With the conclusion of another college basketball season on Monday, a fresh wave of data from the NCAA fills the archives. Of the numerous stat lines and trends, one in particular stands out for Princeton: attendance.
A priori, it has been well known that attendance at Ivy League basketball games is smaller than that for the nation at large. Just this past year, the average Division I men’s basketball game tallied over 4,700 attendees. By comparison, the average Ivy League game in the same year barely eclipsed 1,700.
Furthermore, it is also widely recognized that the men’s teams generally attract larger crowds than the women’s teams at every level of competition. For Princeton, the crowd at men’s basketball games frequently doubles, even triples, that for women’s games.
Yet beyond these broad generalizations lies an even deeper trend: a closer look at attendance rates to women’s games — over the span of a decade — shows attendance is a function of the team’s record while the men’s team, on the other hand, draws a consistent showing regardless of performance.
Starting on the men’s side, we analyze the correlation between attendance and the team’s winning percentage over the past decade, where attendance is defined as the average number of attendees per home game, as maintained by the University’s official athletics department. The findings show that, for the men’s varsity team, attendance has been largely independent of success over the past decade. Specifically, the correlation coefficient was under 0.01, suggesting little to no relationship.
For the women’s team, however, the analogous data shows a correlation coefficient of 0.572. As a stark example, we can look at the 2009 and 2010 seasons. In 2009, the women’s basketball team balanced a very respectable 50% winning percentage. But, the following year, the Tigers went undefeated in the Ivy League and advanced to the first round of the NCAA Tournament, during which the team averaged over 2,500 attendees per home game, a mark that was nearly equal to the average for the men’s games.
Before translating the correlation between winning and larger audiences into a causal claim, we will analyze a few potential confounding variables. First we observe that attendance at women’s basketball games is in fact not a function of time. Since 2006, the average attendance has increased and then decreased on an annual basis.
Furthermore, average attendance to women’s basketball games has remained amazingly stagnant in the past decade, suggesting that large variations in attendance are unusual and in need of explanation. In fact, since 2006, the national average at women’s basketball games has been 1,569 with a standard deviation of just 33, a minuscule fraction.
To explain such grand variation in attendance at women’s basketball games in Jadwin Gymnasium, we are left to believe that success remains a crucial draw for attendees. On the other hand, regardless of the season, attendance at men’s games has been remarkably stable. While the underlying economic and social causes for this correlation can only be inferred, basketball fanatics should think carefully about which games they attend in the future.
