MODEL ADOPTION #2520: REJECT — LEDGER BENCH: Apply filter: Min confirmations >= 6 — improves win rate from 31.1% to 37.7% — E8 Intelligence Research

DECISION: REJECT ELEMENT: No single rule survives — the 6-confirmation filter is a statistical artifact, not an E8-geometry edge. EXPECTED IMPACT: Win-rate gain of +2–3pp out-of-sample at best, but trade frequency drops 20–30%, reducing absolute monthly PnL by 10–20% — net negative for profit. The +6.7pp is inflated by n=416 and a single favorable window (2 removed losses). RISK: Overfits to recent regime; filters out early-momentum winners that hit TP2 in trending markets; breaks E8 timing rhythm by delaying entries past the Gods-Clock confirmation window, degrading polarity alignment. TEST: Run a 90-day walk-forward on 3-month rolling windows (2019–2026) across all pairs, requiring: (1) win-rate improvement ≥ +3pp sustained in ≥70% of windows, (2) monthly PnL reduction ≤ 5%, (3) no increase in average time-to-TP2. If any window shows win-rate < 33% or PnL drop > 15%, reject permanently. Also shadow-test filtered-out signals live for 30 days — if their win rate > 40%, the filt Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com