LAB #4003 NEUTRAL: LEDGER BENCH: Apply filter: Session = LONDON_OPEN — improves win rate from 32.5% to 36.2% (+3.7pp, n — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Session = LONDON_OPEN — improves win rate from 32.5% to 36.2% (+3.7pp, n=348) SAME-WINDOW EFFECT: Applying the LONDON_OPEN filter to the current live window would have excluded 8 of the 10 trades shown, as their entry timestamps fall outside the typical London session (07:00–16:00 UTC+2). The two surviving trades — NZDUSD SHORT (02:43) and USDJPY SHORT (03:57) — are both early-morning entries, and both are currently LOSS or open with negative drift. This filter would have removed the two WINs (GBPUSD, EURUSD) and the one LOSS that occurred during London hours, leaving a net effect of removing 2 wins and 3 losses, which would have *reduced* win rate in this specific window from 30% (3/10) to 0% (0/2) — a stark contradiction to the backtest claim. The +3.7pp improvement is not visible here; instead, the filter would have worsened live performance by cutting profitable London entries. OVER-TIME PROJECT Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com