Methodology

How the edge engine works.

BetEdge is a decision-support analytics tool, not a tipster. The numbers below come from a deterministic pipeline you can audit — and every result we publish stays public, win or lose. Here is the whole thing, end to end.

1. 1 · Market intake

   We pull live odds across the books through the day and treat Pinnacle as the sharp anchor — the closest thing the market has to an efficient price, because it moves on money, not opinion. No scraping of injuries or insider tips: just the prices, refreshed on a schedule.

2. 2 · De-vig fair value

   Every bookmaker price carries a margin (the vig). We strip it out of the sharpest line to recover a true fair-value probability for each outcome. This is deterministic math — the same inputs always produce the same number. No black box, no model trained to 'predict winners'.

3. 3 · Edge detection

   We compare the fair value to the price you can actually get at a book you can bet. When a book is priced above fair value, that gap is the edge — a measured mispricing, expressed post-tax. We surface a market only when the gap clears our threshold.

4. 4 · Risk & sizing labels

   Each surfaced edge carries context so it isn't read as a guarantee: the size of the gap, a sample-size note, and a Quarter-Kelly stake suggestion. Bigger isn't better — longshot prices are the noisiest, and we label them as such.

5. 5 · Result tracking, in public

   Every call is timestamped before kickoff to an append-only public ledger. We grade ourselves on model calibration — do our probabilities match reality — and show Closing-Line Value, wins and losses both, including the negatives. Nothing is deleted or backdated.

Informational only · not a bookmaker or tipster · not financial advice · 18+.