LAB #4151 NEUTRAL: LEDGER BENCH: Apply filter: Session = NY_OPEN — improves win rate from 32.5% to 38.9% (+6.4pp, n=781 — E8 Intelligence Research

Abstract

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Session = NY_OPEN — improves win rate from 32.5% to 38.9% (+6.4pp, n=781) SAME-WINDOW EFFECT: Applying the NY_OPEN session filter to the live window above would have excluded all trades timestamped outside the NY session (roughly 13:30–20:00 UTC / 15:30–22:00 SAST). Of the 10 listed trades, only the GBPUSD SHORT at 11:50 SAST (09:50 UTC) and the NZDUSD SHORT at 10:20 SAST (08:20 UTC) fall outside NY_OPEN; the rest are either late NY or pre-NY. Excluding those two would remove one WIN (NZDUSD) and one OPEN (GBPUSD), leaving a mixed record — the remaining 8 include 2 wins, 3 losses, 2 opens, 1 rejected. The backtest's +6.4pp win-rate gain does not translate cleanly here; the live sample is too small and the filter would have cut a winner, suggesting the edge is not visible in this window. OVER-TIME PROJECTION: Over 3–6 months, a NY_OPEN-only filter could plausibly improve win rate if the backtest's n= Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com