 |
| Rangers | 10 | 17.9 |
| Giants | 6 | 5.5 |
| Angels | 4 | 34.7 |
| Brewers | 4 | -21.6 |
| Athletics | 2 | 22.4 |
| Cubs | 1 | 51.6 |
| Indians | 1 | -52.6 |
| Red Sox | 1 | 26.7 |
| Yankees | 1 | -13.1 |
| Royals | -10.4 | |
| Astros | -2 | -42.1 |
| Reds | -2 | 43.1 |
| White Sox | -4 | -19 |
| Rockies | -5 | -27.7 |
| Braves | -6 | 18.3 |
| Tigers | -8 | -29.7 |
| Dodgers | -9 | 64.6 |
| Pirates | -17 | 23.1 |
| Mariners | -18 | -3.1 |
| Nationals | -19 | 20.5 |
| Rays | -20 | 57 |
| Orioles | -22 | 28.5 |
| Cardinals | -26 | -46.9 |
| Marlins | -26 | -45.5 |
| Blue Jays | -48 | -26 |
So if it’s not about the field itself, and it’s not about the interaction between the field and the specific players about it, is there another source of foul-ground-inspired home field advantage? What about day/night splits? Perhaps some players know their foul ground better and can navigate it better at night?
| Home Team | HFA Night |
|---|---|
| Padres | 27 |
| Phillies | 25 |
| Dbacks | 21 |
| Mets | 14 |
| Twins | 14 |
| Angels | 7 |
| Rangers | 6 |
| Dodgers | 3 |
| Giants | 1 |
| Red Sox | 1 |
| Rockies | 1 |
| Royals | -1 |
| Pirates | -3 |
| Tigers | -3 |
| Athletics | -4 |
| Braves | -4 |
| Astros | -5 |
| Reds | -5 |
| Indians | -6 |
| Brewers | -7 |
| Nationals | -7 |
| Cubs | -10 |
| Rays | -10 |
| White Sox | -10 |
| Yankees | -11 |
| Orioles | -18 |
| Cardinals | -21 |
| Mariners | -24 |
| Marlins | -26 |
| Blue Jays | -37 |
Well, it works in Philadelphia for some reason. Guess I’ve got a story for the next time Cody Asche ranges deep into foul territory in a big spot in a big night game for the Phillies next year.
With a phone full of pictures of pitchers' fingers, strange beers, and his two toddler sons, Eno Sarris can be found at the ballpark or a brewery most days. Read him here, writing about the A's or Giants at The Athletic, or about beer at October. Follow him on Twitter @enosarris if you can handle the sandwiches and inanity.

I don’t know if you can actually check statistics on this, but is it possible that a team like the Phillies had fewer fly ball fouls in their home foul ground than their opponents?
Yeah, it seems this could likely be caused by a pretty direct correlation of IFFB rate.
I thought that would be the case, too. A quick look yielded these results (IFFB, home split):
Athletics,0.131
Blue Jays,0.116
Braves,0.116
Mets,0.112
Royals,0.111
Marlins,0.109
White Sox,0.108
Yankees,0.108
Orioles,0.107
Rays,0.105
Mariners,0.105
Dodgers,0.103
Phillies,0.101
Astros,0.101
Rangers,0.1
Pirates,0.096
Red Sox,0.095
Twins,0.094
Cardinals,0.092
Tigers,0.09
Nationals,0.088
Padres,0.086
Cubs,0.086
Angels,0.084
Rockies,0.083
Reds,0.081
Brewers,0.08
Diamondbacks,0.08
Giants,0.079
Indians,0.071
There doesn’t appear on first look to be that strong a correlation between high Home IFFB rate and Eno’s HFA.
If I’m not mistaken this number would correlate to HTFO, not HFA.
HFA would correlate to (Home team IFFB rate / Away team IFFB rate).
The point is, if your team just simply hits more popups at home than whoever it is that the away team is does, then it would skew your numbers. So there is a lot of room for random error here, especially over single-year IFFB numbers.
The results of this study simply aren’t reliable since they aren’t statistically significant.
We should be using two-year samples since the study above is a two-year sample. But I agree, I don’t see this matching up either.
It may not be only correlation with the home team’s IFFB rate, but also with their opponents IFFB%. Notice how the 4 out of the top 5 NL teams in IFFB% are NL East teams. So the Phillies and Mets may play a disproportionate number of their games against high IFFB% teams.
Where would one get this data? I would be interesting in looking at [ HTIFFB/HTFA ] / [ RTIFFB/RTFO ]
I’ve always wondered about the impact of foul territory on the game. I feel like you just scratched the surface here. For instance, while there may not be a strong home/road split, it appears pretty clear that more foul outs occur in Oakland. How does that affect the A’s wOBAs?
Also, the A’s corner infielders had high UZRs relative to the Phillies. Is that because they cover more ground and are thus better defenders, or is it possible that they get credit for balls that other defenders don’t simply because the field is big enough to allow it. How do defensive metrics take field dimensions into account?
I swear I wrote my comment below before seeing this. Good thoughts.
Very true, there’s likely no strong standard in place for calculating park effects in UZR. They try to account for it, for sure, but there’s still a strong chance that UZR is affected by this.
As with pretty much all areas of scientific study, knowing the areas in which your tool lacks precision is the most important part of using the tool.
I would be curious (RE: Barton’s comment) who gains and loses the most value offensively and defensively with variations in foul ground. Wouldn’t UZR overvalue A’s corner IFs because they are able to make plays in zones other fielders literally cant? I believe UZR considers the ballpark when determining the zone and difficulty of a play, but then there is a pretty small sample size regarding how to score a play in A’s foul ground.
Could try to see who hits the most foul balls and how they do in the big foul territory parks…
RTFO needs to be reduced to account for the fact that home teams don’t always bat in the bottom of the 9th or don’t have complete half innings in the bottom of the 9th or in extra innings. This year road pitchers recorded 5.4% fewer outs than home pitchers. Of course, the more games a team wins at home the larger the adjustment that needs to be made.
Duh. Didn’t think of this. Unsure how to account for it though.
My research on foul area vs. strikeouts has demonstrated a remarkable effect, And one that has changed over the years. While I haven’t updated my foul areas to all modern stadiums, it can be easily done. Lists of stadiums, mostly older ones, can be found on web site and blog, along with foul areas. I’m sure analyses posted here could be adjusted per square or linear foot foul area dimension.
On your last point: so then we could say they get to x amount of foul pop-ups vs foul square foot? Hm. That’s going to be 95-99% I’d bet. These are all still ‘infield flies’ basically, which is I think the main reason why HFA is hard to show.
When a park is ranks as pitcher-friendly, hitter-friendly, or neutral, is the amount of foul-ball territory considered? Also, would appreciate a column in with the amount of foul ball territory per park in the stats. I got it that Oakland has a lot, but what is the ranking of other parks?
Amount of foul territory is considered when labelling a park pitcher or hitter friendly.
I would also like a square footage ranking of mlb foul territory.
Er, “when a park is ranked”. This is related to the comment above, can you show the number of foul popups per square foot of foul area? Does it matter how far the “thicker” foul territory extends? I would think a short but wide territory near the plate would result in more foul outs than a long narrow foul territory, but I could be wrong–depending on the skill of the catcher vs. skill of other fielders.
For those interested in FA, AL stadiums ’69-2000 FA listed here: http://www.beyondtheboxscore.com/2011/9/2/2401228/foul-area-vs-strikeouts-1969-2000. Relationship of Strikeouts/ sq. ft. FA changed after’69 compared to previously. Probably fewer foul balls with lower strike zone. There’s been a decrease in popups recently as we’ll.
So if you had around 225 of these foul balls where would your two standard deviations be? The distribution in the table above seems like a random distribution. Would be interesting to see it proven not to be.