Home-Field Advantage and Game 7

The Dodgers’ reward for having the best record in the majors this year, for recording three more wins than the Astros during the regular season, is this: they host the final game of the 2017 major-league campaign, Game 7 of the World Series, tonight.

Home-field advantage is a real phenomenon. In the game’s history, World Series Game 7 victors have been split evenly, 19-19, between home and road teams. But that’s a fairly small sample. In baseball, the home team typically wins 54% of regular-season games, and that rate holds up as a constant decade after decade, era after era.

Conventional wisdom suggests that home-field advantage is the result of multiple advantages enjoyed by the home team: the right to bat last, the ability to tailor a roster specifically to the park, a certain comfort with the surroundings, and the absence of travel-related fatigue (such as might be experienced by the visitors). All of those elements, perhaps, play a role. But the salient factor behind home-field advantage is umpire bias, particularly regarding borderline strike-ball calls. University of Chicago behavioral economist Tobias Moskowitz and Sports Illustrated writer L. Jon Wertheim documented convincing evidence in support of that observation in their book Scorecasting.

The authors used pitch-tracking data to quantify this concept.

While QuesTec, and then PITCHf/x, were created in part to better evaluate umpires and create a more uniform strike zone, 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.

Consider the following charts created for a post published earlier this year exploring the nature of home-field advantage. The trend lines remain more or less parallel in the years for which we have data.

outofzonecalledstrike

zonecalledball

What creates the bias? In the case of the home-plate umpire — just as with officials in other sports — it appears to be a desire (consciously or subconsciously) to please the crowd. Moskowitz and Wertheim demonstrate that, the more intense the environment — think Duke basketball playing at home at Cameron Indoor Arena — the more intense the bias for the home team on 50-50 calls.

So the Dodgers’ reward for the producing the regular-season’s best record is a higher probability of ending up on the favorable end of some borderline calls. Perhaps that effect will be enhanced by a raucous Dodger Stadium and its excellent sound system, which blasts sound waves from center field to the stadium’s grandstand, essentially a concrete bowl nestled within a ravine just west of downtown Los Angeles. From my experience covering games there, the stadium has excellent acoustics. It can trap and hold noise, increasing the intensity of key moments.

Jeff wrote before the series that the strike zone would likely favor the Dodgers in part because of their excellent framing tandem of Austin Barnes and Yasmani Grandal. The Astros’ less capable group is led by Brian McCann.

In preparation for this piece, I employed Baseball Savant’s detailed zones, which examine the 50-50 borderline area of the zone, to see if the Dodgers have benefited on the edge of the zone in the series’ first six games. It appears they’ve had a slight advantage as far as balls called and swings offered on borderline pitches.

Consider the following tables, with data from Games 1 through 6.

Astros’ Pitching Success in Borderline zone
Called strikes Called Strike% Balls Ball % Borderline pitches Borderline pitch % Opponent swings Opponent swing %
Home 49 28.5% 48 27.9% 172 35.59% 74 43.0%
Road 38 22.1% 46 26.7% 172 42.5% 86 50.0%

Dodgers’ Pitching Success in Borderline zone
Called strikes Called Strike% Balls Ball % Borderline pitches Borderline pitch % Opponent swings Opponent swing %
Home 34 22.2% 33 21.6% 153 32.9% 85 55.6%
Road 44 25.0% 42 23.9% 176 43.8% 89 50.6%

So how, if at all, should this knowledge influence managerial decision-making and player strategy in the most important of all games? We know that the road pitching staff should generally expect fewer favorable calls. So who could be more adversely affected?

As a team, the Dodgers ranked fourth in baseball in targeting those borderline zones this year (39.0% of total pitches, and 39.5% at home), while the Astros ranked 18th in pitches thrown in those borderline zones (38.4%). That’s a small percentage difference.

Of course, we’re not concerned with aggregate team totals since each manager seems to trust only a few bullpen options beyond his starting pitchers.

We know Lance McCullers and Yu Darvish are starting. We’ll see how long either lasts. Never will a leash be shorter than in a Game 7, for which almost all hands are on deck. In a series throughout which each manager has shown a willingness to bullpen, tandem, and respect the Third Time Through the Order Penalty, it stands to reason that McCullers and Darvish are unlikely to be standing on the mound for the game’s final outs.

The Dodgers should be less concerned with their nibbling pitchers at home. Dave Roberts probably has a clear plan he’d like to follow. I’d guess it’s something like this: get two quality turns through the Astros’ lineup with Darvish and then turn the ball over to Clayton Kershaw and Kenley Jansen, both of whom have excellent control (at least when not fatigued). I’d expect Barnes to be assigned catching duties, given he’s earned the lion’s share of playing time in the postseason and has been the better per-pitch framer this season (and the best in the majors per pitch).

But what about the Astros?

The following list contains the Astros’ pitchers most likely to be used tonight. They’re ranked here by their nibble factor, the percent of pitches thrown to 50-50 zones:

Astros’ Pitchers Ranked by Borderline Pitch %
Player Total Pitches % of pitches
1 Dallas Keuchel 2194 43.16
2 Ken Giles 952 41.39
3 Joe Musgrove 1764 39.51
4 Francisco Liriano 1750 39.37
5 Brad Peacock 2248 39.01
6 Lance McCullers Jr. 2015 38.26
7 Charlie Morton 2391 37.85
8 Will Harris 670 37.46
9 Chris Devenski 1282 36.90
10 Collin McHugh 1080 36.76
11 Luke Gregerson 983 35.10

What we have here, essentially, is the order in which Astros’ pitchers are likely to be penalized by an away game. For instance, the Astros might want to be careful in how they employ Dallas Keuchel.

Consider, for example, Keuchel’s borderline pitch results during Game 1 at L.A:

Kuechel was able to get some calls below the zone and inside against lefties, but there wasn’t a ton of help there.

As I wrote about before the series, no pitcher throws to the borderline area of the lower zone more than Keuchel. Because of that and because of the Dodgers’ right-handed bats and their swing planes, it seemed like L.A. represented a tough matchup for the Houston lefty. That has largely proven to be true.

Keuchel should be particularly challenged on the road against the most disciplined offense in baseball. He recorded the third-lowest zone rate (37.0%) in the majors among pitchers who threw at least 100 innings this season. This is typical for Keuchel, who has lived on the edge as much as any major-league pitcher.

Keuchel has significant splits this year: a .236 wOBA at home versus a .305 mark on the road, and a 13.9-point strikeout- and walk-rate differential (K-BB%) at home versus a 12.9-point mark on the road. Keuchel is also on short rest.

In Game 1, Keuchel wasn’t able to get the inside strike against righties. Keuchel really wanted this pitch, for example:

He didn’t get it, though, and Justin Turner homered later in the at-bat.

While overthinking such decisions can cause paralysis by analysis, an umpire’s inclination to favor the home team is real and has to be considered by a manager. On this stage, the smallest of things can add up. The borderline 1-1 pitch is never more crucial. Umpire bias could play roles in shaping Game 7 and determining a 2017 World Series champion. And the Dodgers should enjoy this edge. After all, they earned it.

We hoped you liked reading Home-Field Advantage and Game 7 by Travis Sawchik!

Please support FanGraphs by becoming a member. We publish thousands of articles a year, host multiple podcasts, and have an ever growing database of baseball stats.

FanGraphs does not have a paywall. With your membership, we can continue to offer the content you've come to rely on and add to our unique baseball coverage.

Support FanGraphs




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.

newest oldest most voted
stever20
Member
Member
stever20

1 thing looking- in the dh era- there’s been 13 World Series gone to 7 games… In those 13 game 7’s been no dh for 8 before tonight. In those 8 games- the NL has won 6 of 8. Last AL team to win a no dh game 7 was Kansas City all the way back in ’85(the last time the World Series was either all dh or non dh). So that would make it seem that it’s a pretty big advantage for the Dodgers tonight.

stever20
Member
Member
stever20

also looking at the dh info- last 10 World Series- so going back to the 2008 World Series…. Been exactly 29 games in both NL and AL parks…
@ NL Parks- NL has a 18-11 advantage
@ AL Parks- AL has a 17-12 advantage

david k
Member
david k

This is interesting, because I’ve heard two narratives as to which league as the advantage regarding the DH. At first glance, you’d think the NL would have an edge because their pitchers have had more “practice” hitting and bunting runners over. However, the delta batting average between NL and AL pitchers may not be all that much, and then factor in pinch hitters will take away a couple of pitcher ABs anyway. Plus, AL teams are built to have a slugging DH, while most NL teams don’t have a comparable player on their bench, so the AL might actually have an advantage overall. But the numbers shown here indicate no clear advantage for one league over the other, small sample size to be sure.