Data analytics spread consistently over the years for answering several questions in all the fields, included sports. The aim of this work is to analyze the offensive performance of NBA players in terms of the amount of minutes taken to exceed a given point threshold during the post All-Star season segment. The final goal is to perform variable selection, identifying which are the main players` achievements that significantly impact the outcome. Survival analysis methods, in particular Cox regression and LASSO Cox have been used. Results suggest that attempting a higher number of two- and three-point shots, having been selected for the All-Stars game and gaining more double doubles increase the probability of exceeding the threshold and of doing it in a shorter time. Another interesting result regards the number of steals, the only variable selected by the models related to defense and game construction, which was negatively associated to the outcome.
© Copyright 2023 13th World Congress of Performance Analysis of Sport and 13th International Symposium on Computer Science in Sport. IACSS&ISPAS 2022. Veröffentlicht von Springer. Alle Rechte vorbehalten.