MODEL ADOPTION #964: REJECT — LEDGER BENCH: Apply filter: Phase distance MODERATE (5-10deg) — improves win rate from 31. — E8 Intelligence Research

DECISION: REJECT ELEMENT: The phase distance filter (5–10 deg) fails the hostility test: it is a blunt band-pass on a continuous variable, not a discrete E8 geometry timing edge. The backtest improvement (+1.8pp, n=156) is fragile — the sample is small, the effect is within noise, and the filter will reject winning trades that happen to fall outside that band, reducing throughput by 10–20% with no guarantee the win-rate gain persists. The "same-window" projection (57% → 80%) is a cherry-picked 7-trade sample, not evidence. Overfitting to a specific regime is the primary risk; the filter has no theoretical anchor in Gods-Clock rhythm logic or FADE polarity mechanics. EXPECTED IMPACT: Negligible or negative. Win-rate may tick up 1–2pp in the short term, but absolute PnL likely drops due to fewer trades and capped upside from rejected winners. RISK: Regime shift renders the 5–10 deg band irrelevant; filter becomes a drag. Also, the filter's rejection rate is unmeasured — if it kills Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com