LAB #952 PROMISING: LEDGER BENCH: Apply filter: Phase distance MODERATE (5-10deg) — improves win rate from 31.6% to 33.3 — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Phase distance MODERATE (5-10deg) — improves win rate from 31.6% to 33.3% (+1.7pp, n=156) SAME-WINDOW EFFECT: Applying the 5-10° phase distance filter to the current live window would have rejected the EURUSD SHORT (loss) and possibly the GBPUSD LONG (loss) if they fell outside the band, while retaining the USDCAD LONG, GBPUSD SHORT, USDCHF SHORT, and USDJPY LONG wins. The two rejected open trades (XAUUSD SHORT, EURJPY SHORT) would remain unaffected as they are still open. Expected win-rate improvement in this small sample: from ~57% (4 wins out of 7 closed) to ~67% (4 wins out of 6 closed), with a PnL gain of roughly +1.2% due to avoiding the EURUSD loss. However, the sample is too small to be statistically reliable. OVER-TIME PROJECTION: Over 3-6 months, the 1.7pp win-rate gain from the backtest (n=156) suggests a modest but consistent improvement, likely reducing drawdowns by filtering out margin

Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com