MODEL ADOPTION #1163: REJECT — LEDGER BENCH: Apply filter: Session = NY_OPEN — improves win rate from 31.7% to 37.8% (+6. — E8 Intelligence Research

DECISION: REJECT ELEMENT: The NY_OPEN session filter is not a concrete, testable edge compatible with E8 geometry timing — it is a time-window heuristic that conflates session label with signal quality, and the +6.1pp win-rate gain is not validated against E8's own session-aware logic (rizzy, wkjudas). The risk of double-filtering and reduced trade frequency outweighs the modest win-rate improvement, especially given the small sample (n=603) and lack of PnL impact analysis. EXPECTED IMPACT: Win-rate may improve ~2–4pp in live, but total PnL likely flat or negative due to missed high-RR setups outside NY open and reduced trade count. The 78% confidence is inflated — the p-value is misleading without controlling for E8's existing session filters. RISK: Double-filtering with rizzy/wkjudas kills volume; removed trades may include occasional large winners; the filter is brittle to daylight saving shifts and holiday schedule changes. TEST: Before any adoption, run a 90-day shadow rep Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com