spb/anomaly-atlas Public License

1---2project: anomaly-atlas3document: F002 confidence4author: Simon-Pierre Boucher5contact: contact@spboucher.ai6data_source: hfmarketdata.io7created: 2026-08-128status: reviewed9---1011Level: 21213# Confidence — F002 (artifact demonstration)1415**Level 2 — robust.** This is a finding-type C/D entry (artifact16demonstration / methodological), published as a negative result about a17would-be anomaly.1819* **Corrected**: the effect itself survives Hansen SPA — which is the20  demonstration: statistical correction cannot detect mechanism.21* **Artifact mechanism established**: measured in expB (+0.065) and expD22  (+0.132) with print-synchronization applied — the lead persists at23  fresh_fraction 0.996 with artifact-share-from-print-gaps ~0, isolating24  CONTENT staleness (Fisher 1966) as the mechanism; the synthetic gate25  demonstrates the same signature on planted staleness.26* **Out-of-sample**: replicates on the validation split (+0.080) — the27  artifact is structural, unlike every "alpha" in the same pipeline.28* **Robust**: present in two independent windows plus validation, at both29  raw and synchronized treatments.3031**Why not Level 3**: holdout untouched (sealed); magnitude varies with32index-constituent liquidity regime, so the number (not the mechanism)33should be re-measured per period.34