MODEL ADOPTION #299: REJECT — LEDGER BENCH: Apply filter: USDCHF SHORT @ NY_OPEN with 5 conf — improves win rate from 32 — E8 Intelligence Research

DECISION: REJECT ELEMENT: The filter "USDCHF SHORT @ NY_OPEN with 5 conf" is not a precise, testable edge — it conflates a time window (NY_OPEN) with a confidence threshold (5) that is undefined in the E8 rhythm logic. "5 conf" is not a documented E8-node/Gods-Clock parameter. The sample (n=67) is too small to distinguish signal from noise, and the 32.3% baseline suggests the unfiltered strategy is already broken for this pair — fixing a broken leg with a narrow filter is not an edge, it's overfitting. EXPECTED IMPACT: Win rate may appear to rise in the short term, but the filter will likely reduce trade frequency by >50% and miss the majority of USDCHF SHORT opportunities that occur outside NY_OPEN. Net profit impact is likely neutral or negative due to missed trends and reduced sample diversity. RISK: Overfitting to a small, recent window. The filter will cause the model to ignore valid E8 polarity signals during London, Asian, or late US sessions, degrading overall model coherenc Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com