LAB #4140 PROMISING: LEDGER BENCH: Apply filter: Session = OFF_SESSION — improves win rate from 32.6% 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 32.6% to 42.5% (+10.0pp, n=127) SAME-WINDOW EFFECT: Applying the OFF_SESSION filter to the live window above would have removed the three wkjudas trades (GBPUSD LONG, USDJPY SHORT, NZDUSD SHORT) and the two rizzy5m trades from 2026-08-27 (AUDUSD SHORT, XAUUSD SHORT) — all of which were losses or still open. That leaves only the rizzy5m GBPUSD SHORT (WIN), NZDUSD SHORT (WIN), and the rizzy7m AUDUSD SHORT (LOSS) — a 2/3 win rate (66.7%) versus the unfiltered 3/10 (30%). PnL delta is strongly positive, but the sample is tiny and the filter appears to be selecting for rizzy5m/7m intraday entries while discarding wkjudas entirely — which may be the real driver, not the session itself. OVER-TIME PROJECTION: Over 3–6 months, the +10pp win-rate gain (n=127) is plausible if OFF_SESSION entries cluster around London/NY overlap where volatility and trend persisten Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com