There’s been a big brouhaha lately about home field advantage in sports. This has been caused by the publication of a new book, Scorecasting: The Hidden Influences Behind How Sports Are Played and Games Are Won, in which the authors claim that biased umpiring accounts for most of a team’s home field advantage. I believe they make this claim across virtually all sports, though they used PITCHf/x data to study the phenomenon in baseball.
A lot of people are jumping on this assertion. Phil Birnbaum has tackled the subject on his blog, and J-Doug is also writing about it at Beyond the Boxscore. I’d like to throw my towel into this particular ring, because John Walsh covered this subject in detail in the 2011 Hardball Times Baseball Annual.
John gathered two years worth of PITCHf/x data (2008-2009), about 1.5 million pitches worth, and looked to see if ball-strike calling is biased toward the home team. His approach was generally similar to his original work on umpires and strike zones, The compassionate umpire. John did find a bias toward the home team, an average of a little less than one extra ball called in their favor, per game.
Doesn’t sound like much, right? Think again. If you apply the run value count of each strike and ball to the home field bias, well, I’ll let John tell the story:
…it turns out that the value to the batter of miscalling a strike is 0.17 runs, averaging over all ball-strike counts. So in our hypothetical scenario, the home team is scoring 0.17 runs per game more than the visiting team. We can now apply the Pythagorean rule to convert runs scored and allowed into a winning percentage. If I assume a typical value of runs scored of 4.5, I get a home winning percentage of .519. Actually, though, I found the umpire bias to be 0.8 balls per game (not one, as I’ve been using in this example), or 0.14 runs, for a winning percentage of .515.
So, yeah, umpire bias in favor of the home team is pretty small, but it accounts for a little more than a third of the home-field advantage, which is not so small.
John also studies other potential umpire bias, such as whether the speed of the fastball affects whether or not it might be called a strike. The answer, it does.
And that’s just one article in the Hardball Times Annual. I don’t mean to complain, but it does bug me when book publishers and writers make big claims to sell books, but ignore the excellent research that has been performed by hard-core baseball analysts such as John, who has a PhD in Physics and has been a leading expert in baseball analysis for a while.