LAB #2672 NEUTRAL: LEDGER BENCH: Apply filter: ETHUSD SHORT — improves win rate from 31.1% to 50.8% (+19.7pp, n=59) — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: ETHUSD SHORT — improves win rate from 31.1% to 50.8% (+19.7pp, n=59) SAME-WINDOW EFFECT: Over the current 10-trade window, applying the ETHUSD SHORT filter would have had **zero direct effect** — none of the listed trades are ETHUSD. The filter is symbol-specific, so it would not alter any GBPUSD, XAUUSD, AUDUSD, USDCHF, USDCAD, or USDJPY outcomes. The claimed +19.7pp win-rate improvement (31.1%→50.8%, n=59) is from a backtest on ETHUSD shorts only, not on the broader portfolio. Therefore, live PnL over this window would be unchanged — the filter is orthogonal to the trades shown. OVER-TIME PROJECTION: Over 3–6 months, if the filter is applied only to ETHUSD SHORT signals, the expected effect depends on whether the backtest edge is real or curve-fit. With n=59, the 95% confidence interval around a 50.8% win rate is roughly ±12.7pp (binomial), so the true win rate could be as low as ~38% — not clearl Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com