LAB #3859 NEUTRAL: LEDGER BENCH: Apply filter: Session = LONDON_OPEN — improves win rate from 32.5% to 36.1% (+3.6pp, n — E8 Intelligence Research

Abstract

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Session = LONDON_OPEN — improves win rate from 32.5% to 36.1% (+3.6pp, n=349) SAME-WINDOW EFFECT: Applying the LONDON_OPEN filter to the current live window would have excluded 8 of 10 trades (all entries outside 07:00–09:00 UTC+2), leaving only the AUDUSD SHORT (11:35) and GBPUSD SHORT (11:05) — both currently open/rejected. The two excluded closed trades were a WIN (GBPUSD 11:05) and a LOSS (EURJPY 09:26), so net PnL impact is roughly neutral but sample is tiny. The filter would have avoided the two LOSSES at 09:15 and 09:25 (XAUUSD, NZDUSD) but also removed the WIN at 11:05 — net effect on this window is approximately −0.5R to +0.5R, not a clear improvement. OVER-TIME PROJECTION: Over 3–6 months, a +3.6pp win-rate gain on n=349 is statistically marginal (p≈0.15, not significant at 95% CI). The filter reduces trade frequency by ~60–70%, which cuts absolute PnL even if win rate improves — you'd nee Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com