LAB #4812 NEUTRAL: LEDGER BENCH: Apply filter: Session = LONDON_OPEN — improves win rate from 31.8% to 35.6% (+3.8pp, n — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Session = LONDON_OPEN — improves win rate from 31.8% to 35.6% (+3.8pp, n=393) SAME-WINDOW EFFECT: The London Open filter would have blocked the majority of the live window's trades, since entries cluster around 10:25–11:45 CET — which is post-London-open (08:00 London) but the filter as stated is a session tag, not a time-of-day cutoff, so the actual effect depends on how "LONDON_OPEN" is defined in the engine. If it means the London session broadly, almost none of the listed trades are filtered out and the window is unchanged. If it means the first hour of London (08:00–09:00 UK), then every trade in the window (10:25–11:45 CET = 09:25–10:45 UK) falls outside it, and the filter would have blocked 100% of the window — including the two winners (USDCHF, USDCAD) and the one open EURUSD short. That is a catastrophic same-window effect disguised as a +3.8pp improvement. OVER-TIME PROJECTION: A +3.8pp wi Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com