LAB #971 PROMISING: LEDGER BENCH: Apply filter: Phase distance FAR (>10deg) — improves win rate from 31.6% to 32.3% (+0. — E8 Intelligence Research

Abstract

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Phase distance FAR (>10deg) — improves win rate from 31.6% to 32.3% (+0.7pp, n=669) SAME-WINDOW EFFECT: Applying the phase distance FAR (>10deg) filter to the current live window would have rejected the EURUSD SHORT (loss) and potentially the GBPUSD LONG (loss) if their phase distances were ≤10deg, while keeping the USDCAD LONG, GBPUSD SHORT, USDCHF SHORT, and USDJPY LONG (all wins). This would shift the live window from 4 wins, 2 losses, 4 open/rejected to 4 wins, 0 losses, 4 open/rejected, improving win rate from ~67% to 100% on closed trades. PnL delta would be positive, avoiding two losses and preserving gains, though the small sample (n=6 closed) limits statistical significance. OVER-TIME PROJECTION: Over 3-6 months, the +0.7pp win-rate improvement (from 31.6% to 32.3%) suggests a modest but consistent edge, likely reducing low-probability entries that degrade performance. Risk includes overfit Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com