Epsilon reference note

This page belongs to a controlled crawl-diff experiment on PortfolioHQ. It carries ordinary prose so that a crawler measuring overlapping word sequences has something real to compare between two runs.

Concentration is the risk that hides most easily. A portfolio can look balanced by count while two or three positions carry most of the value, and sector exposure can drift a long way from where it started without any single decision having felt wrong. Measuring allocation on every refresh is what turns that drift into something visible.

Back to the lab index