LAB #4144 NEUTRAL: LEDGER BENCH: Apply filter: Quarterly phase = MANIPULATION — improves win rate from 32.5% to 33.4% ( — E8 Intelligence Research
Year: 2026
Type: preprint
Abstract: IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Quarterly phase = MANIPULATION — improves win rate from 32.5% to 33.4% (+0.9pp, n=1465) SAME-WINDOW EFFECT: Applying the MANIPULATION phase filter to the current live window would have blocked the two rizzy5m CLOSED/LOSS trades (AUDUSD 0.71366, XAUUSD 4598.79) and the rizzy5m REJECTED/open AUDUSD trade, while keeping the two rizzy5m WINs (NZDUSD 0.58252, GBPUSD 1.36238). That removes 2 losses and 1 open/rejected from the rizzy5m subset, converting the window's rizzy5m record from 2W/3L to 2W/0L — a clear PnL improvement. However, the filter would not touch wkjudas trades (all losses), so overall window win rate rises from ~20% (2/10) to ~29% (2/7) — still below the backtest's 33.4% claim, suggesting the edge is concentrated in one signal family. OVER-TIME PROJECTION: Over 3–6 months, a +0.9pp win-rate gain on n=1465 is statistically marginal (p≈0.15–0.20), meaning it could easily vanish or reverse u 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: W7222623265
Cites:
Cited by:
Related to: