MODEL ADOPTION #437: REJECT — LEDGER BENCH: Apply filter: USDCHF SHORT — improves win rate from 32.4% to 62.7% (+30.4pp, — E8 Intelligence Research

DECISION: REJECT ELEMENT: The filter "USDCHF SHORT only" is a symbol-direction bias, not a timing or geometry edge. It cannot be cleanly grafted onto E8-node/Gods-Clock rhythm logic because it ignores the phase-locked entry conditions that define the model's core edge. The claimed +30.4pp win rate is likely a backtest artifact from a period where USDCHF happened to trend favorably for shorts — no evidence this persists across regimes. EXPECTED IMPACT: Win rate may temporarily rise, but total trade count drops sharply (only USDCHF SHORT signals survive). Net PnL likely degrades because the model loses diversification and misses profitable longs or other pairs that E8 timing would otherwise capture. The one live trade blocked (rizzy5m, winner) confirms immediate PnL loss. RISK: Overfitting to a 161-sample window; regime shift in USDCHF (e.g., sudden bullish reversal) would turn the filter into a loss magnet. Also introduces symbol-specific dependency that violates the model's pair- Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com