# NBA2Vec: Dense feature representations of NBA players

## TL;DR

NBA2Vec introduces a neural network model that creates dense, meaningful representations of NBA players by predicting game outcomes, enabling better understanding of player roles and potential strategic applications.

## Contribution

The paper presents NBA2Vec, a novel Word2Vec-based neural network that generates player embeddings from play-by-play data without relying on hand-crafted features.

## Key findings

- Achieved 0.3 K-L divergence with empirical play distribution
- Embeddings align with player positions and styles
- Accurately predicts playoff game outcomes

## Abstract

Understanding a player's performance in a basketball game requires an evaluation of the player in the context of their teammates and the opposing lineup. Here, we present NBA2Vec, a neural network model based on Word2Vec which extracts dense feature representations of each player by predicting play outcomes without the use of hand-crafted heuristics or aggregate statistical measures. Specifically, our model aimed to predict the outcome of a possession given both the offensive and defensive players on the court. By training on over 3.5 million plays involving 1551 distinct players, our model was able to achieve a 0.3 K-L divergence with respect to the empirical play-by-play distribution. The resulting embedding space is consistent with general classifications of player position and style, and the embedding dimensions correlated at a significant level with traditional box score metrics. Finally, we demonstrate that NBA2Vec accurately predicts the outcomes to various 2017 NBA Playoffs series, and shows potential in determining optimal lineup match-ups. Future applications of NBA2Vec embeddings to characterize players' style may revolutionize predictive models for player acquisition and coaching decisions that maximize team success.