About macfax
Project Overview
macfax is a comprehensive college basketball analytics platform that aggregates and analyzes data from multiple sources to provide advanced metrics, efficiency ratings, and predictive models for all 365 NCAA Division I men's basketball teams.
Our mission is to make advanced basketball analytics accessible and understandable, combining the best features of sites like KenPom and T-Rank with original analysis and visualizations.
Data Sources
We aggregate data from three primary sources, each providing unique insights:
KenPom.com
Provides adjusted efficiency metrics (AdjEM, AdjO, AdjD), tempo, luck ratings, and strength of schedule.
- Adjusted for opponent quality and pace
- Updated daily throughout the season
- Gold standard for efficiency metrics
Bart Torvik (T-Rank)
Comprehensive Four Factors data, shooting splits, and advanced team statistics.
- Detailed Four Factors breakdown (eFG%, TOV%, ORB%, FTR)
- 2P% and 3P% shooting splits
- Wins Above Bubble (WAB) and Barthag ratings
CBB Analytics
Additional adjusted metrics and supplemental data validation.
- Alternative adjusted efficiency calculations
- Cross-validation of Four Factors
- Independent data verification
Methodology
Data Collection
We use automated Python scripts with browser automation (Playwright) to collect data from public sources daily. All scraping is done ethically with appropriate rate limiting and respects robots.txt policies.
Data Normalization
Team names are normalized across all three sources using a comprehensive mapping system. This ensures that "UConn", "Connecticut", and "CONN" all refer to the same team, enabling seamless cross-dataset analysis.
Metric Calculations
Our derived metrics (margins, edges) are calculated using standard formulas:
- eFG% Margin: Team eFG% minus Opponent eFG%
- Turnover Edge: Forced TOV% minus Team TOV%
- Rebounding Edge: ORB% minus (100% - DRB%)
- FTR Margin: Team FTR minus Opponent FTR
Update Frequency
Data is scraped and processed daily during the season. The website is rebuilt with fresh data each day to ensure you're always seeing the most current statistics.
Technology Stack
Frontend
- Next.js 14 (App Router)
- TypeScript
- Tailwind CSS
- Apache ECharts
- TanStack Table
Data Pipeline
- Python 3.14
- Playwright
- pandas + BeautifulSoup
- SQLite
- CSV/JSON export
Disclaimers & Attribution
Not Affiliated: This project is not affiliated with, endorsed by, or officially connected to KenPom.com, barttorvik.com, or CBB Analytics. We are an independent project created for educational and analytical purposes.
Data Attribution: All efficiency metrics are calculated using methodologies pioneered by Ken Pomeroy. Four Factors analysis is based on Dean Oliver's work. We gratefully acknowledge these contributions to basketball analytics.
Accuracy: While we strive for accuracy, this tool is provided "as is" for informational and entertainment purposes. Always verify critical information with official sources.