LAB #1891 NEUTRAL: LEDGER BENCH: Apply filter: XAUUSD SHORT — improves win rate from 31.9% to 55.2% (+23.2pp, n=29) — E8 Intelligence Research
Year: 2026
Type: preprint
Abstract: IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: XAUUSD SHORT — improves win rate from 31.9% to 55.2% (+23.2pp, n=29) SAME-WINDOW EFFECT: Applying the XAUUSD SHORT filter would have removed the two XAUUSD SHORT trades in the live window (both closed as LOSSES at 4051.79 and 4066.35). That removes 2 losses from the 10 listed trades, shifting the window win rate from 5/10 (50%) to 5/8 (62.5%) — a +12.5pp improvement, though the sample is tiny. PnL delta is positive but modest, as the two removed losses likely offset a few wins; no wins are sacrificed since no XAUUSD SHORT in the window was a winner. The filter would not affect any other instrument or direction, so its immediate effect is purely loss-removal, not signal improvement. OVER-TIME PROJECTION: Over 3–6 months, the filter's claimed +23.2pp win-rate gain (n=29) is statistically weak — the confidence interval is wide (±18pp at 95% CI), and the effect could easily shrink to zero or reverse. Th 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: W7210271794
Cites:
Cited by:
Related to: