LAB #2036 NEUTRAL: LEDGER BENCH: Apply filter: NZDUSD LONG — improves win rate from 31.9% to 35.9% (+4.0pp, n=64) — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: NZDUSD LONG — improves win rate from 31.9% to 35.9% (+4.0pp, n=64) SAME-WINDOW EFFECT: Applying a NZDUSD LONG-only filter would have had zero direct impact on the 10 live trades listed, as none are NZDUSD. However, the filter would have excluded any hypothetical NZDUSD SHORT entries during that window — but since none occurred, the immediate PnL delta is nil. The +4.0pp win-rate improvement (n=64) is statistically weak (p≈0.35), so even if a few NZDUSD trades had been present, the expected PnL change would be within noise. No trade in the current window would have been altered, so the live effect is effectively neutral for this specific sample. OVER-TIME PROJECTION: Over 3–6 months, a LONG-only NZDUSD filter could reduce drawdowns if the pair's short-side losses are systematic, but the sample size (n=64) is too small to distinguish signal from noise. The risk is overfitting to a recent regime — Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com