LAB #4127 NEUTRAL: LEDGER BENCH: Apply filter: Phase distance MODERATE (5-10deg) — improves win rate from 32.5% to 32.8 — E8 Intelligence Research

Abstract

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Phase distance MODERATE (5-10deg) — improves win rate from 32.5% to 32.8% (+0.3pp, n=201) SAME-WINDOW EFFECT: Applying the MODERATE (5-10deg) phase-distance filter would have excluded trades falling outside that band. In the current window, the two rizzy5m SHORTs on NZDUSD (0.58252 WIN, 0.59416 LOSS) and the AUDUSD SHORT (0.71366 LOSS) are prime candidates for exclusion if their phase distance fell outside 5-10deg — but without per-trade phase data, the net effect is ambiguous. The backtest claims +0.3pp win-rate on n=201, which over this ~10-trade window translates to roughly +0.03 wins (statistically invisible). PnL delta would be negligible, likely within ±0.5R, and could easily flip negative if the filter accidentally removes one of the two WINs while keeping losses. OVER-TIME PROJECTION: Over 3-6 months, a +0.3pp win-rate improvement on a 32.5% base is marginal — roughly 1 extra win per 333 tra Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com