LAB #4539 NEUTRAL: LEDGER BENCH: Apply filter: Session = NY_OPEN — improves win rate from 31.5% to 38.3% (+6.7pp, n=802 — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Session = NY_OPEN — improves win rate from 31.5% to 38.3% (+6.7pp, n=802) SAME-WINDOW EFFECT: Applying the NY_OPEN filter to the current live window would have excluded all six trades timestamped between 09:56 and 11:50 CET (which correspond to NY pre-open/early session in winter), including the three CLOSED/WINs (USDCHF, USDCAD, GBPUSD) and the two REJECTED/open positions. This would have removed 3 wins and 1 loss from the settled set, leaving only the two wkjudas losses (GBPUSD, USDJPY) and the NZDUSD open — flipping the window from +2 net wins to −2 net losses. The backtest's +6.7pp win-rate gain is not visible here; in fact, the filter would have destroyed PnL in this specific sample, suggesting the edge is session-dependent but not uniformly positive. OVER-TIME PROJECTION: Over 3–6 months, the NY_OPEN filter may genuinely improve win rate if the 31.5%→38.3% gain is robust and not curve-fit to a Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com