LAB #4899 NEUTRAL: LEDGER BENCH: Apply filter: USDCHF SHORT — improves win rate from 31.9% to 49.3% (+17.5pp, n=304) — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: USDCHF SHORT — improves win rate from 31.9% to 49.3% (+17.5pp, n=304) SAME-WINDOW EFFECT: The filter only touches USDCHF SHORT signals, and the live window contains zero USDCHF SHORT trades — the two USDCHF entries are both LONG (one closed win, one rejected/open). So the applied filter would have changed nothing in this window: same trades, same losses on EURUSD/XAUUSD/AUDUSD shorts, same USDCHF LONG win. The +17.5pp claim is therefore untested by the current live sample and cannot be credited or debited here. Note the window's actual damage came from short-side FX/metal trades, which this filter does not address. OVER-TIME PROJECTION: If the 49.3% vs 31.9% split holds out-of-sample on n=304, USDCHF SHORT is a genuine regime/edge asymmetry and the filter is worth ~+17pp on that subset — but USDCHF SHORT is a small slice of total flow, so portfolio-level impact is modest unless USDCHF shorts are fre Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com