NBA Player Data Statistics and Predictions Based on Spark

Abstract

Entering the 21st century, the sports field has begun to rely on advanced data to establish fine-grained athlete evaluation metrics, such as the impact of an athlete’s presence on the team and performance in critical moments. These metrics are included in statistics and analysis. This experiment analyzes player datasets from NBA games, processing and statistically analyzing data indicators such as player rankings, changes brought by player age and career length, and changes in the proportion of three-point shots. The rating score proposed in this study is used as a measurement indicator to predict Luka Doncic’s performance since entering the league. The model’s performance is evaluated by calculating the root mean square error between the predicted and actual data. Analyzing and predicting player game data is crucial for evaluating player performance, formulating tactics, and management strategies. It not only helps management and coaching teams make more informed draft and trade decisions but also reveals similarities and differences among players through techniques such as cluster analysis, guiding personalized training and tactics.