MODEL ADOPTION #1031: REJECT — LEDGER BENCH: Apply filter: Session = NY_OPEN — improves win rate from 31.5% to 37.5% (+6. — E8 Intelligence Research

Abstract

DECISION: REJECT ELEMENT: The NY_OPEN session filter is not compatible with E8 geometry timing. E8-node/Gods-Clock rhythm logic operates on fractal phase transitions that are session-agnostic. Forcing a fixed UTC window (13:00-16:00 UTC) introduces a calendar-based constraint that violates the model's core premise: that edge arises from harmonic timing, not human-defined market hours. The +6.0pp win-rate improvement (n=608) is likely a liquidity artifact, not a structural edge — it will degrade when DST shifts, holiday schedules, or news events alter NY session behavior. The 72% confidence is inflated given the small sample and lack of out-of-sample validation across regime changes. EXPECTED IMPACT: Win rate would temporarily rise ~3-5pp, but absolute PnL would drop 40-60% due to trade count collapse. Profit per trade may increase slightly, but total profit falls. The model becomes brittle — dependent on a single session's liquidity profile. RISK: Overfitting to a narrow time window Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com