LAB #2951 NEUTRAL: LEDGER BENCH: Apply filter: USDCHF SHORT — improves win rate from 31.3% to 59.1% (+27.7pp, n=171) — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: USDCHF SHORT — improves win rate from 31.3% to 59.1% (+27.7pp, n=171) SAME-WINDOW EFFECT: Over the current live window, applying the USDCHF SHORT filter would have removed the one USDCHF SHORT trade (currently REJECTED/open) — no closed PnL impact. The remaining 9 trades are unaffected, so win rate stays at 5/9 (55.6%) vs. the unfiltered 5/9 — no delta. The filter's claimed +27.7pp win-rate improvement is not visible in this small sample; it only removes a single open position, so expected PnL delta is ~0. OVER-TIME PROJECTION: Over 3–6 months, the filter would eliminate all USDCHF SHORT signals. If the backtest's 59.1% win rate on that subset is real, you'd avoid a historically losing directional bias — but the n=171 is modest and likely regime-dependent (USDCHF short has been punished in recent CHF strength). Risk: the filter may overfit to a specific volatility/trend regime; if CHF weakens, you'l Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com