Our Mission
College Hoops Data was built with one goal: to help college basketball fans see the game differently. We believe that the best basketball analysis combines the eye test with data-driven insights, and our platform brings advanced analytics to every NCAA Division I team and game in a way that is accessible, visual, and actionable.
Whether you are a casual fan tracking your alma mater, a bracket enthusiast preparing for March Madness, or a serious analyst studying matchup tendencies, College Hoops Data gives you the tools to understand the game at a deeper level.
What We Offer
Live Scores and Box Scores: Real-time scores for every Division I game with detailed box scores, player statistics, and game flow data. Our scoreboard updates automatically so you never miss a crucial moment.
NET Rankings and Quad Records: The official NCAA NET rankings updated daily, with complete quadrant record breakdowns that the Selection Committee uses to evaluate tournament teams. Understand the difference between a Quad 1 road win and a Quad 4 home win.
AI-Powered Predictions: Our CHD Scout prediction model analyzes team efficiency metrics, player form, home court advantage, strength of schedule, and historical matchup data to forecast game outcomes. Each prediction includes a detailed breakdown of which team has the edge and why.
Tournament Simulator: Simulate remaining schedules to see how future results would impact a team's tournament resume. Get tournament grades from A+ to F based on the same criteria the Selection Committee evaluates.
Team and Player Analytics: Deep dive into every Division I team with offensive and defensive efficiency ratings, shooting percentages, rebounding rates, turnover margins, and more. Track individual player performance with hot and cold streak indicators.
AI-Generated Articles: Game previews and recaps powered by AI that analyze matchups, highlight key players, and provide context for every game.
Our Prediction Model
The CHD Scout prediction model uses a multi-factor approach to forecasting game outcomes. Key inputs include adjusted offensive and defensive efficiency, NET rankings, recent team form, player hot/cold streaks, and venue-specific home court advantage calculations.
The model uses a dual-architecture approach: a primary model for standard predictions and a specialized close-game model that activates when projected margins are tight. This dual approach helps avoid overconfident predictions in what are genuinely uncertain matchups.
We continuously backtest and refine the model throughout the season to improve accuracy.
Technology
College Hoops Data is built with modern web technologies. The frontend is a Flutter web application providing a fast, responsive experience across desktop and mobile devices. Our backend uses Supabase for real-time data storage and retrieval, with Firebase Cloud Functions handling data synchronization, article generation, and prediction calculations.
Data is sourced from publicly available NCAA and ESPN feeds and is updated in real-time during live games.