LAB #4185 HARMFUL: LEDGER BENCH: Apply filter: XAUUSD SHORT @ NY_OPEN with 6 conf — improves win rate from 32.6% to 80. — E8 Intelligence Research
Year: 2026
Type: preprint
Abstract: IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: XAUUSD SHORT @ NY_OPEN with 6 conf — improves win rate from 32.6% to 80.0% (+47.4pp, n=5) SAME-WINDOW EFFECT: Over the live window shown, this filter would have blocked the XAUUSD SHORT rizzy5m entry at 4598.79 (2026-08-27 09:15) — which was a CLOSED/LOSS. No other XAUUSD trades appear in the window, so the only direct effect is removing one losing trade. That would shift the live record from 3W/4L to 3W/3L, improving win rate from ~42.9% to 50.0% — but that's a single-trade artifact, not a systemic edge. The claimed +47.4pp on n=5 is statistically meaningless (95% CI roughly ±40pp), so the apparent boost is likely noise. OVER-TIME PROJECTION: Over 3–6 months, this filter would cut XAUUSD SHORT entries at NY_OPEN with 6 conf — a narrow slice of the trade universe. If the true edge is zero, you'd expect regression to ~35–45% win rate, not 80%. The risk is overfitting to a tiny sample: 5 trades cannot Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com
Author: Caldin, Andrew Stewart
Language: English
Subfield: Finance
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7222897240
Cites:
Cited by:
Related to: