MODEL ADOPTION #3915: REJECT — LEDGER BENCH: Apply filter: Min confirmations >= 6 — improves win rate from 32.5% to 38.2% — E8 Intelligence Research

DECISION: REJECT ELEMENT: No single rule survives — the ≥6 confirmations filter is a blunt instrument that conflates noise-filtering with E8 geometry timing. It does not respect the distinct confirmation logic of `wkjudas` (which may require fewer confirmations by design) and risks suppressing high-momentum breakouts that only confirm late. The backtest gain (+5.7pp) is not isolated to a clean, strategy-specific mechanism. EXPECTED IMPACT: Win rate might rise 3–6pp, but total trade count drops ~15–20%, and net profit could fall if the removed trades include outsized winners (e.g., the GBPUSD SHORT win from 2026-08-26). The live window shift (37.5% → 40%) is within noise (n=8). No evidence this filter improves risk-adjusted return — only raw win rate, which is a misleading metric without R:R context. RISK: Global application would likely degrade `wkjudas` edge (different confirmation cadence). Also, the filter may block early entries in fast trends, causing missed breakouts that E Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com