Can Machine-Learning Methods Predict the Outcome of an NBA Game?

10 Pages Posted: 25 Jul 2018

Date Written: March 1, 2018

Abstract

Game outcome prediction is a well known problem to predict, yet there are aspects, features, and advantages that aren’t taken into account in most modules. In this paper, we investigate various features and aspects of a prediction outcome for a team vs team, this includes basic known features such as home court advantage, number of days between games and winning streaks as well as less conventional features such as player similarity score, gambling odds, team specifications. The analysis is based on descriptive statistics and while it is mainly restricted to the National Basketball Association, it can be projected towards other sports as well.

Keywords: NBA, Machine Learning, Game Prediction, Basketball, Sports, Regression, Classification

Suggested Citation

Lieder, Nachi, Can Machine-Learning Methods Predict the Outcome of an NBA Game? (March 1, 2018). Available at SSRN: https://ssrn.com/abstract=3208101 or http://dx.doi.org/10.2139/ssrn.3208101

Nachi Lieder (Contact Author)

Independent ( email )

No Address Available

Register to save articles to
your library

Register

Paper statistics

Downloads
264
Abstract Views
1,160
rank
114,777
PlumX Metrics