How it works

The engine pulls warm-up-extended market data and computes indicators, pivots and level statistics. The analysis is authored into one canonical payload, then both reports render behind a strict QA gate.

Every Pro report logs falsifiable predictions — exact levels and an exact window — before the session opens. Nothing is vague; each call can be proven right or wrong.

The free Snapshot opens for everyone; the Pro report unlocks with a subscription. Files are served from a CDN, so it's fast no matter how many people read it.

After the window closes, the engine grades each prediction against the actual price tape — Hit, Miss or No-trigger — with no human nudging the result.

Results land in an append-only ledger. Rows are never edited, re-tuned or cherry-picked, and a calibration table appears once ten reports are scored.