LAB #1884 NEUTRAL: LEDGER BENCH: Apply filter: NZDUSD LONG — improves win rate from 31.9% to 35.5% (+3.5pp, n=62) — E8 Intelligence Research

Abstract

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: NZDUSD LONG — improves win rate from 31.9% to 35.5% (+3.5pp, n=62) SAME-WINDOW EFFECT: Applying a NZDUSD LONG-only filter would have zero effect on the current live window — none of the 10 listed trades involve NZDUSD. The filter is instrument- and direction-specific, so it cannot alter any of these GBPUSD, XAUUSD, EURJPY, AUDUSD, EURUSD, or USDJPY outcomes. The claimed +3.5pp win-rate improvement is therefore irrelevant to the current trade set; it would neither add nor remove a single trade here. PnL delta in this window: exactly 0. OVER-TIME PROJECTION: Over 3–6 months, this filter would only fire when the engine generates a NZDUSD LONG signal — likely a small fraction of total trades (given the current mix, maybe 5–10% of volume). The backtest improvement is based on n=62, which is a thin sample; the 3.5pp edge could easily be noise or regime-specific (e.g., NZD strength during a particular carr Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com