Ten things I didn’t know last week

Interleague home field advantage

Did you notice anything weird about the intracity interleague games? In Chicago, the Cubs won all three games in Wrigley and the White Sox won all three games in Comiskey. That’s an extreme home field advantage, even though both parks are in Chicago. If you just listened to the games, you sometimes couldn’t tell who had more fans in the stands. But the park apparently had an impact (I know, I know. Small sample size).

Other intracity splits also favored the “home” team, but not quite as extremely. In New York, the Mets were 3-0 in Shea and the Yankees were 2-1 in Yankee Stadium. In Anaheim of Los Angeles, the Angels and Dodgers were both 2-1 in their home parks. In these three sets of intracity games, the “home team” was 15-3. Same teams, same cities. Same home cooking.

The home field advantage didn’t play out for the rest of the National League. Overall, they were 55-70 at home in interleague games; 47-79 in AL parks. That’s different, but not extremely different. Not like those intracity games.

EDIT: Big Oops. As several readers have pointed out, I got my New York stadiums (stadia?) mixed up. The Mets were 3-0 at Yankee Stadium, and the Yankees were 2-1 at Shea. So … never mind.

What the analysts say

Beyond interleague games, teams have been strong at home this year. In baseball, home teams usually win about 54 percent of the time. This year, home teams have won 56.4 percent of the time (through Tuesday’s games). That doesn’t sound like a big difference, but MGL (one of the more conservative interpreters of current statistics) felt that it was a big enough difference to wonder out loud whether home teams are cheating this year.

At Baseball Prospectus, Nate Silver is wondering about home field advantage, too (subscription required). As usual, Nate’s article is thoughtful and insightful; he outlines seven variables that might contribute to home field advantage:
{exp:list_maker}Asymmetry
Hitter/Pitcher Balance
Fences and Caroms
Weather
Domes
Attendance and Atmosphere
Travel Distance {/exp:list_maker}In the article, he rated each home park within each category and found that a dome has had the most impact on home field performance over the past four years. I think more detailed analysis is forthcoming.

Historical home field advantage

So let’s add some perspective here. This is the home field advantage (HFA) of major league teams for each of the last umpteen decades:

 Decade      HFA
  1870      .519
  1880      .572
  1890      .598
  1900      .552
  1910      .540
  1920      .542
  1930      .552
  1940      .544
  1950      .539
  1960      .540
  1970      .538
  1980      .540
  1990      .535
  2000      .540

See why we consider 1900 the beginning of “modern” major league baseball? Things were just crazy before then. A .600 home field advantage in the 1890s? Crazy.

Since 1910, the single highest years in total home field advantage have been:
{exp:list_maker}1931: .582
1978: .573
1930: .571
1945: .567
1926: .565
1918: .565 {/exp:list_maker}Those are the only years in which the total home field advantage has been higher than the 56.4 percent we’ve got at the midpoint of this year. This year’s mark is high, but it wouldn’t be the highest historically. And we’ve still got half a year to go.

We’ll take a closer look at 1978 in a minute.

Umpire bias

By the way, that radical home field advantage in 1890 reminds me of Matt Souders’ (otherwise known as SABRMatt) SABR presentation last Saturday. Matt did a whole bunch of complicated math on historical baseball stats and found that umpires were indeed very biased in their decision-making in the 1800s. In fact, the “umpire factor” was as important as the “home park factor” in those early days.

Since the 1900s, umpiring has been consistently a (relative) nonissue. The weird thing is that the bias trended somewhat up (in other words, umpires have become more biased) throughout much of the last century, only to decline a bit since Questec was introduced. I can’t vouch for the math, but the results were thought-provoking.

Phil Birnbaum also presented his debunking of the umpire racism study (summarized by John Beamer in this post) at the SABR conference. Point taken: Be careful when you use numbers to “accuse” anyone of bias.

By the way, Phil has also posted the latest clutch hitting study by Dick Cramer and Pete Palmer (also presented at SABR), along with a quick review of clutch hitting studies and some relevant links. Good reading, unless you’re soooo tired of clutch hitting studies.

All-time highest home field advantage

Back to home field advantage. Here’s a list of the 10 teams that have had the highest home field advantage in a single year (after 1910). To rank the teams, I subtracted each team’s “away” winning percentage from its “home” winning percentage.*

  Year  HomeTm      Home%    Away%   Diff
  1945  PHA         .527     .171   .356
  1949  BOS         .792     .455   .338
  1996  COL         .679     .346   .333
  1987  MIN         .691     .358   .333
  1978  HOU         .617     .296   .321
  1952  BOS         .649     .338   .312
  1933  CHN         .709     .400   .309
  1949  PHA         .675     .377   .299
  2003  COL         .605     .309   .296
  1961  BOS         .617     .321   .296

*This probably isn’t quite the right way to rank home field advantage. When a team has an extreme winning percentage (positive or negative), it’s harder to make it more extreme. In other words, it would be harder for the Phillies to perform worse than their .171 “away” percentage in 1945 than it would be for the 1949 Red Sox to perform worse than their .455. However, I couldn’t think of a way to do it, and I’m not convinced I really need to. And I’m lazy.

Several teams appear on the list more than once. The Red Sox are on it three separate times, the Rockies and Philadelphia Athletics twice. Plus, two appearances by dome teams: the Astros and Twins.

You probably understand that Fenway and Coors are unique and idiosyncratic fields. The A’s played in Philadelphia’s Shibe Park from 1909 to 1954, and I don’t know why their home record reached record levels in the 1940s. As far as I can tell, there were no major changes to the ballpark during the time. It may have had more to do with the composition of the team those years. Or it may have just been a fluke.

The 1978 Royal team was one of the drivers of the general 1978 peak, but they weren’t alone. Five other teams also notched a home/away difference of .200 or more (the Pirates, Royals, Red Sox (again, the Red Sox!), Phillies and Indians). Also, there wasn’t a single team with a negative home/away difference in 1978. That has happened only about a dozen times since 1900.

How about the worst yearly home teams of all time? Here’s the same top 10 list, in reverse:

  Year  Team     Home%   Away%   Diff
  1994  CHN       .339    .537  -.198
  1998  KCA       .363    .531  -.168
  1948  BRO       .468    .623  -.156
  1948  PHA       .468    .623  -.156
  1980  BOS       .444    .595  -.150
  1981  KCA       .404    .554  -.149
  2001  CIN       .333    .481  -.148
  1928  SLN       .545    .688  -.143
  1923  BSN       .286    .416  -.130
  1912  CHA       .442    .571  -.130

Well, looky there. The 1948 Philadelphia Athletics have the fourth-worst home record of all time. They just couldn’t make up their minds, could they? And the Kansas City Royals make the list twice. Don’t forget, 1981 was a strike-shortened year.

By the way, the Rockies are the all-time champion in winning at home. Counting all years, they have been a .547 team at home and .395 team away. They’re followed by the Astros, Philadelphia Athletics, Marlins, Red Sox and Tampa Bay. The worst “home team” of all time? The Orioles, only .543 at home vs. .493 on the road.

Bonus Graph

So here’s a fancy graph of each team’s annual home field advantage over the decades:

It’s a terribly nerdy thing called a “box whisker” graph, but I like the way it looks. For each decade you can see the spread of each team’s home field advantage. The blue line in the middle is the median home field winning percentage of each decade (calculated as the difference between each team’s home winning percentage and away winning percentage). The red boxes cover the inner quartiles, the vertically extended lines cover the outer quartiles, and the triangles are outliers.

Most of the teams I listed above are triangles on the graph.

If you spot something meaningful on the graph or in any of the data, leave me a note on Ballhype.

K.C. and the Sunshine Band inspired Disco Demolition Night

One of the highlights of the SABR conference was the Q&A session by Mike Veeck and Mark Shapiro. Veeck was the marketing genius behind Disco Demolition Night, one of the most notorious ballpark events ever. He tells a great story about the event, including a long buildup, a crowning finish and an afterword (“My Dad was proud”).

My favorite part of the story is that Veeck was inspired when a car drove by him blasting “That’s the Way I Like It” by K.C. and the Sunshine Band. He thought: “Haven’t we had enough of this music?” (or something like that). The rest is infamy.

I always hated that song.

The biggest home field advantages this year

Where were we? Oh yeah. Now that we’ve reviewed the history of home field advantage, let’s take a look at this year’s biggest home achievers.
{exp:list_maker}Atlanta: 28-14 at home (.667) and 12-29 on the road (.293). A difference of .374. Note that if the Braves keep up this pace, they’ll set the record for largest home field advantage since 1900.
Chicago Cubs: 33-10 at home (.767) and 17-23 on the road (.425). A difference of .342. At this pace, they’d finish second to the record.
Boston: 31-10 at home (.756) and 18-24 on the road (.429) for a difference of .328. That would be the sixth-highest difference since 1900. {/exp:list_maker}These are the three teams driving this year’s pace. Why these three teams? Well, the Red Sox’ Green Monster is apparently once again rearing its big green frontispiece. And the Cubs seem to have their home mojo working this year. Could be the fickle winds of fate. But the Braves? I have no idea. The difference is in the offense: The Braves have scored 5.2 runs per game at home and 3.8 runs per game on the road.

The odd thing is that the Atlanta Braves are the second-worst “home team” of all time (behind the Orioles) with a .544 record at home and .487 on the road.

There’s one other interesting team home/away split this year:
{exp:list_maker}San Francisco: 14-25 at home (.359) and 22-22 on the road (.500) for a difference of -.141. {/exp:list_maker}That would be one of the top 10 worst home/away splits of the past 100 years.

I expect all of these teams to even out over the rest of the year. But I’ll be interested to see what Nate, MGL and others have to report on the subject.

What happened to the Mitchell 89

Another one of the SABR presentations that rates a mention here was called Did Steroid Use Enhance the Performance of the Mitchell 89? The Effect of Performance-Enhancing Drugs on Offensive Performance from 1995–2007. The presenters were a bunch of academics in biostatistics, and their paper is going to appear in the Journal of Quantitative Analysis in Sports.

In a nutshell, the authors found that batters who took steroids increased their production by 7-12 percent during the years the Mitchell Report found that they took steroids. However, those who took HGH had a negligible change in their production. The authors used Runs Created per 27 outs to determine the change.

I always find baseball research by academics to be a bit frustrating, and this presentation was no exception. Folks like this tend to be good at statistics (biostatistics is a hot field these days) but not as insightful about specific baseball statistics. In this case, I wonder why the authors would use a statistic as “gross” as Runs Created when a component analysis would be a lot more helpful.

Although limited, the study was worth noting. I hope others build upon it.

Stupid Rangers

I mentioned last week that the Rangers were on a roll of “consistent averageness.” I defined consistent averageness as staying within one full game of .500 for consecutive days. The record is 56 games, and I was rooting for Texas to break it.

Unfortunately, they won three consecutive games the last three days (including two against the Yankees) and now stand a game and a half above .500. What’s worse, I miscounted last week: The Rangers managed only 36 straight days of averageness in the end. The major league record remains 56.