LAB #2023 PROMISING: LEDGER BENCH: Apply filter: Session = NY_OPEN — improves win rate from 31.9% to 38.1% (+6.2pp, n=619 — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Session = NY_OPEN — improves win rate from 31.9% to 38.1% (+6.2pp, n=619) SAME-WINDOW EFFECT: Applying the NY_OPEN filter would have excluded all 10 live trades listed, as every entry timestamp falls outside the NY_OPEN session (all are between 09:55 and 20:00 UTC+2, which spans London/European hours, not NY open). The current window shows 4 wins, 4 losses, 1 open, 1 pending — so the filter would have eliminated both wins and losses equally, yielding zero trades and zero PnL delta. This is a null result for the live window; it neither saves nor costs anything, but it also provides no evidence the filter works on recent market conditions. OVER-TIME PROJECTION: Over 3-6 months, a +6.2pp win-rate improvement on n=619 suggests a real edge if the session filter isolates a regime with better signal-to-noise (e.g., higher liquidity, tighter spreads, more directional momentum). However, the risk is overfitt Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com