MODEL ADOPTION #449: REJECT — LEDGER BENCH: Apply filter: Session = NY_OPEN — improves win rate from 32.4% to 38.5% (+6. — E8 Intelligence Research

DECISION: REJECT ELEMENT: The NY_OPEN filter is a time-window heuristic, not a geometry-timed edge. It lacks alignment with E8-node/Gods-Clock rhythm logic, which operates on fractal phase transitions, not arbitrary session boundaries. The +6.2pp win-rate improvement is likely a regime artifact (high-liquidity bias) and will not survive regime shifts or low-volatility periods. The 70-80% trade frequency collapse destroys absolute profit potential even if win rate improves, and the net PnL delta in your own sample was zero. EXPECTED IMPACT: Win rate may rise 3-5pp in high-liquidity regimes, but absolute profit drops 50-70% due to missed trades. Over 6 months, the filter underperforms the unfiltered model in total return. RISK: Overfitting to a specific historical window (n=589 is small for regime dependency). The filter will miss strong E8-timed entries that occur outside NY_OPEN (e.g., London close, Tokyo breakouts). Degrades model adaptability. TEST: Before any reconsideration Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com