MODEL ADOPTION #971: REJECT — LEDGER BENCH: Apply filter: Phase distance FAR (>10deg) — improves win rate from 31.6% to — E8 Intelligence Research

DECISION: REJECT ELEMENT: The +0.7pp win-rate improvement (31.6% → 32.3%) is statistically fragile (n=669, delta ~2 trades), and the live window "100% win rate" projection is a cherry-picked artifact of 6 closed trades. The filter is not a concrete, testable edge — it is a distance threshold that has not been validated against E8 phase geometry timing or shown to survive out-of-sample. EXPECTED IMPACT: Negligible or negative — the filter would reduce trade frequency by an unknown amount (likely >30%), and the win-rate gain is too small to offset lost opportunity cost. In trending markets where phase distance compresses, this filter would reject high-reward entries. RISK: Overfitting to a single backtest window; reduced trade volume degrades model throughput and total PnL; the filter may interact destructively with the existing FADE polarity edge, which already accounts for phase geometry. TEST: Before any adoption, run a 1,000-signal shadow-test comparing filtered vs. unfiltere Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com