Aging Decline in Basketball Career Trend Prediction Based on Machine Learning and LSTM Model

2025-09-30

AI Summary

This study addresses the modeling and prediction of age-related performance decline in NBA players. Methodologically, it proposes a unified analytical framework integrating unsupervised learning with temporal modeling: first, an autoencoder extracts low-dimensional representations of career trajectories, followed by K-means clustering to yield interpretable player development archetypes; subsequently, cluster-specific LSTM models are trained for personalized, dynamic forecasting of individual season-level performance. Compared to conventional statistical approaches or monolithic deep learning models, the framework achieves significantly improved generalization across players and seasons, outperforming baseline methods on key metrics—including prediction error for points per game (PTS) and Player Efficiency Rating (PER). Its core contribution lies in the principled coupling of career trajectory typology with individualized time-series forecasting, establishing a transferable methodological paradigm for aging modeling and athletic career management in sports science.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning
• Planning, Routing, and Scheduling: Model-Based Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: Studies of user behavior, including longitudinal effects of personalized systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

The topic of aging decline on performance of NBA players has been discussed in this study. The autoencoder with K-means clustering machine learning method was adopted to career trend classification of NBA players, and the LSTM deep learning method was adopted in performance prediction of each NBA player. The dataset was collected from the basketball game data of veteran NBA players. The contribution of the work performed better than the other methods with generalization ability for evaluating various types of NBA career trend, and can be applied in different types of sports in the field of sport analytics.

Problem

• Predicting NBA players' performance decline with aging trends
• Classifying career trajectories using autoencoder and K-means clustering
• Forecasting individual player performance through LSTM deep learning models

Innovation

• Autoencoder with K-means clustering for career classification
• LSTM deep learning method for performance prediction
• Dataset collected from veteran NBA players game data