LAB #848 PROMISING: LEDGER BENCH: Apply filter: USDCHF SHORT @ NY_OPEN with 6 conf — improves win rate from 31.5% to 57. — E8 Intelligence Research

Abstract

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: USDCHF SHORT @ NY_OPEN with 6 conf — improves win rate from 31.5% to 57.6% (+26.0pp, n=33) SAME-WINDOW EFFECT: Over the current trade window, the filter would have blocked the USDCHF SHORT via rizzy5m entry at 0.81212 (CLOSED/WIN), removing a winning trade. It would not have affected any other trades since no other USDCHF SHORT was taken. The net PnL delta would be negative, as a confirmed winner is excluded, reducing overall win rate and profitability in this small sample. OVER-TIME PROJECTION: Over 3-6 months, the filter shows a strong improvement in win rate (+26pp) based on n=33, but the sample is small and may be period-specific. Risk includes overfitting to a narrow time window (NY_OPEN) and pair-specific behavior that may not persist. If the edge is real, it could reduce drawdowns and improve consistency, but the small sample size means a high chance of regression. HOW TO TEST: Run a shadow- Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com