No black box theatrics. Here is exactly what happens between raw NFL data and the pick on your screen.
Every night the pipeline gathers the current week's inputs: the schedule, player-prop lines, injury reports and official depth charts, plus contextual signals. Everything is rebuilt for the live week so picks reflect the roster and situation as they stand now.
A production stack of fourteen models scores the full universe of qualifying props across quarterback, running back, wide receiver and tight end markets. Some markets use a canonical frozen architecture; four are refit weekly as new outcomes accumulate.
A raw model score is never shown to you. Once a pick qualifies, its confidence is remapped through historical accuracy curves, so the percentage you see reflects how often comparable picks have actually hit.
Before each game window, the relevant picks lock into an immutable pre-kickoff record. Later line moves or data refreshes overlay the frozen row rather than replacing it — no hindsight rewrites.
Only rows that clear the qualification threshold reach your board — just the final, calibrated, frozen prediction, with player, market, line, side and confidence. Never the raw internal scoring.
After games finish, results are tied back to the exact frozen pre-kickoff prediction and realized stats — no cherry-picking a friendlier version after the fact.
For market-specific context, see the NFL player props guide, rushing props, receiving props and the prop glossary.
For the historical accuracy behind every step above — by model, by SHARP rank and by AI Slap size — see the results and accuracy page. For a plain-English look at how machine learning models are actually trained and validated, read how AI and ML prop models work.
