Public summary
A small cluster of near-identical trade outcomes could push a risk-adjusted score calculation into an effectively-zero denominator, producing a wild, meaningless spike instead of a real reading. That single degenerate reading then got averaged into a broader out-of-sample summary, corrupting it. Root-caused, fixed, regression test added. The same iteration caught a second variant with the same underlying cause: too few, too-similar trade outcomes to make the score meaningful. The calculation now treats any near-constant outcome series as undefined rather than computing a number from it - a public report safety guard, not a performance claim.
Public record
- Record type, publication date, and public-safe summary.
- Aggregate evidence posture and operating boundary.
- Whether the record belongs to research, runtime, safety, evidence, governance, or architecture.
Protected review
- System names, market-specific routing, and implementation notes.
- Protected dashboards, sensitive records, and control actions.
- Detailed operator decision notes that could expose live decision mechanics.