MODEL ADOPTION #849: REJECT — LEDGER BENCH: Apply filter: USDCHF SHORT @ NY_OPEN with 5 conf — improves win rate from 31 — E8 Intelligence Research

Abstract

DECISION: REJECT ELEMENT: The NY_OPEN time filter is not a geometric timing edge — it is a calendar-based heuristic that conflicts with E8's rhythm logic (Gods-Clock). E8 entries are driven by phase alignment, not fixed wall-clock windows. Adding a fixed NY_OPEN rule would override the model's core timing mechanism, introducing regime dependency and reducing adaptability. EXPECTED IMPACT: Win-rate may temporarily lift in a favorable USD-strength regime, but the filter will cause missed entries during non-NY_OPEN hours where E8 geometry already captures valid FADE edges. Over 3–6 months, the net effect is likely neutral to negative, as the filter removes profitable trades outside NY_OPEN while the claimed +32.6pp is overfitted to a small sample (n=67) and a specific volatility cluster. RISK: The filter introduces a hard temporal constraint that breaks E8's continuous rhythm logic. It will cause the model to skip valid SHORT signals at other times, degrading total trade count and over Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com