Statistics and story: baseball's two languages
George Herman “Babe” Ruth is a baseball legend. He’s the “Great Bambino.” The “Sultan of Swat.”
From 1914-1935, Ruth consistently crushed the ball, breaking nearly every standing batting record and tallying a career 714 home runs. By many accounts, the Boston Red Sox star pitcher turned New York Yankee slugger is still considered the greatest baseball player of all time.
But, should he be?
Since the first pitch of Major League Baseball, fans have used two languages to talk about the game -- statistics and story. However, as long as the two languages have coexisted, they have been at odds when speaking of the greats.
Is it possible to compare Ruth’s performance between the foul lines to contemporary players?
Lifelong baseball fan and professor within the Department of Statistics Daniel Eck noticed, generally, baseball statistics seem to place higher values on the performances of players from past eras, specifically pre-integration. At that time the talent pool was smaller with fewer having access to the league. To right this, Eck and graduate student Shen Yan set to work on an era-adjusted statistical method to compare the greats.
“I feel like I know baseball really well, but I didn’t know what this meant in greater context,” Eck explained of his initial findings. He connected with Adrian Burgos, professor with the Department of History, whose research focuses on US Latino history, sport history, and African American history. From their interdisciplinary approach, they’ve reexamined the baseball greats by considering their impact on the game through the lens of who, when, and how they played.
“What our study allows us to do is to reset these people, assess their performance, and gain a new appreciation for the circumstances they played in,” Eck said.
The Full House Model assigns players with a statistic “through a principled balancing of how well a player performed ‘versus their peers’ and the size of the MLB eligible population,” their website says. “Under this model, great all-time statistics requires that an MLB player is both better than their peers and played during a time in which the MLB eligible population is large. In this way, the model constructs an even playing field that extends across eras.”
Their Full House Model uses theories from a book of the same name written by Stephen Jay Gould.
“Most importantly, we find that the greatest modern players, including several top African American and Latino players, now sit atop the greatest all-time lists of historical baseball players while conventional wisdom ranks such players lower,” they write in their findings.
The Illinois researchers are using a baseball statistic known as Wins Above Replacement, or WAR, to base their framework. This stat is relatively new in the age of baseball. It measures a player's value to his team in all areas of the game by indicating how many additional wins he is worth over a replacement-level player, according to MLB.com. Various resources calculate WAR slightly differently; however, using their model with different WAR baselines, their reexamined top baseball players remain consistent with Willie Mays, Barry Bonds, Roger Clemons, Ruth, Henry “Hank” Aaron, and Alex Rodriguez in the mix.
“(This model) is going to change the rankings of who the greatest players are, and this is a debate that happens every single day,” Eck said.
“Baseball has two languages – the written words and stats,” Burgos explained. “What we were able to do was bring our expertise together in narrative form, but also understand the stats the way many individuals seek to understand the game, by looking at the history behind the production of the stats and the play that produced the stats. Within the baseball historian community, oftentimes there is this difficulty bridging between the historical analyses and the statistical analyses. We have, with great intention, sought to bridge that with this project.”
For Eck and Burgos, this project goes beyond baseball by considering the eras of wartime, segregation, and early integration using interdisciplinary tools that help in studying the past.
“We had to make an adjustment for that historical reality,” Burgos said. “To me that’s what makes this project worthwhile. That’s one of the beauties of scholarship. It can give us tools to examine those things and view things differently. I’m constantly telling my students, when you find new evidence; it doesn’t mean the past has changed, it means how we understand the past has changed. If we gain new perspectives and outlooks and tools, we understand the past differently and have a new appreciation for it.”