LAB #4088 NEUTRAL: LEDGER BENCH: Apply filter: Session = OFF_SESSION — improves win rate from 32.5% to 42.5% (+10.0pp, — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Session = OFF_SESSION — improves win rate from 32.5% to 42.5% (+10.0pp, n=127) SAME-WINDOW EFFECT: Applying the OFF_SESSION filter would have blocked the 15:08 GBPUSD long (wkjudas), the 14:36 USDJPY short (wkjudas), and the 02:43 NZDUSD short (wkjudas) — all three are OFF_SESSION entries. The 02:43 NZDUSD is still open, but the two closed ones are losses; removing them would flip the window from 2W/5L to 2W/3L on closed trades, a raw win-rate jump from ~28.6% to 40%. PnL delta is positive but modest — you avoid two losing trades, yet you also lose the potential upside of any future OFF_SESSION winners that might have occurred in this small sample. The 09:15–11:35 rizzy entries are all ON_SESSION and unaffected, so the filter only trims the wkjudas tail, not the core bleed. OVER-TIME PROJECTION: Over 3–6 months, a +10pp win-rate gain on n=127 is statistically weak (standard error ~4.4pp, so the true Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com