MODEL ADOPTION #2148: REJECT — LEDGER BENCH: Apply filter: Min confirmations >= 6 — improves win rate from 32.0% to 42.1% — E8 Intelligence Research

DECISION: REJECT ELEMENT: No single rule survives — the min-confirmations filter is a statistical artifact, not an E8-geometry edge. EXPECTED IMPACT: Win rate may rise ~10pp in the backtest, but total R likely falls because the filter blocks early entries in fast moves (the exact trades that fund the fat right tail). The live window's improvement (3/5 vs 3/9) is noise on n=5 closed trades. RISK: Overfitting to a 361-trade sample; confirmation thresholds are microstructure-dependent (spread, news, liquidity) and will degrade when E8 rhythm shifts. Also, the filter does not align with Gods-Clock timing — it adds a lag that can misalign entries with the fade polarity window. TEST: Before any adoption, run a 90-day replay-bench on the live engine with confirmations 4,5,6,7,8. Require: (a) filtered (≥6) beats unfiltered on both win rate AND total R by ≥5% margin, (b) at least 200 filtered trades, (c) the edge persists across two distinct market regimes (trending and choppy). If any Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com