Application of Linear Regression in Basketball Analytics: Illustrated by Two Case Studies

As sports have evolved over the last hundred years, scientific approaches have proven essential for studying athletic phenomena. Linear Regression, a method reveals relationship between dependent variable and independent variables, is favored by statisticians by its prediction power and computational efficiency. This article explores the application of linear regression as a scientific method in basketball analytics by separately examining key factors influencing NBA players' skill levels and salaries. Among independent variables such as duration of playtime, steals per game, efficiency rating and professional age in correspondence of dependent variables, the skill levels and salaries of players are selected based on data provided by American Basketball League. To investigate this further, the author constructed regression models with R program to assess significant factors affecting two dependent variables respectively. Moreover, visualization charts and plots were made to ensure the validity of the regression models. Lastly, results indicated that determinants like duration of playtime, professional age showed low p-value and VIF (variance inflation factors) implying significance and no risk of multicollinearity.