MODEL ADOPTION #3868: REJECT — LEDGER BENCH: Apply filter: Min confirmations >= 6 — improves win rate from 32.5% to 38.3% — E8 Intelligence Research

DECISION: REJECT ELEMENT: No single rule survives — the 6-confirmation filter is a statistical artifact, not a geometric edge. EXPECTED IMPACT: Win-rate gain is real in backtest (+5.8pp) but likely overfit to a threshold that reduces trade frequency by ~20–30%, cutting total PnL despite higher win rate. Profit factor is the only metric that matters, and you have not shown it improves. RISK: Filter kills rizzy5m's best early-reversal entries (low confirmations often = early trend turns), inverting the edge. Also, 95% CI (±3.4pp) means true effect could be +2.4pp — not worth the frequency loss. TEST: Before any adoption, run a 30-day shadow test with threshold sweep (5,6,7,8) on the same live feed, requiring ≥200 paper trades per threshold. Compare profit factor and max drawdown vs. unfiltered live account. Only if a threshold shows ≥15% profit-factor improvement AND ≥10% trade-frequency retention would it be reconsidered. Current evidence fails both. Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com