Mastering the Numbers Game – Sports Illustrated Coverage
-
June 14, 2013
America’s favorite pastime, Major League Baseball, is well underway for the 2013 season with teams chasing the coveted pennant. Many University of Texas at Austin Petroleum and Geosystems Engineering (UT PGE) students and alumni are tuning into the games to watch the action unfold, but UT PGE assistant professor, Eric Bickel, is actually guiding the players’ success on the field through strategic number crunching.
Bickel’s baseball research started in graduate school at Stanford University in 1995 when he began working with the baseball team, creating data mining software, ChartMine®. The software helped the Cardinals collect data on every pitch thrown in a game, guiding the approach to batters from opposing teams – specifically, Stanford’s rival team: the University of Southern California. Soon, the software was purchased by 300 schools across the country with about 30 percent of Division I schools using it, including UT Austin.
After graduating from Stanford, Bickel realized the decision-making aspect of the game via baseball coaches and players was not quite accurate. He began writing papers to explain players and coaches’ mistakes and how to correct the errors, increasing their victory margins.
“One paper I wrote was how to act on different pitch counts,” Bickel said. “Sometimes the batter will just let a pitch go by on purpose. If it’s three balls, no strikes, a lot of times the coach will say, ‘Don’t swing at the pitch, no matter what.’”
Bickel said most people don’t understand why a batter would take a pitch on a 3-0 count, because the likelihood of a fastball down the middle is more than 90 percent. However, Bickel’s research proves that taking a pitch on a 3-0 count increases the likelihood of a batter getting on base.
“About 38 percent of all batters eventually get on base,” Bickel said. “At 3-0, 77 percent of batters eventually get on base. Suppose you’re sitting there with a 3-0 count. If you let the pitch go by, and the pitcher throws a strike, you’re down to a 63 percent chance of getting on. If you instead put that ball in play, you only have a one-third chance of getting on base. Your choice is to put the ball in play and have a one-third chance of getting on base, or take a strike and still have a 63 percent chance of getting on base. That’s why you take it.”
Pointing out this mistake has led to Bickel becoming a household name in the baseball world. His paper gained so much attention that Evan Gattis, Atlanta Braves catcher and early front-runner for the National League Rookie of the Year Award, even mentioned his work in a June issue of Sports Illustrated.
“The Sports Illustrated coverage came out of the blue,” Bickel said. “[Gattis] emailed me sometime in January 2012 thanking me for one of the papers I had written about specific batting averages by pitch count. He told me the paper was extremely helpful and thanked me for writing it. He’s just the nicest guy. I talked with him a bit, asking him about the paper and where he was playing. He was in the minor leagues for the Braves, at the time. That was the only conversation we had, and then a friend of mine called me to tell me about the article.”
Like Gattis, many of Bickel’s students in his graduate decision making class resonate with baseball as an example of the decision making process. Analyzing baseball decisions is not as far off from engineering as it seems, Bickel said.
“Engineers are really good at breaking down complicated issues into smaller parts and putting it all together,” Bickel said. “Engineering is just an approach to thinking and decision making. Petroleum engineers, specifically, are making high risk decisions on a daily basis impacting our energy resources.”
