LAB #3886 PROMISING: LEDGER BENCH: Apply filter: USDCHF LONG — improves win rate from 32.5% to 39.3% (+6.8pp, n=140) — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: USDCHF LONG — improves win rate from 32.5% to 39.3% (+6.8pp, n=140) SAME-WINDOW EFFECT: Applying a USDCHF LONG-only filter would have blocked the one live USDCHF trade in the window (rizzy7m LONG at 0.80231, which closed as a LOSS). That trade was a loser, so removing it would have improved the window's net PnL by that loss amount and raised the win rate from 4/10 to 4/9 (44.4% vs 40%) — a modest but real improvement. However, the filter would not affect any other trades, so the overall PnL delta is limited to avoiding that single loss, not a systemic shift. OVER-TIME PROJECTION: Over 3–6 months, a USDCHF LONG-only filter could meaningfully reduce exposure to a pair/strategy combination that historically underperforms on the short side. The backtest improvement (+6.8pp win rate, n=140) suggests a stable edge, but the sample is small and may be regime-dependent — if USDCHF trends shift, the filter co Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com