LAB #4465 PROMISING: LEDGER BENCH: Apply filter: Session = OFF_SESSION — improves win rate from 31.5% to 41.9% (+10.3pp, — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Session = OFF_SESSION — improves win rate from 31.5% to 41.9% (+10.3pp, n=129) SAME-WINDOW EFFECT: Applying the OFF_SESSION filter would have blocked the two wkjudas entries on 2026-09-01 (GBPUSD LONG at 15:08 and USDJPY SHORT at 14:36) — both closed as LOSSES, removing 2 losing trades from the 10-trade window. It would also have blocked the rizzy5m EURJPY SHORT at 11:35 (REJECTED/open) and the rizzy7m AUDUSD LONG at 09:56 (REJECTED/open), but those were not closed outcomes. Net effect on the closed set: 6 wins, 2 losses remain (75% win rate vs. current 60%), and PnL improves by removing two full losses — though the sample is too small to attribute the +10.3pp backtest gain to this window alone. OVER-TIME PROJECTION: Over 3–6 months, the filter should reduce exposure during low-liquidity/off-session hours where spreads widen and false breakouts cluster — likely lifting win rate toward the 41.9% back

Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com