In the 2023 NBA offseason, several playoff-contending teams fired their coaches and I became curious about what types of candidates they should pursue. Mike Budenholzer, Monty Williams, and Doc Rivers quickly became candidates for other highly-desirable positions which led me to wonder if a first time head coach can produce better results; Will Hardy and Joe Mazzulla each had tremendous regular season success in their first seasons last year. This repository contains a Jupyter Notebook that investigates the impact of coaching experience on winning percentages in the NBA. The dataset spans from the 2000 season to 2023, including statistics on regular-season wins, playoff wins, and other relevant information. This analysis aims to provide insights into how a coach's experience and performance in previous seasons affect their ability to lead their team to victory.
NBA_Coaching_Analysis.ipynb: The main Jupyter Notebook that contains all code, analyses, and visualizations.
- Python 3.x
- Jupyter Notebook
- Pandas
- Selenium
- BeautifulSoup
- Pandas for data manipulation and analysis
- Selenium for web scraping
- BeautifulSoup for parsing HTML and extracting data
The data is scraped from the Basketball Reference website for NBA coaches from the 2000 to 2023 seasons. The notebook uses Selenium for web scraping and collects information on regular-season and playoff wins for each coach and team.
- Data Preprocessing: Cleaning and preparing the scraped data for analysis.
- Regular Season Winning Percentage: Analyzing how the overall seasons and regular-season wins affect the winning percentage.
- Playoff Performance: Investigating how coaches with previous playoff experience perform in subsequent seasons.
- Coaching Hire Analysis: Understanding the effectiveness of hiring experienced coaches for "win-now" scenarios.
If you find any issues or have suggestions for additional analyses, please feel free to open an issue or make a pull request.
Special thanks to Basketball Reference for providing the data.