LAB #1802 PROMISING: LEDGER BENCH: Apply filter: Phase distance TIGHT (<=2deg) — improves win rate from 31.9% to 36.3% (+ — E8 Intelligence Research

Abstract

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Phase distance TIGHT (<=2deg) — improves win rate from 31.9% to 36.3% (+4.3pp, n=411) SAME-WINDOW EFFECT: Over the current live window, applying a phase-distance filter (≤2°) would have excluded the two XAUUSD SHORTs (entries at 4051.79 and 4066.35) and likely the AUDUSD SHORT — all three were losses. That removes 3 losing trades from 10, leaving 7 trades with 4 wins and 3 losses (including the open EURUSD), shifting win rate from ~40% to ~57% on this small sample. PnL delta is positive but modest — the filter avoids three full losses, though it also risks cutting future winners that happen to sit just outside the tight phase band. OVER-TIME PROJECTION: Over 3–6 months, the +4.3pp win-rate gain (n=411) is statistically meaningful but not transformative — it likely trades off frequency for precision, meaning fewer trades overall and lower gross PnL per unit time. The main risk is overfitting to a spe Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com