Basketball Data Science by Zuccolotto Paola; Manisera Marica;

Author:Zuccolotto, Paola; Manisera, Marica;
Language: eng
Format: epub
Publisher: CRC Press LLC
Published: 2019-12-23T00:00:00+00:00

Very similar remarks, although less detailed since the growth of the tree is less deep, can be drawn from the second CART, obtained with the “Rio 2016” dataset.

According to the evidence identified by the CART models, shots are not all alike. For this reason, the authors used the trees to develop new shooting performance measures, taking into account the circumstances under which each shot has been attempted. For example, a 2-point shot attempted in the last 2 seconds of the shot clock has a scoring probability of approximately 40%, unlike a shot attempted in the first 7 seconds, which has a scoring probability greater than 65%: a shooting performance measure should take into account this evidence and give higher merit to a basket made when the scoring probability is lower. For each shot type T (2P: 2-point, 3P: 3-point, FT: free-throw), let JT be the set of attempted shots of type T and xij the indicator assuming value 1 if the jth shot of the ith player scored a basket and 0 otherwise. The new shooting performance of player i for shot type T is given by

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.