Presenter: Nathan Wiggins
Authors: Nathan Wiggins
Faculty Advisor: Joshua Price
Institution: Southern Utah University
Every year, millions of people fill out brackets for the NCAA March Madness Tournament. Strategies for picking a potential winning bracket range from strenuously analyzing regular season statistics to comparing college mascots. These strategies naturally make the curious mind consider whether one comparison in particular is effective at predicting which teams will do well in the tournament. In particular, this paper investigates whether or not the number of rebounds per game that a team averages has a measurable impact on how far a basketball team progresses in the NCAA College Basketball Tournament. Using data from the 2013 – 2019 College Basketball Tournaments, we construct a series of linear probability models using the ordinary least squares method to analyze this impact and find that each additional rebound per game averaged by a team results in a 2.62 percentage point increase in the probability that the team will progress to the Round of 32 and a 1.99 percentage point increase in the probability that the team will progress to the Sweet 16 when controlling for other performance metrics. Further, we fail to find that rebounding has a measurable impact on the probability that a team will progress past the Sweet 16. These results are rigorously tested for robustness and applications of the results are discussed.