To view the project on GitHub, click here.
One of my favorite things about working with data is how it can challenge assumptions—and this project did exactly that.
I set out to explore whether WNBA teams exhibit more statistical balance than NBA teams, using Python to analyze key offensive metrics across both leagues. What I found not only reinforced some long-standing differences in style but also raised bigger questions about how performance is structured and valued in pro basketball.
📊 The Setup
Using available 2022-2024 season data for both leagues, I:
- Applied linear weights modeling to assess how various stats (assists, rebounds, turnovers, etc.) contribute to team scoring
- Used regression analysis to measure how strongly each stat correlated with points per game
- Ran a Team of Clones analysis—a method where we simulate a team made entirely of one player’s average stats—to see how dependent a team is on a single standout player
🔍 What I Found
- The WNBA’s statistical sample was more limited, both in size and in overall raw performance. Across the board, most WNBA teams trailed NBA teams in raw offensive metrics (like FG%, assists, and scoring averages).
- Despite this, the top WNBA players held their own in the Team of Clones analysis. Some even approached or matched the clone efficiency of NBA stars.
- This contrast highlights a key strategic difference:
💡 The NBA is structured around star-centric offenses, while the WNBA fosters more balanced team play.
In other words, while NBA teams often rely on standout individual production, WNBA teams tend to share the load, creating a more evenly distributed stat profile—even if it doesn’t always result in higher raw output.
💡 Why It Matters
This project goes beyond basketball. It’s a case study in how data can reveal the hidden structure of performance systems. Whether you’re building a team in sports, business, or tech, this kind of analysis can help answer questions like:
- Are we over-indexing on top performers?
- How does balance affect long-term success?
- What are the trade-offs between standout stats and team synergy?
And of course, it underscores how crucial data access and context are in drawing fair comparisons—something we should always be aware of when analyzing across groups or systems.
🛠️ Tools Used
- Python: pandas, statsmodels, matplotlib, seaborn
- Modeling Methods: Linear weights, regression analysis, simulation-based scoring models
- Data Storytelling: Clear visualizations and clone-based evaluation for cross-league comparison
👋 Let’s Connect
This project was a blend of sports analytics, strategy, and curiosity—and I’m always looking for opportunities that let me bring those strengths to the table. If you’re hiring, collaborating, or just want to talk data, I’d love to connect.
📬 madelineludwig13@gmail.com
🌐 Explore more of my work