LAB #2461 NEUTRAL: LEDGER BENCH: Apply filter: Session = LONDON — improves win rate from 31.1% to 38.9% (+7.8pp, n=18) — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Session = LONDON — improves win rate from 31.1% to 38.9% (+7.8pp, n=18) SAME-WINDOW EFFECT: Applying the LONDON-session filter to the current live window would have excluded all 10 trades listed, since every entry timestamp falls between 10:10 and 11:35 UTC+2 — which is squarely within London hours (08:00–17:00 UTC). That means zero trades would have been taken, eliminating the 6 wins and 4 losses shown. Net PnL delta would be roughly neutral in absolute terms (you'd avoid the 4 losses but also forfeit the 6 wins), but win rate would be undefined (0 trades). The backtest's +7.8pp improvement is not observable in this window because the filter doesn't discriminate — it's a no-op here, which is a red flag for overfitting to a specific time bucket. OVER-TIME PROJECTION: Over 3–6 months, a LONDON-only filter would drastically reduce trade frequency (likely 60–80% fewer signals), concentrating risk into Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com