LAB #3987 NEUTRAL: LEDGER BENCH: Apply filter: USDCHF LONG — improves win rate from 32.5% to 38.7% (+6.3pp, n=142) — E8 Intelligence Research

Abstract

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: USDCHF LONG — improves win rate from 32.5% to 38.7% (+6.3pp, n=142) SAME-WINDOW EFFECT: Applying a USDCHF LONG-only filter would have blocked the single USDCHF LONG trade in the window (entry 0.80231, CLOSED/LOSS). That trade was a loss, so removing it would improve the window's raw win rate from 3/9 (33.3%) to 3/8 (37.5%) — a small positive shift, but not the +6.3pp claimed in the backtest. No other trades involve USDCHF, so the filter's live effect here is limited to deleting one losing trade; PnL impact is minor and directionally positive, but the sample is far too small to validate the backtest edge. OVER-TIME PROJECTION: Over 3–6 months, a USDCHF LONG-only filter could plausibly raise win rate if the backtest's 38.7% vs 32.5% is real and not overfit to a specific regime. However, the risk is high: USDCHF is a low-volatility, trend-reversing pair where LONG filters often just remove short-side l Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com