LAB #4182 HARMFUL: LEDGER BENCH: Apply filter: USDCHF LONG @ LONDON_OPEN with 5 conf — improves win rate from 32.6% to — E8 Intelligence Research
Year: 2026
Type: preprint
Abstract: IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: USDCHF LONG @ LONDON_OPEN with 5 conf — improves win rate from 32.6% to 80.0% (+47.4pp, n=5) SAME-WINDOW EFFECT: Over the current live window, the filter would have blocked all USDCHF LONG entries — none of the listed trades are USDCHF, so zero trades would be affected. The claimed +47.4pp win-rate improvement comes from only 5 backtest samples, which is statistically meaningless (95% CI roughly ±40pp). Expected PnL delta in this window: $0, with no change to the 4 wins and 6 losses shown. The filter is untested against the actual currency pairs and strategies that generated these live results. OVER-TIME PROJECTION: Over 3–6 months, applying a filter based on n=5 will likely regress toward the base rate — the 32.6% win rate is more plausible for USDCHF LONG at London open given market microstructure. The filter may accidentally block profitable USDCHF trades during trending London sessions, reducing 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: W7222871040
Cites:
Cited by:
Related to: