MODEL ADOPTION #4626: REJECT — LEDGER BENCH: Apply filter: Min confirmations >= 6 — improves win rate from 32.2% to 37.3% — E8 Intelligence Research

DECISION: REJECT ELEMENT: No single rule survives — the ≥6 confirmation filter is a statistical artifact, not a geometric edge. E8 timing already encodes confirmation thresholds via Gods-Clock phase alignment; adding a fixed count filter would distort the FADE polarity entry window and reduce trade frequency without a proven PnL-per-trade gain. EXPECTED IMPACT: Win rate might rise 2–4pp in backtest, but absolute profit likely falls 15–25% due to fewer trades, and the unresolved three trades (USDCHF, EURJPY, AUDUSD) could easily flip the live delta negative if they win. No evidence that filtered-out early entries have worse R-multiple or MAE in live conditions — only a win-rate shift from a small n=871 sample. RISK: Overfitting to a regime with many false early signals; confirmation counts are broker-feed-latency dependent, so live vs backtest parity is unverified; rejecting fast breakouts that never reach 6 confirmations could miss the exact momentum spikes E8 FADE polarity is de

Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com