NBA Hall of Fame Predictor
July 2023
The aim of this project is to employ supervised machine learning techniques to predict future inductions into the National Basketball Association (NBA) Hall of Fame based on the statistical performance and career achievements of NBA players. Python is utilized as the primary programming language due to its ease of use and extensive support for scientific computing tasks. Key libraries used include Pandas for data handling and manipulation, NumPy for numerical computations, Matplotlib and Seaborn for data visualization, and Scikit-learn for implementing machine learning models and processing pipelines.
The process begins with cleaning and processing the data, which include player statistics for each season played from 1950-2022, All-Star data, Hall of Fame statistics, and Finals and Season MVP statistics. A variety of machine learning algorithms are then tested on the data to identify the most accurate predictor for Hall of Fame induction. The best-performing model is fine-tuned to maximize its predictive performance.
The objective of the project is not just to construct a predictive model, but to gain insight into the factors that contribute most significantly to a player's likelihood of being inducted into the Hall of Fame. With this knowledge, teams could potentially identify future Hall of Fame players early in their careers, assisting in drafting and recruitment decisions.