About Capitol Signals
Capitol Signals is a weekly publication that scores congressional stock trades against the legislation, contracts, and lobbying activity connected to each member's committee assignments. The scoring engine ingests public disclosures, cross-references them across five federal data sources, and produces a 0 to 100 composite suspiciousness rating for every flagged trade. Editorial briefs unpack the highest-scoring correlations each week.
Who built this
Capitol Signals is built and maintained by Brian Chaplow, a data scientist and retired Navy veteran. The project started as a capstone-adjacent experiment in correlating public financial disclosures with government activity and grew into a fully automated, open-source analytics pipeline.
Why this exists
The STOCK Act of 2012 requires members of Congress to disclose their stock trades within 45 days. Those filings are public, but scattered across multiple systems and difficult to cross-reference with the legislative and procurement activity that could explain suspicious timing. Capitol Signals closes that gap by joining five federal data sources into a single correlation engine that runs every night.
The goal is transparency, not accusation. A high score means multiple public signals aligned in time and topic. It does not prove wrongdoing. The data, methodology, and source code are all public so that journalists, researchers, and oversight bodies can verify the work and draw their own conclusions.
Data sources
| Source | Data | Update |
|---|---|---|
| Quiver Quantitative | STOCK Act trade disclosures | Daily |
| Congress.gov API | Bills, committee assignments, member rosters | Weekly |
| USAspending.gov | Federal contract awards | Daily |
| Senate LDA API | Lobbying disclosure filings | Quarterly |
| yfinance | Market prices, sector classification | Daily |
Contact
For corrections, methodology questions, or data inquiries: [email protected].
Source code and issue tracker: github.com/brianchaplow/Capitol_Signals.
License
All data, editorial briefs, and methodology artifacts published on this site are released under the CC0 1.0 Universal public-domain dedication. No paywall, no registration, no restrictions. Journalists, researchers, and civic-tech developers can redistribute the dataset, fork the scoring pipeline, or incorporate the weekly briefs into their own reporting without attribution.
Subscribe to the weekly brief
One brief per week. CC0 data drops. No tracking pixels. Powered by Buttondown.
Free. Double-opt-in. Buttondown handles delivery and unsubscribe.