Data analytics play a critical role in an ever-increasing segment of people’s lives, and baseball is no exception. Brought to the public’s attention through Michael Lewis’s 2003 bestselling book, Moneyball, later made into a movie starring Brad Pitt, baseball analytics have been a part of Seton Hall since the late 1980s. For John Saccoman, professor of mathematics and computer science, his interest in both statistics and by playing baseball converged in the 1970s with a regular game of Strat-OMatic, a dice baseball game he took up with his cousin that they still play online to this day. The interest sustained Saccoman for nearly 50 years; he’s taught classes on it and co-authored three books on the subject with Michael Huber and Father Gabriel Costa, an innovator in the field in his own right. Seton Hall editor Pegeen Hopkins spoke with Saccoman recently to learn more.
What is sabermetrics, or baseball analytics?
“Sabermetrics” was coined by Bill James, a well-known baseball analytics person. SABR is the acronym for the Society for American Baseball Research. James’s definition is very simple: the search for objective truth and knowledge about baseball. Baseball analytics is an all-encompassing term that takes into account sabermetrics, but also performance metrics such as spin rate, launch angle for hitting and speed as the ball comes off the bat, for example.
How did Seton Hall get involved in sabermetrics?
In 1988, Father Gabe Costa offered a one-credit course in sabermetrics, the first known credit-bearing course in America on baseball statistics. I was a guest speaker for that class and we team taught it for a while.
What is the field’s appeal?
Baseball analytics can appeal to both the baseball fan and the mathematician. If you know about baseball, you know that if you score a run, it means somebody had to hit the ball so you could advance and score. Analytics, as a mathematical model, tries to divorce performance from context and say, “How many runs would you expect someone to produce for his or her team?” And then, a certain number of runs translates to a certain number of wins for which the player is responsible. As with any mathematical model, the more data you have, the more accurate your model can become. Virtually every Major League Baseball team now has a well-staffed analytics department that produces a report before every game for the manager. It can be used to suggest an optimal batting order or what pitcher to use.
What do people dislike about baseball analytics?
There’s a backlash because it’s different. Statistics should be used within reason, one part of the whole evaluation process.
Now that analytics are widely used, are the advantages that teams had in when using stats to manage diminished?
Except everybody’s got their own spin on it. I started a project that a student, Karl Hendela ’19, did an honors thesis on with a statistic called Wins-Above-Replacement (WAR). WAR allows you to compare players differently: pitchers with other kinds of players or people from different eras. But there’s no agreed upon formula for WAR. With a batting average, everybody knows what the formula is. With WAR, Baseball Reference (a statistics website) has its own standard, FanGraphs (another website) has its own, using different formulas. What we’ve been doing is to devise our way that closely aligns with the others. We called it SHU-WAR.
What do you hope students or others interested in analytics will take away from it?
You don’t necessarily have to love baseball, but it helps. These things are extendable to other fields, as well.
