Strategies · Core
Lexical Density of Company Filings
Score filings on information density (less boilerplate, more content words) and sort the cross-section on that score.
Definition
Lexical density ≈ content words / total words (or a modern embedding analog). Long high-density, short low-density, with sector controls. The economic story is disclosure quality or complexity as a risk.
Why it matters
It is another filing-based alternative signal. Easy to proxy a sector or a law-firm template.
Case
A sector that adopted a new MD&A template. Density moved together. The LS residual was zero.
How to read it
Control industry and size. Then a hold-out after LLM-written filings became common.
No query has been run yet. Which is tragically normal for most knowledge systems, but we are trying to evolve past decorative databases.