LAB #2950 NEUTRAL: LEDGER BENCH: Apply filter: ETHUSD SHORT — improves win rate from 31.3% to 54.0% (+22.7pp, n=63) — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: ETHUSD SHORT — improves win rate from 31.3% to 54.0% (+22.7pp, n=63) SAME-WINDOW EFFECT: Applying the ETHUSD SHORT filter would have zero effect on the current live window — none of the 10 trades listed are ETHUSD. The claimed +22.7pp win-rate improvement (n=63) is not represented in this sample, so no PnL delta would occur here. The filter is orthogonal to the current book, meaning it neither helps nor hurts these specific positions. Any benefit would only materialize if ETHUSD SHORT signals were present in future windows. OVER-TIME PROJECTION: Over 3–6 months, the filter could meaningfully lift overall win rate if the 54% vs 31.3% split holds out-of-sample. However, n=63 is small, and the improvement may be curve-fit to a specific regime (e.g., ETHUSD trending down in that period). Risk: the filter could suppress valid long-side or counter-trend SHORT setups that were profitable, reducing trade fr Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com