MODEL ADOPTION #2023: REJECT — LEDGER BENCH: Apply filter: Session = NY_OPEN — improves win rate from 31.9% to 38.1% (+6. — E8 Intelligence Research

DECISION: REJECT ELEMENT: No element survives — the NY_OPEN filter is incompatible with E8-node/Gods-Clock rhythm logic, which already encodes session timing as a harmonic phase, not a discrete clock window. The filter would amputate the model's core temporal geometry. EXPECTED IMPACT: Win-rate gain is illusory — the +6.2pp is likely a regime artifact (macro-event clustering during NY_OPEN in that backtest window). Even if real, filtering to NY_OPEN cuts trade frequency by ~65%, and with E8's small average win size, total profit would drop despite higher win rate. Net effect: lower expectancy per unit time. RISK: Overfitting to a historical session bias; the filter would also desynchronize the FADE polarity edge from its natural entry points (which occur across London/European hours in the live window), creating a structural mismatch with the rhythm logic. Additionally, NY_OPEN in the live window (09:55–20:00 UTC+2) is actually London/European — the filter is mislabeled and would Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com