Slashdot Mirror
A Computer Umpires Its First Pro Baseball Game
An anonymous reader writes: Baseball has long been regarded as a "game of inches." Among the major professional sports it arguably requires the greatest amount of precision — a few extra RPMs can turn a decent curveball into an unhittable one, and a single degree's difference in the arc of a bat swing can change a lazy popup into a home run. As sensor technology has improved, it's been odd to see how pro baseball leagues have made great efforts to keep it away from the sport. Even if you aren't a fan of the game, you're probably familiar with the cultural meme of an umpire blowing a key call and altering the course of the game.
Thus, it's significant that for the first time ever, sensors and a computer have called balls and strikes for a professional game. In a minor league game between the San Rafael Pacifics and the Vallejo Admirals, a three-camera system tracked the baseball's exact position as it crossed home plate, and a computer judged whether it was in the strike zone or not. The game went without incident, and it provided valuable data in a real-life example. The pitch-tracking system still has bugs to work out, though. Dan Brooks, founder of a site that tracks ball/strike accuracy for real umpires, said that for the new system to be implemented permanently, fans must be "willing to accept a much smaller amount of inexplicable error in exchange for a larger amount of explicable error."
Thus, it's significant that for the first time ever, sensors and a computer have called balls and strikes for a professional game. In a minor league game between the San Rafael Pacifics and the Vallejo Admirals, a three-camera system tracked the baseball's exact position as it crossed home plate, and a computer judged whether it was in the strike zone or not. The game went without incident, and it provided valuable data in a real-life example. The pitch-tracking system still has bugs to work out, though. Dan Brooks, founder of a site that tracks ball/strike accuracy for real umpires, said that for the new system to be implemented permanently, fans must be "willing to accept a much smaller amount of inexplicable error in exchange for a larger amount of explicable error."
The MLB rule book expressly prohibits arguing balls and strikes, partly because it's so hard for the home plate umpire to make those calls. For the most part, MLB umpires do a really good job of getting calls right. It's really hard to determine accurately whether the ball crossed within a certain rectangle at any point as it crosses home plate. The strike zone is a three dimensional construct that varies from one hitter to another depending on their height. There's nothing easy about what the umpires are asked to do. For the most part, MLB players tolerate a strike zone that's not perfect provided it's not unreasonably large or small and that it's consistent for both teams throughout the game. That said, it's not perfect, and there are situations where this really could make a difference.
Yadier Molina is one of the best catchers in baseball, known primarily for his defense, especially picking runners off base, and for being a good hitter as catchers go. He also does a good job of calling the game for his pitcher, that is what pitch he asks the pitcher to throw, and where he sets up and tries to get the pitcher to locate the pitch. He's probably the best catcher in MLB right now at this. But he's also really good at fooling the umpire in what's known as framing the pitch. That is, how he moves the glove after catching the baseball can fool the home plate umpire into calling borderline balls as strikes and effectively expanding the strike zone for his pitchers. Long considered a myth, there's actually real evidence now with new data that shows catchers like Molina can have large impacts on a game by framing pitches. Not directly related to Molina, but a similar issue, is when a pitcher badly misses the catcher's target, but the pitch is still a strike. An example would be the catcher setting up on the outside corner but the actual pitch being on the insider corner. The pitch may well be a strike, but because the pitcher missed the target so badly, it will almost always be called a ball.
Greg Maddux was a really great pitcher who was successful in fooling hitters and getting a lot of strikeouts with a fastball around 90 mph. He's a Hall of Fame pitcher and deservedly so. He fooled hitters for about two decades and his stats in the shortened seasons of 1994 and 1995 rank among the very best stats for a season in the history of baseball, right up with Bob Gibson's 1968 season and Pedro Martinez's ridiculously good 2000 season. However, in addition to pinpoint control of his pitches and good movement, he also seemed to get a lot of pitches called strikes that were off the plate. That is, he got some strikes "on the corner" that were really about six inches off the plate. His teammate and fellow Hall of Fame pitcher Tom Glavine, was also frequently associated with getting a wide strike zone.
While the umpires generally do a really good job, I can see ways where this really can help standardize the strike zone and solve some problems. I don't think this will ever replace human umpires altogether, but it frees them up to focus on other things like managing the pace of play (a huge problem for baseball), looking for balks, and determining whether a hitter went around on a checked swing. And I suspect that in college and lower levels of baseball, human umpires will always call balls and strikes.