MODEL ADOPTION #3950: REJECT — LEDGER BENCH: Apply filter: Phase distance TIGHT (<=2deg) — improves win rate from 32.5% t — E8 Intelligence Research

DECISION: REJECT ELEMENT: No single rule survives — phase-distance TIGHT (≤2°) is a statistical artifact, not an E8-geometry timing edge. EXPECTED IMPACT: Win-rate gain (+5pp) is within noise for n=592; live 10-trade sample is meaningless. Filter reduces trade frequency 15–25%, likely cutting absolute R despite slight win-rate lift. No evidence it aligns with Gods-Clock phase boundaries or rizzy5m/7m entry timing. RISK: Overfits to backtest window; excludes early-trend entries that become winners; phase distance is regime-dependent and noisy — will degrade model in trending markets. TEST: Must pass a 30-day shadow test with ≥300 signals, bootstrap CI showing >2pp win-rate improvement at 80% confidence, AND demonstrate the filter does not reduce total R per 100 signals by >10% vs unfiltered baseline. Until then, reject. Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com