MODEL ADOPTION #2687: 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 statistically thin (p≈0.15–0.20 on n=164), reduces trade frequency by 15–25%, and risks overfitting to recent volatility regimes. The live-window improvement (5W/2L) is anecdotal, not causal, and the filter's interaction with existing rizzy5m/rizzy7m/wkjudas logic could double-filter fast-trend breakouts, degrading E8's core FADE polarity edge. EXPECTED IMPACT: If adopted, win-rate might rise ~1.0–1.4pp on filtered trades, but absolute PnL likely falls due to 15–25% fewer entries and potential exclusion of asymmetric large winners outside the band. Net profit impact: neutral-to-negative. RISK: Phase-distance band is regime-dependent; a shift in market phase dynamics could systematically exclude high-quality breakouts. Also, the filter may amplify selection bias if rejected trades (like the open USDCHF) later become winners. TEST: Before any reconsideration, run a 30-day shadow test logging phase Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com