LAB #2710 NEUTRAL: LEDGER BENCH: Apply filter: USDCHF LONG @ NY_OPEN with 6 conf — improves win rate from 30.9% to 62.5 — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: USDCHF LONG @ NY_OPEN with 6 conf — improves win rate from 30.9% to 62.5% (+31.6pp, n=16) SAME-WINDOW EFFECT: Over the current live window, the filter would have blocked the USDCHF SHORT (rejected/open) and likely any USDCHF LONG that did not occur — since no USDCHF LONG is present, the filter would have zero direct impact on the 10 listed trades. The claimed +31.6pp win-rate improvement is based on n=16 backtest samples, which is statistically fragile (95% CI roughly ±24pp), so the real edge could be near zero. PnL delta in this window: unchanged, because no USDCHF LONG was triggered. OVER-TIME PROJECTION: Over 3–6 months, if the filter is real, it would only affect USDCHF LONG entries at NY_OPEN with conf=6 — a narrow slice of the book. The main risk is overfitting to a small sample: 16 trades is not enough to distinguish a genuine regime filter from noise, especially given the live engine already Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com