BBMISports
DashboardRankingsPlayoff PulseThis Week's PicksModel PerformanceResearchAbout
Research series · 1 installment

NCAA March Madness 2026.

The basketball model's first live tournament, graded in public: what the bracket picks returned, where March went wrong, and the input-layer fix tested across six re-run seasons.

← All research
01NCAAB · Methodology

You asked about the basketball model. Here are the answers, including the ugly one.

Three questions came up over and over after the first live season: was that a one-year fluke, how does the thing actually work, and what on earth happened in March? A six-season walk-forward, a diagnosis we could have ducked and didn't, and a fix we tested before we believed it.

57.1% on 2,017 live picks · 28–28 in the bracket · six seasons re-run · March mispricing cut 54%Jul 27 · 14 min

Independent sports analytics and model projections, for informational purposes only. Rankings, predictions, and probabilities are model outputs, not financial or betting advice, and past performance does not guarantee future results.

© 2026 BBMI SportsModel ChangelogPrivacyTerms