Evidence
What is proven — and what is not
Historical replay, delayed public examples, and forward capture are separate evidence classes.
Historical replay (V2)
4,892 completed 30-trading-session outcomes from a frozen parent universe of 5,517 campaigns.
- • Parent frozen universe: 5,517 campaigns
- • Evidence-eligible: 4,943
- • Complete 30 trading sessions: 4,892
- • Excluded: 574
- • Pending 7 / 30 / 90 trading sessions: 19 / 51 / 68
- • Complete 30td parent coverage: 88.67%
- • 54.97% positive completed outcomes · +1.26% median endpoint
- • Positive and negative outcomes included · real SEC and market data · strict as-of replay
Historical V1 is frozen and invalid for performance claims. Current customer metrics bind only to Historical V2. Verify at /archive.
Delayed public examples
Current policy releases Public copies 7×24 hours after the immutable VIP timestamp. Legacy three-day releases keep their recorded timestamps and are labelled as legacy — they are not seven-day delay proof. Historical replay examples are not original live VIP delivery proof.
Browse public delayed signals →Forward validation
Since 2026-08-05, every new campaign is captured with immutable source and decision evidence. Forward results will be published after predefined sample thresholds.
Forward status →Rejected ideas
Weak hypotheses stay visible so marketing cannot quietly drop failures.
How we reject weak ideas →Advanced methodology
Technical release hashes, replay classes, and research notes live under methodology for auditors — not as homepage marketing language.
Methodology →