MODEL ADOPTION #1849: REJECT — LEDGER BENCH: Apply filter: USDCHF SHORT @ ASIAN with 5 conf — improves win rate from 32.0 — E8 Intelligence Research

Q: What does this research mean for the field?

The proposed USDCHF SHORT @ ASIAN + 5 conf trading filter is a statistical artifact resulting from overfitting to a small sample size (n=19), and its implementation would reduce absolute returns rather than providing a genuine trading edge. Novelty: ClaimNovelty.INCREMENTAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

Andrew Stewart Caldin

doi:10.5281/zenodo.22324992

September 5, 2026Open Access

MODEL ADOPTION #1849: REJECT — LEDGER BENCH: Apply filter: USDCHF SHORT @ ASIAN with 5 conf — improves win rate from 32.0 — E8 Intelligence Research

View Full Paper

Ask AI

Bookmark

Share

Key Points

To evaluate whether applying an Asian-session confluence filter to USDCHF short positions provides a genuine statistical edge or constitutes an overfit artifact.
Analyzed the performance metrics of a session-filtered rule on a sample of historical USDCHF short trades (n=19).
Modeled expected 60-day return distributions, trade frequency reductions, and regression to the mean.
Outlined walk-forward validation protocols requiring a 12-month evaluation split by quarter with a minimum sample size threshold (≥30 filtered trades per quarter).
The filter was rejected as a statistical artifact on n=19, with the apparent win-rate improvement over 32.0% projected to regress to a true edge of only 0–10 percentage points.
Trade frequency decreased by ~40%, reducing expected 60-day net performance to a delta of −1.5R to +0.5R rather than the anticipated +2.5R gain.
Session-specific filtering introduced regime vulnerability to Asian liquidity shocks and contradicted the system's session-agnostic entry timing logic.

Implication

Quantitative evaluation demonstrates that session-based trade filtering induces overfitting in currency trading models, suggesting that apparent win-rate improvements reflect small-sample noise.

Abstract

DECISION: REJECT ELEMENT: No single rule survives — the USDCHF SHORT @ ASIAN + 5 conf filter is a statistical artifact on n=19, not an E8-geometry-timed edge. EXPECTED IMPACT: Win-rate gain is illusory; true edge likely 0–10pp after regression to mean. Filter would cut trade frequency by ~40% (USDCHF SHORT is rare), reducing absolute R despite higher win rate. Over 60 days, expect net PnL delta of −1.5R to +0.5R, not the projected +2.5R. RISK: Overfits to a recent SNB/dollar regime; Asian-session liquidity shifts (e.g., Tokyo intervention) could invert the edge. Also, blocking EURUSD SHORT (currently open) risks missing a winner — the filter is not session-agnostic and violates E8's rhythm logic, which already times entries by Gods-Clock, not by currency-pair session. TEST: Before any re-adoption, run a 12-month walk-forward on USDCHF SHORT with 5 conf, split by quarter, requiring ≥30 filtered trades per quarter. Additionally, replay the last 60 live days with the filter applie Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Andrew Stewart Caldin (2026) studied this question.

synapsesocial.com/papers/6a9e85d0c3034f961570e4cf https://doi.org/10.5281/zenodo.22324992

View Full Paper

Ask AI

Bookmark

Share

MODEL ADOPTION #1849: REJECT — LEDGER BENCH: Apply filter: USDCHF SHORT @ ASIAN with 5 conf — improves win rate from 32.0 — E8 Intelligence Research

Key Points

Implication

Abstract

Discussion

Cite This Study

MODEL ADOPTION #1849: REJECT — LEDGER BENCH: Apply filter: USDCHF SHORT @ ASIAN with 5 conf — improves win rate from 32.0 — E8 Intelligence Research

Key Points

Implication

Abstract

Discussion

Cite This Study

Keep reading

More related papers

Keep reading

More related papers