Prediction of the margin of victory only from team rankings for regular season games in NCAA men's basketball

摘要

The main objective of this paper is to investigate the extent to which themargin of victory can be predicted solely by the rankings of the opposing teamsin NCAA Division I men's basketball games. Several past studies have modeledthis relationship for the games played during the March Madness tournament, andthis work aims at verifying if the models advocated in these papers stillperform well for regular season games. Indeed, most previous articles haveshown that a simple quadratic regression model provides fairly accuratepredictions of the margin of victory when team rankings only range from 1 to16. Does that still hold true when team rankings can go as high as 351? Do themodel assumptions hold? Can we find semi- or non-parametric methods that yieldeven better results (i.e. predicted margins of victory that more closelyresemble actual results)? The analyses presented in this paper suggest that theanswer is "yes" on all three counts!