LAB #4778 NEUTRAL: LEDGER BENCH: Apply filter: Phase distance TIGHT (<=2deg) — improves win rate from 31.8% to 36.3% (+ — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Phase distance TIGHT (<=2deg) — improves win rate from 31.8% to 36.3% (+4.5pp, n=644) SAME-WINDOW EFFECT: The filter would have removed a large share of the 10 live trades in the window, since most entries are rizzy5m/rizzy7m shorts clustered at similar times — phase distance is exactly the kind of correlated feature that fires across a basket. The two closed losses (XAUUSD, AUDUSD) and the USDCHF win are the only resolved outcomes; a +4.5pp win-rate lift on n=644 implies roughly 29 extra wins per 644, but on a 10-trade window the expected delta is ~0.45 trades — statistically indistinguishable from noise. Net PnL effect is likely small and could go either way, since the filter may have blocked the USDCHF winner as easily as the losers. OVER-TIME PROJECTION: Over 3–6 months, if the +4.5pp edge is real and stable, expect win rate to drift from ~31.8% to ~36.3% on filtered setups, with trade count dro Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com