Price at publication
Evaluation uses the recorded reference price—not a favourable price selected after the result.
Public evidence · complete historyEvery result.
Evidence protocolRead methodology →Every result.
Nothing hidden.
A future subscriber should never need to trust a screenshot. Every published selection will retain its captured price, result, profit or loss and proof reference.
Settled public picks0no deleted losses
Net profitPendingsettlement-derived
Verified ROIPendingpublic picks only
Closing line valuePendingcoverage pending
Max drawdownPendingpeak-to-trough units
Calibration evidence
53 / 500 current-method observations
99 settled fixtures are retained across archived cohorts. Older methodology results remain visible but never inflate the current model gate.
0 · Start100 · Early signal500 · Review cohort
Daily cohort · collecting
Win probability is not edge.
One pick may be issued each UTC day for a fixture within seven days. It must still clear the 55% probability floor, positive expected value and current two-sided price checks.
Total picks33 settled · 0 pending
Avg. win probability58.77%estimated before kickoff
Actual hit rate33.33%1W · 2L
Daily ROI-33.33%-1.00u
Daily CLV8.38%100.0% coverage
Max drawdown1.00uderived from immutable order
Append-only public record
Profit claims lockedSelection ledger
DateSelectionCaptured priceResultP&LProof
No public selection has been published yet.
The ledger is intentionally empty. Shadow observations and editorial model views never enter verified performance.
Every outcome retained
Wins, losses, voids and corrections remain visible under the same settlement policy.
One performance cohort
ROI, CLV and drawdown use public selections only. Research and editorial views stay outside.