MODEL ADOPTION #2838: REJECT — LEDGER BENCH: Apply filter: USDCHF SHORT — improves win rate from 31.2% to 59.1% (+27.9pp, — E8 Intelligence Research

DECISION: REJECT ELEMENT: No element survives — the USDCHF SHORT filter is a regime-dependent statistical artifact, not an E8-geometry-compatible edge. EXPECTED IMPACT: Win-rate gain is illusory in live context: only 1 qualifying trade in the current window (blocked, zero PnL delta). Over 3–6 months, if USDCHF SHORT frequency is low (as seen), the +27.9pp boost translates to <1% overall portfolio win-rate change, while reducing trade count and total PnL if the filter removes profitable CHF-weak trends. The 95% CI (±7.5pp) still leaves the true effect positive, but the edge is not tied to E8 timing or polarity — it's a currency-pair/direction bias that can flip with macro regime. RISK: Overfitting to a CHF-strength or USD-weakness regime. If CHF weakens (e.g., SNB intervention, risk-on), the filter blocks high-R:R shorts that E8 rhythm would have timed correctly. Hard rule also conflicts with E8's polarity logic, which already accounts for pair-specific momentum — adding a static Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com