MODEL ADOPTION #1938: REJECT — LEDGER BENCH: Apply filter: Min confirmations >= 6 — improves win rate from 31.9% to 42.0% — E8 Intelligence Research

DECISION: REJECT ELEMENT: No filter added — the "min confirmations >= 6" rule is not compatible with E8 geometry timing. E8-node entries are defined by the Gods-Clock rhythm; confirmations are a lagging indicator that breaks the FADE polarity timing edge. Rizzy5m low-confirmation entries are not "noise" — they are early-phase E8 reversals. Filtering them would systematically remove the highest-R breakout entries in trending regimes, which the backtest's 352-trade window cannot capture. EXPECTED IMPACT: Win rate would rise ~10pp in the tested window, but total R per month would likely drop 15–25% because the filter removes early entries that capture the full FADE move. The +10pp is a statistical artifact of a loss-heavy, choppy regime; in trending markets, the filter would cause late entries and missed reversals, degrading the model's core edge. RISK: Overfitting to a 352-trade sample. The filter is a volatility-regime proxy, not a timing rule. It would blind the model to E8's e Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com