LAB #923 PROMISING: LEDGER BENCH: Apply filter: USDCHF LONG @ LONDON_OPEN with 6 conf — improves win rate from 31.7% to — E8 Intelligence Research

Abstract

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: USDCHF LONG @ LONDON_OPEN with 6 conf — improves win rate from 31.7% to 61.1% (+29.5pp, n=18) SAME-WINDOW EFFECT: Over the current trade window, no USDCHF LONG @ LONDON_OPEN trades with 6 conf appear in the live record — the only USDCHF trade is a SHORT via rizzy5m (win). Applying this filter would have added zero new trades or altered any outcomes. The win-rate improvement from 31.7% to 61.1% is based on n=18, which is too small to be statistically reliable — a single loss could swing the rate by ~5.6pp. The PnL delta would be negligible in this window. OVER-TIME PROJECTION: Over 3-6 months, if the 61.1% win rate holds with larger sample (n>100), this filter could meaningfully improve overall strategy profitability by reducing drawdowns from low-conviction USDCHF longs. However, the risk is severe: the 31.7% baseline suggests the underlying signal is weak, and the 6-conf filter may be overfitting t Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com