In-play fade-the-goal: when a goal goes in, back the CONCEDING side at the real live price if the conditioned posterior still shows >=10% edge. The bayes replay's best surviving policy (+11.7% ROI over 1,324 bets on the own-goal-corrected sample, 6/10 winning seasons vs +1.1% control). Placed by the watcher, max 2 per contest, savepoint-safe. The definitive grade is this live sample at real odds.
football · Match result · status: live · betting forward from 2026-07-07 · same $1000 start as every other strategy
| Bet | Stake | Odds | To win | Model edge | P&L | Status |
|---|---|---|---|---|---|---|
Norway win in Norway v England · Match result @ 5.50 | $5.00 | 5.50 | $22.50 | +121% | -5.00 | lost |
"Model edge" = model probability × odds − 1 at the moment of betting: the profit the model expected per $1. Positive edge can still lose (single bets usually do) — the question is whether edges are positive on average AND the model's probabilities are honest.