MODEL ADOPTION #1803: REJECT — LEDGER BENCH: Apply filter: Phase distance MODERATE (5-10deg) — improves win rate from 31. — E8 Intelligence Research

DECISION: REJECT ELEMENT: No element survives — the 5–10° phase-distance filter is a statistical mirage, not a geometric edge. EXPECTED IMPACT: +1.4pp on n=156 is within noise (±4.5pp CI); even if real, the 20–30% trade-frequency cut likely nets negative total PnL, as filtered trades include high-RR winners (e.g., EURJPY SHORT). RISK: Overfitting to a narrow band that is regime-dependent; phase-distance distribution shifts with volatility, making the band irrelevant or harmful. Also, the filter contradicts E8 rhythm logic — phase distance is a continuous variable, not a discrete state; chopping it into 5–10° creates arbitrary boundaries with no geometric basis. TEST: Before any reconsideration, run a 60-day shadow test on the exact instrument set with ≥200 signals per arm (all vs. 5–10° band) AND adjacent bands (0–5°, 10–15°). Require p<0.05 on win-rate delta AND a net PnL improvement after accounting for reduced frequency. Also verify the band's stability across two distinct v Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com