MODEL ADOPTION #3886: REJECT — LEDGER BENCH: Apply filter: USDCHF LONG — improves win rate from 32.5% to 39.3% (+6.8pp, n — E8 Intelligence Research

DECISION: REJECT ELEMENT: No element survives — USDCHF LONG-only filter is a pair-direction bias, not an E8 geometry timing edge. It does not align with the model's rhythm logic and risks overfitting to a small sample (n=140, +6.8pp) that is regime-dependent. EXPECTED IMPACT: Win-rate lift is likely illusory outside the tested window; PnL delta is negligible (blocking one losing trade in live window). Over 3–6 months, could reduce exposure to a weak short side, but no evidence this is stable across USDCHF regimes or correlated pairs (EURUSD, GBPUSD). RISK: Overfitting to a single pair/direction; if USDCHF trends shift, filter becomes neutral or harmful. Also, it ignores the core E8-node/Gods-Clock timing — adding a non-geometric filter dilutes model coherence and may introduce hidden correlations that degrade future performance. TEST: Must pass a 12-month rolling 3-month window backtest (≥60 simulated trades, win-rate improvement ≥5pp, no increase in avg loss) AND a 30-day shad Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com