MODEL ADOPTION #1144: REJECT — LEDGER BENCH: Apply filter: Min confirmations >= 5 — improves win rate from 31.7% to 33.3% — E8 Intelligence Research

DECISION: REJECT ELEMENT: The filter "min confirmations >= 5" is not a concrete, testable edge compatible with E8 geometry timing — it is a statistical heuristic derived from a backtest with n=1329, not a rhythm or polarity rule. The 1.6pp win-rate improvement is marginal and likely noise, especially given the small sample and the fact that the live model already operates at ~50% win rate (vs backtest's 31.7%). The filter would reduce trade frequency without a demonstrated PnL improvement — the projection of "3 wins, 0 losses" is speculative and ignores the opportunity cost of rejected high-reward trades. EXPECTED IMPACT: Negligible or negative. Win rate may rise ~1-2pp, but trade count drops ~10-20%, and average PnL per trade could fall because low-confirmation trades in strong trends are often high-reward. Net PnL likely flat or slightly negative. RISK: The filter will reject high-reward entries during strong directional moves (e.g., EURUSD SHORT at 1.13594 was a valid fade that w Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com