LAB #2815 NEUTRAL: LEDGER BENCH: Apply filter: Min confirmations >= 7 — improves win rate from 31.1% to 40.0% (+8.9pp, — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Min confirmations >= 7 — improves win rate from 31.1% to 40.0% (+8.9pp, n=10) SAME-WINDOW EFFECT: Over the 10 live trades listed, applying a minimum of 7 confirmations would have filtered out the three rizzy5m entries with lower confirmation counts (GBPUSD 1.34407, GBPUSD 1.34461, and XAUUSD 4048.74). Of those, two were wins and one was a loss, so the immediate win-rate would drop from 60% (6/10) to 50% (4/8) — not the +8.9pp improvement seen in your backtest. The PnL delta is roughly neutral to slightly negative here, because you'd be removing two winning trades and one losing trade, netting out near zero but with fewer total trades. OVER-TIME PROJECTION: Over 3–6 months, the filter likely reduces trade frequency by ~20–30% (since rizzy5m is your most active signal). The backtest's +8.9pp gain is based on n=10, which is statistically meaningless — the margin of error is ±30pp at 95% confidence. The Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com