LAB #3962 NEUTRAL: LEDGER BENCH: Apply filter: USDCHF LONG — improves win rate from 32.5% to 39.0% (+6.6pp, n=141) — 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 39.0% (+6.6pp, n=141) SAME-WINDOW EFFECT: Applying the USDCHF LONG filter would have blocked the single USDCHF LONG trade in the current window (rizzy7m entry at 0.80231, CLOSED/LOSS). That trade was a loss, so the filter would have removed one losing trade from the 10-trade sample, shifting the live win rate from 3/10 (30%) to 3/9 (33.3%) — a modest improvement, but not the +6.6pp claimed in backtest. The PnL delta is small (one loss avoided), and no other trades in the window are USDCHF LONG, so the effect is isolated and not statistically meaningful at n=9. OVER-TIME PROJECTION: Over 3–6 months, this filter could plausibly lift win rate by a few points if USDCHF LONG trades are systematically weak in your model — but the backtest sample (n=141) is thin, and the +6.6pp improvement may be noise or regime-specific. Risk: overfitting to a single pair/dire Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com