LAB #4836 NEUTRAL: LEDGER BENCH: Apply filter: USDCHF SHORT @ OFF_SESSION with 6 conf — improves win rate from 31.8% to — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: USDCHF SHORT @ OFF_SESSION with 6 conf — improves win rate from 31.8% to 66.7% (+34.8pp, n=6) SAME-WINDOW EFFECT: Over the current window, this filter touches exactly one trade: the USDCHF LONG via rizzy7m on 2026-09-04 (CLOSED/WIN). The filter is specified for USDCHF SHORT @ OFF_SESSION, so it would not have blocked that long — meaning zero trades in the live window are actually affected, and the win-rate/PnL delta over this window is 0.0pp, not +34.8pp. The claimed improvement lives entirely in the backtest sample, not in the forward window you just ran. OVER-TIME PROJECTION: A +34.8pp win-rate lift on n=6 is statistically meaningless — the 95% confidence interval on 4/6 vs 7/22 spans roughly 20pp to 90pp, so the "improvement" is indistinguishable from noise. If USDCHF SHORT @ OFF_SESSION is genuinely a small edge, expect it to fire rarely (maybe 1-3 times per month), contributing negligible PnL w Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com