How the model works
Six data sources, a gradient-boosted baseline, and a Claude-powered reasoning layer. Every projection is back-tested weekly so claimed confidence actually means something.
Inputs
Every projection pulls from play-by-play, market, weather, injury, and athletic data — refreshed daily during the season.
Play-by-play, EPA, CPOE, success rate, 2006–present
Air yards, separation, time to throw, coverage grades
Depth charts, injury reports, snap counts
Game-day weather: wind, precipitation, temperature, indoor/outdoor
Implied totals, spreads, line movement across sharp books
Athletic testing, 40-yard, vertical, three-cone percentiles
Pipeline
Daily pulls from six data sources, normalized into a unified Postgres schema. De-duplication, missing-value handling, and outlier flags before any modeling touches it.
Weighted rolling splits (3-game, 5-game, 10-game) for every usage metric. Defensive matchup adjustments by coverage scheme. Game-script priors from Vegas implied totals.
Gradient-boosted regression trained on 2006–2024 play-level data. Position-specific models (QB, RB, WR, TE) with cross-validation by season to prevent look-ahead bias.
Claude 4.7 interprets the projection against the prop line, matchup context, and injury data. Outputs a confidence score, direction, and written reasoning trace.
Compares our projection against book lines. Flags mismatches where the model differs from the market by 2+ standard deviations after weather and injury adjustments.
Weekly back-test vs. actual outcomes. Recalibrate confidence bands so that claimed 70% picks actually hit 70% of the time. Miss rate is public on the pricing page.
Back-test · 30-day rolling
Confidence calibration measures how closely our claimed confidence (e.g. 70% of picks hit) matches actual hit rate over a rolling 30-day window. Lower drift = better calibrated.
Philosophy
Confidence maxes out at ~85% for a reason. Anyone claiming 95%+ is either miscalibrated or lying.
We publish the full 30-day window and disclose when we regress. Streak-hunting is marketing, not methodology.
If the data is missing or stale, we flag the projection as low-confidence — we don't fill gaps with priors and pretend.
Every tool is free during beta — AI prop analyzer, player database, 2026 draft prospects, WNBA hub. No signup required.