MODEL ADOPTION #999: REJECT — LEDGER BENCH: Apply filter: NZDUSD SHORT — improves win rate from 31.6% to 33.7% (+2.0pp, — E8 Intelligence Research

DECISION: REJECT ELEMENT: The NZDUSD SHORT filter is not a concrete, testable edge compatible with E8 geometry timing. It is a blunt instrument that removes all NZDUSD SHORT entries regardless of rhythm phase, polarity alignment, or Gods-Clock window. E8-node logic requires filters to be conditional on timing or polarity state, not static pair-direction bans. EXPECTED IMPACT: +2.0pp win rate on a sample of 98 is statistically fragile (CI ±3pp). Even if real, the filter reduces trade count and may cut profitable NZDUSD SHORT runs during favorable regimes, lowering net PnL. No evidence this edge is regime-independent or E8-compatible. RISK: Overfitting to a short sample; regime shift where NZDUSD SHORT becomes strongly profitable; model loses adaptability by hard-coding a pair-direction exclusion without timing context. TEST: Before any adoption, run a replay-bench on the last 6 months of E8 live data (not the original n=98 set). Require that the filter improves win rate by at le Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com