MODEL ADOPTION #1052: REJECT — LEDGER BENCH: Apply filter: Session = NY_OPEN — improves win rate from 31.6% to 37.7% (+6. — E8 Intelligence Research

Abstract

DECISION: REJECT ELEMENT: The NY_OPEN filter (13:00–20:00 UTC) is a time-window heuristic, not a geometric edge. It does not align with E8-node/Gods-Clock rhythm logic, which operates on harmonic intervals, not arbitrary market sessions. The backtest n=605 is too small to rule out overfitting to a specific volatility regime, and the live shadow-test already showed a *decrease* in win rate (55.6% → 50%) on the current window. The filter would also cut trade count by ~40–50%, likely reducing absolute PnL even if win rate improves marginally. EXPECTED IMPACT: Win rate might rise to ~37% but absolute PnL would drop due to fewer trades and removal of high-R:R setups that occur outside NY open. Net effect: likely negative or neutral. RISK: Overfitting to NY session volatility patterns that break during holidays, central bank events, or regime shifts. The filter is incompatible with E8 timing geometry, introducing a non-harmonic constraint that degrades the model's structural coherence. T Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com