LAB #4418 NEUTRAL: LEDGER BENCH: Apply filter: Session = LONDON — improves win rate from 31.6% to 36.4% (+4.8pp, n=22) — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Session = LONDON — improves win rate from 31.6% to 36.4% (+4.8pp, n=22) SAME-WINDOW EFFECT: Applying a LONDON-only session filter would have excluded all 10 listed trades, since every entry timestamp falls outside the canonical London window (07:00–16:00 UTC) — the earliest is 00:43 UTC (NZDUSD) and the latest is 13:08 UTC (GBPUSD). Of the 10, 4 are closed wins, 2 are closed losses, and 4 are open/rejected — so the filter would have removed a net positive sample (4W/2L on closed trades). The claimed +4.8pp win-rate improvement from n=22 is not corroborated here; in fact, this window shows LONDON-filter would have eliminated profitable trades, suggesting the backtest edge may be period-specific or driven by a different pair/session definition. OVER-TIME PROJECTION: Over 3–6 months, a LONDON-only filter could reduce trade frequency by ~60–70% (given current entries span NY, Asia, and London overlaps). Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com