EXPLORING GAME PERFORMANCE IN NBA PLAYOFFS
This study aimed to identify performance profiles in NBA playoffs and describe performances in the first and the last games of the series. Thirty games from the official box-scores and player-tracking data of the 2014/15 NBA playoffs were considered. A k-means cluster analysis was performed to group the players according to their game performance profiles and a discriminant analysis was conducted to identify the game-related statistics that best discriminated the groups. The first function correctly classified 64.2% of the cases and the second function classified 29.9% of the cases. The cluster analysis identified four different performance profiles and from the discriminant analysis emerged several offensive and defensive variables to classify the players in the clusters. The identified trends help to improve understanding of the game during different stages of playoffs. Coaches may use this information to better prepare their teams for different game scenarios.
Key words: game statistics, players profile, discriminant analysis
Copyright (c) 2018 Kinesiology
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
At Faculty of Kinesiology we recognize that access to quality research is vital to the scientific community and beyond. Kinesiology is non-profit journal and all costs of publishing and peer review process are covered by the publisher itself or other funding sources like Ministry of Science and Education of the Republic of Croatia. Full text papers are also available free of charge at http://hrcak.srce.hr/kineziologija. There are no restrictions on self archiving of any form of paper (preprint, postprint and publisher's version).
Articles are distributed under the terms of the CC BY - NC 4.0
Kinesiology does not charge any fees to authors to submit or publish articles in our journal.