A data visualization project built for DS 4200. Five visualizations, three static (Matplotlib) and two interactive (Altair and D3.js v7), to examine how BARTHAG, ADJOE, ADJDE, WAB, and shooting metrics relate to NCAA Tournament outcomes across 3,885 team-seasons from 2013–2024.
Live site: https://ytpatel3.github.io/cbb-analytics/
Kaggle: College Basketball Dataset
- Originally sourced from barttorvik.com.
- Covers NCAA Division I, seasons 2013–2019 and 2021–2024 (2020 omitted since tournament was canceled).
- Predicting the Unpredictable: Predicting the March Madness Champion Using Statistical Modeling by Jack Sweeney (2025).
- Seed Distribution and Upset Probabilities in the NCAA Men’s Basketball Tournament by S. H. Jacobson, J. G. King, & E. C. Sewell (2011).
| # | Type | Tool | Description |
|---|---|---|---|
| 1 | Static | Matplotlib | Box plot: BARTHAG distribution by major conference |
| 2 | Static | Matplotlib | Radar chart: performance profiles of champions vs. runners-up |
| 3 | Static | Matplotlib | Multi-faceted scatter: WAB vs. ADJOE, 2P_O, 3P_O |
| 4 | Interactive | Altair | Efficiency scatter plot: (ADJOE vs. ADJDE) with season slider and legend click |
| 5 | Interactive | D3.js | Stacked bar chart: tournament outcome distribution by efficiency tier, with metric dropdown and click-to-highlight |