Everybody knows MLB umpiring needs an overhaul, right? As usual, Bud Selig has been slow to adopt measures common in other professional (and even kids') leagues, this time with video replay. And with a couple of high-profile gaffes -- sorry, that means you, Jim Joyce -- putting the umps under even more scrutiny, there's been something of a groundswell to modernize how MLB calls the games.
So the crack researchers at ESPN looked back at a sample of two weeks worth of games to figure out just how many calls the umps miss. Their verdict? That the umpires get roughly 20 percent of close calls wrong. Pretty damning, no? Well, not exactly.
As baseball stats guru Nate Silver points out, the key question here is what constitutes a "close call". ESPN's researchers looked at a total of 184 games and found 47 clearly missed calls. On a per-game basis that's actually not that bad -- one bad/missed call every four games.
But that doesn't fit ESPN's narrative of an epidemic of poor umpiring marring the sport. So instead of reporting that MLB's umpires mess up a couple of times a week, they restricted their analysis to "close calls" -- an obviously subjective category they never define. And voilá: suddenly there's a veritable crisis of officiating, just like everybody always knew.
Of course, there's a compelling argument that mistakes simply shouldn't happen in baseball. Unlike football or basketball, there aren't real judgment calls on fouls. Either someone was out or safe; it was foul or fair. Video review could easily solve most of those problems. But you don't need to misleadingly report your sample to make that case, as ESPN did.
Which just reminds us, there are three types of lies: lies, damned lies, and statistics.