Take Me Out to the Ball Game (We Can Go by Taxi) | NYU Tandon School of Engineering

Take Me Out to the Ball Game (We Can Go by Taxi)


In 1910 a sportswriter named Hugh Fullerton and Johnny Evers, the second baseman for the Chicago Cubs, collaborated on a book they called Touching Second: The Science of Baseball, which promised to examine how the game developed into an “exact mathematical sport.” “As a problem in geometry,” they wrote, “baseball, in all of its departments, may be reduced to exact figures."

The pair would no doubt be amazed and impressed by the work of Claudio Silva, a professor of computer science and engineering and the head of disciplines for the Center for Urban Science and Progress, who has taken that assertion to a whole other level. Along with independent visualization researcher Carlos Dietrich, Silva has developed a visual analytics system dubbed Baseball4D, which provides for the first time the ability to analyze each and every play on the field and allows both fans and industry officials to answer previously unanswerable analytics questions: are the best fielders really those who make dramatic and graceful-looking catches or are seemingly mundane players actually watching the ball and anticipating its trajectory better? Where should an infielder be positioned to best catch a ball traveling at a particular velocity?

Dietrich and Silva have been collaborating with MLB Advanced Media on the development of the system--which involves groups of high-performance cameras installed throughout the ballpark along with software that produces highly interactive visualizations of the game in unprecedented detail. It is expected to be of particular benefit to coaches and scouts, who can use the new data stream to decide which players to recruit and how best to leverage their abilities. It is also being seen as a possible boon to amateur sabermetricians--those dedicated fans who delight in endlessly analyzing every aspect of gameplay. (The term sabremetrics is derived from the acronym SABR—short for the Society for American Baseball Research, a group founded in the early 1970s whose statistics-crunching approach to the game was popularized by sportswriter Bill James.)

The system is now being deployed at Miller Park in Milwaukee, Target Field in Minnesota, and Citi Field in New York, with plans for it to be operational in every major-league ballpark in the nation by Opening Day in 2015.

Silva’s paper, “Baseball4D: A Tool for Baseball Game Reconstruction & Visualization,” will be presented in November at the IEEE (Institute of Electrical and Electronics Engineers) VIS 2014 conference, which will bring together the world’s top researchers and practitioners in the field of visualization.

Baseball4D is just one of the exciting projects in which Silva is involved. Another is TaxiVis, which focuses on another fascinating data set: taxi trips, which he asserts can serve as valuable sensors. The information associated with taxi trips, Silva and his co-researchers show, can provide unprecedented insight into many different aspects of city life, from economic activity and human behavior to mobility patterns.