Terms & Disclaimers

Last updated: March 2026

Prediction Accuracy

All predictions, win probabilities, and analytics on Ubunifu Madness are provided for informational and entertainment purposes only. Our model achieves a Brier score of 0.1413 on historical tournament data, but past accuracy does not guarantee future results. The NCAA tournament is inherently unpredictable — no model can guarantee outcomes. A 70% win probability means the underdog still wins 3 out of 10 times. Use these predictions as one input alongside your own knowledge, not as the sole basis for any decision.

Not for Gambling

Ubunifu Madness is designed exclusively for bracket prediction analytics, education, and entertainment. This platform is not intended to facilitate, encourage, or support sports gambling or wagering of any kind. We do not provide odds, point spreads, or any betting-related information.

Data Sources & Attribution

Historical game data is sourced from the Kaggle March Machine Learning Mania 2026 competition dataset, used under the CC BY 4.0 license.

Live scores, team records, rosters, and tournament bracket data are sourced from ESPN. Team logos and branding are property of their respective institutions.

Massey ordinal rankings aggregate data from multiple independent ranking systems as compiled by Kenneth Massey.

No Affiliation

Ubunifu Madness is not affiliated with, endorsed by, or sponsored by the NCAA, ESPN, any collegiate athletic conference, or any university. "March Madness" and "Final Four" are registered trademarks of the NCAA.

Use at Your Own Risk

This platform is provided "as is" without warranty of any kind, express or implied, including but not limited to the warranties of accuracy, completeness, or fitness for a particular purpose. In no event shall Ubunifu Madness or its contributors be liable for any damages arising from the use of predictions or analytics provided on this platform.

Open Source

Ubunifu Madness is an open-source project built for the Kaggle March Machine Learning Mania competition. Our methodology is fully transparent — see the How It Works page for a complete explanation of our models and data pipeline.

Questions about these terms? Ask the Madness Agent.