SageFusion
  • Research access
  • Firms
  • Insights
  • Podcast
  • The Wealth Almanac
  • About
  • Contact Us
  • Login

Does copying Congress stock trades work? What SageFusion's test of U.S. House trades found

10/8/2026

0 Comments

 
Does copying Congress stock trades work? U.S. House trades, 2018 to 2026, compared with comparable stocks. Label: Verified (no edge). SageFusion by Investment Science.

By Michael Kelly | Published October 9, 2026 | Updated October 9, 2026

Short answer: No, not in SageFusion's test. Copying the disclosed stock trades of U.S. House members, from the day each filing became public, did not beat comparable stocks by any reliable margin from 2018 to 2026. The research log labels the study Verified (no edge). Senate trades were not part of the test.

Key points:

  • SageFusion tested every U.S. House periodic transaction report from 2018 to 2026, about 21,850 stock trades by 205 members (SageFusion research log, October 2026).
  • Copying those trades from the day they were made public showed no reliable gain over comparable stocks in either period, and the study is labeled Verified (no edge) (recorded October 6, 2026).
  • Senate trades were not tested, and the study uses today's listed stocks, so companies that disappeared are missing (SageFusion research log, October 2026).

Read more: https://www.sagefusion.co/does-copying-congress-trades-work.html

How do members of Congress disclose their stock trades?

The STOCK Act of 2012 requires members of Congress to report many securities trades in periodic transaction reports. A report can come weeks after the trade itself, and the deadline can be as late as 45 days from the trade. House reports are published by the Office of the Clerk.

Reports show a range for the size of each trade, not an exact amount. By the time a tracker shows a trade, the price has often already moved.

What did SageFusion test?

SageFusion read every U.S. House periodic transaction report from 2018 to 2026: about 5,976 filings covering about 21,850 stock trades by 205 members, buys and sales.

Each trade was followed from the day its filing became public, which is the earliest anyone else could act on it, and compared with the average of stocks with similar volatility and similar recent performance on the same day.

The result was checked in two separate periods, 2018 to 2021 and 2022 to 2026.

Did copying U.S. House trades do better than comparable stocks?

No. In both periods, the trades U.S. House members made, buys or sales, showed no reliable gain over comparable stocks once the filings were public. Some results were a little ahead and some behind, and none stood out from chance. The study is labeled Verified (no edge) in the SageFusion research log (recorded October 6, 2026).

Larger buys, and stocks bought by two or more members in the same month, also showed no reliable gain. Stocks members sold tended to rise about as much as the ones they bought, which looks more like members holding popular stocks than like inside information.

Why might following disclosed trades not help?

Delay: weeks can pass between the trade and the public filing. Mixed motives: many trades are routine, such as rebalancing, tax moves or managed accounts. Crowding: popular trackers mean many people see the same filing at once. Broad holdings: members often own the same large companies as everyone else.

None of this is a comment on any member, party or policy. The test only asks whether following the filings after they were public would have helped.

What are the limits of this test?

U.S. House only: Senate filings are not in this study yet, so the result says nothing about the Senate. The study uses prices of stocks that are listed today, so companies that disappeared are missing, which can flatter results. It is a test of the past, not a forecast.

Questions about copying Congress stock trades

Does copying Congress stock trades work?

Not in SageFusion's test of U.S. House trades. Following every House periodic transaction report from 2018 to 2026, from the day each filing became public, showed no reliable gain over comparable stocks, in both an earlier and a later period. The study is labeled Verified (no edge). Senate trades were not tested.

Can I legally copy Congress stock trades?

The periodic transaction reports are public records that anyone can read on the House Clerk's website, and many trackers repackage them. SageFusion's study only asks whether following those filings after they became public would have helped, and in this test it did not. Nothing here is legal advice or investment advice.

Is SageFusion investment advice?

No. SageFusion is a research tool, not investment advice. It shows how ideas held up in past data, including the ones that failed, like copying disclosed U.S. House trades. It does not tell anyone what to buy or sell. Talk to a licensed professional before making investment decisions.

Sources: U.S. House of Representatives, Office of the Clerk, Financial Disclosure Reports, https://disclosures-clerk.house.gov/; Congress.gov, STOCK Act (Public Law 112-105), https://www.congress.gov/bill/112th-congress/senate-bill/2038; SageFusion, Does copying Congress trades work?, https://www.sagefusion.co/does-copying-congress-trades-work.html; SageFusion, Research log (label as of October 6, 2026), https://www.sagefusion.co/research.html

About the author: Michael Kelly is the founder of Investment Science and SageFusion, and the author of The Wealth Almanac, a #1 Amazon Best Seller in Wealth Management, Economic History and Investing Portfolio Management (Kindle, October 2025). He has more than a decade in banking and financial technology, with roles at Citigroup, Barclays, UBS, and Mizuho Securities. He is an Adjunct Instructor at Hofstra University's Frank G. Zarb School of Business, and holds a Stonier Diploma from the ABA Stonier Graduate School of Banking and a Wharton Leadership Certificate.

Research tool, not investment advice. Studies use daily prices of today's listed stocks unless stated.

0 Comments

What did public companies tell the SEC from October 5 to October 8, 2026? A plain-English guide

10/8/2026

0 Comments

 
What companies told the SEC, October 5 to October 8, 2026: a calendar range and Form 144, Schedule 13G and Form 8-K filings. SageFusion by Investment Science.

By Michael Kelly | Published October 9, 2026 | Updated October 9, 2026

Short answer: From October 5 to October 8, 2026, SageFusion logged 911 notable SEC filings from 587 public companies it tracks. The most common were planned insider sales (Form 144), new large passive shareholders (Schedule 13G) and director and officer changes (Form 8-K, Item 5.02). A filing is a disclosure, not a forecast.

Key points:

  • SageFusion logged 911 notable SEC filings from 587 public companies from October 5 to October 8, 2026 (SEC EDGAR).
  • The most common kind was planned insider sales (Form 144), with 264 filings in that span (SEC EDGAR, October 8, 2026).
  • In SageFusion's research, the test of planned insider sales (Form 144) against random stocks is labeled Tested, inconclusive (SageFusion research, October 8, 2026).

What did companies file this week?

From October 5, 2026 to October 8, 2026, SageFusion logged 911 notable filings from 587 public companies on SEC EDGAR, across 32 kinds of events. The most common kinds were:

  • Planned insider sale (Form 144): 264 filings
  • New large passive shareholder (Schedule 13G): 108 filings
  • Director or officer change (Form 8-K, Item 5.02): 83 filings
  • Signed a material agreement (Form 8-K, Item 1.01): 78 filings
  • Chief financial officer departure: 56 filings
  • Activist shareholder updated its Schedule 13D: 50 filings

What do these filings mean in plain English?

A Form 144 is a notice that a company insider plans to sell shares. Many of these sales are pre-set plans or shares sold to cover taxes on stock pay.

A Schedule 13G is filed by a large shareholder that says it is a passive investor. A Schedule 13D is filed by a large shareholder that may try to influence the company.

A Form 8-K is a report of a major event. Item 5.02 covers directors and officers joining or leaving, and Item 1.01 covers signing a major contract.

Does an insider sale mean the stock will fall?

Not on its own. SageFusion compared stocks after these filings with random stocks picked the same day, using years of filings, and the research labels are below.

  • Planned insider sale (Form 144): Tested, inconclusive (most of the history is recent, and the result was too small to call)
  • New large passive shareholder (Schedule 13G): Verified (no edge)
  • New large shareholder that may seek influence (a first Schedule 13D, not later updates): Verified (no edge)

Verified (no edge) means stocks after that filing did no better than random stocks picked the same day, both before 2016 and from 2016 on, after trading costs. Tested, inconclusive means the result was too small or too recent to call either way.

Where can I read these filings myself?

Every filing is public and free on SEC EDGAR full-text search. Search a company name or a form type such as 144, 13G, 13D or 8-K.

What are the limits of this data?

These summaries are produced automatically from the SEC full-text search index for the public companies SageFusion tracks, so they are a sample of everything filed, and they can miss or mislabel a filing. The tests use prices of companies that are listed today, so companies that disappeared are missing. That can hide some of the worst outcomes after warning-sign filings such as insider sales. Always read the original filing before relying on it.

Questions about this week's SEC filings

What is a Form 144 filing?

A Form 144 is a notice filed with the SEC when a company insider or affiliate plans to sell restricted or control shares. Many are routine sales under pre-set trading plans or to cover taxes on stock pay. The filing says a sale is planned. It does not say why, and it is not a forecast for the stock.

What is the difference between a Schedule 13G and a Schedule 13D?

Both are filed by investors that own a large stake in a public company. A Schedule 13G is for passive holders, such as index funds, that do not plan to influence the company. A Schedule 13D is for holders that may try to influence the company, for example by seeking board seats, so it often draws more attention.

Is the SageFusion filing summary investment advice?

No. SageFusion is a research tool, not investment advice. The filing summary restates public SEC disclosures in plain English and counts them. It does not say whether to buy or sell anything. Read the original filing on SEC EDGAR and talk to a licensed professional before making investment decisions.

Sources: U.S. Securities and Exchange Commission, EDGAR full-text search (filings dated October 5, 2026 to October 8, 2026), https://www.sec.gov/edgar/search/; U.S. Securities and Exchange Commission, Form 8-K, https://www.sec.gov/answers/form8k.htm; SageFusion, filing summaries (as of October 9, 2026) and research labels (as of October 8, 2026), https://www.sagefusion.co/research.html

About the author: Michael Kelly is the founder of Investment Science and SageFusion, and the author of The Wealth Almanac, a #1 Amazon Best Seller in Wealth Management, Economic History and Investing Portfolio Management (Kindle, October 2025). He has more than a decade in banking and financial technology, with roles at Citigroup, Barclays, UBS, and Mizuho Securities. He is an Adjunct Instructor at Hofstra University's Frank G. Zarb School of Business, and holds a Stonier Diploma from the ABA Stonier Graduate School of Banking and a Wharton Leadership Certificate.

Research tool, not investment advice. Studies use daily prices of today's listed stocks unless stated.

0 Comments

How does SageFusion test its picks against random stocks, and what failed?

10/8/2026

0 Comments

 
Our picks vs random stocks: a ranked list of picks beside a field of randomly scattered stocks. SageFusion by Investment Science.

By Michael Kelly | Published October 9, 2026 | Updated October 9, 2026

Short answer: SageFusion compares its nightly picks with random stocks picked the same night, under the same entry and exit rules, and publishes the ideas that fail. In a three-year replay (a backtest rebuilt after the fact, not a live record), SageFusion's own picks did not beat random stocks, and that result sits in the public research log.

Key points:

  • SageFusion's nightly picks are compared with random stocks picked the same night, under the same entry and exit rules (SageFusion research log, October 2026).
  • In the SageFusion research log's three-year replay, a backtest rebuilt after the fact, the nightly picks did not beat random stocks (recorded October 7, 2026).
  • Ideas such as moon phases and stock trades by members of the U.S. House are labeled Verified (no edge) in the SageFusion research log (October 2026).

What is the random-pick test?

The random-pick test asks one plain question: would a random stock, picked the same night and handled under the same entry and exit rules, have done just as well? SageFusion runs this comparison on its nightly picks and on its chart setups.

If an idea cannot do better than chance, it is not a skill. It is luck, and luck does not repeat on schedule.

Did SageFusion's own picks beat random stocks?

In a three-year replay recorded in the research log, SageFusion's own nightly picks did not beat random stocks picked the same night under the same entry and exit rules. A replay is a backtest rebuilt after the fact, not a live record, and this one has known flaws: the ranking used odds estimated partly on the same years, and companies that later disappeared are missing. Those flaws, if anything, help the picks, and they still did not beat random. A live record, logged in real time, started in October 2026.

That result is published on purpose. A research tool that hides its own misses is asking for blind trust, and SageFusion would rather show the work.

What is the failure log?

The failure log is the part of the SageFusion research log that records ideas that did not hold up. Every study gets one of a few labels.

Verified means the result held up on data from 2005 to 2015 and again on data from 2016 to 2026 that the test never saw, after trading costs. Verified (no edge) means the idea was tested on an earlier period and again on a later period the test never saw, and it did not beat chance or a matched comparison. A result like that is hard to rescue: the known limits of these tests, such as missing delisted companies, tend to flatter ideas, not hide them. Tested, inconclusive means the test was too limited to call either way. Failed or retracted means the idea broke down, or an earlier result was found to be wrong and taken back.

Which popular ideas did not hold up?

A few examples from the SageFusion research log, with the label each one carries today:

  • Moon phases and stock prices: Verified (no edge)
  • Sudden spikes in Google searches for a stock: Tested, inconclusive (large companies only, five years of data, no edge found)
  • Stock trades disclosed by members of the U.S. House: Verified (no edge)
  • The tone of news coverage about a company: Verified (no edge)
  • SageFusion's own older score dashboards: Verified (no edge)

Some of these were Michael Kelly's own ideas from years ago. They are in the log anyway.

Why publish failures at all?

Most trading claims are never compared with chance, so readers cannot tell skill from luck. Publishing the misses next to the results gives readers a way to judge the whole record, not only the highlights.

It is also the idea behind The Wealth Almanac: look at what usually happened before trusting a story about what will happen next.

What are the limits of this research?

The studies use daily prices of stocks that are listed today, so companies that disappeared are missing, which can flatter long-only results. The three-year replay of the nightly picks has the same limit, and its ranking also used odds estimated partly on the same years. Sector and size labels are today's labels applied to past dates. A rerun on an archive that includes delisted companies is still on the to-do list.

Questions about the random-pick test and failure log

What does it mean when SageFusion says an idea has no edge?

It means the idea was tested on years of past data, in an earlier period and again in a later one, and did not do better than chance or a matched comparison. The label is Verified (no edge). It can change if a better test or new data changes the answer. It is not a judgment on the person who had the idea, only on whether the idea helps.

Is SageFusion investment advice?

No. SageFusion is a research tool, not investment advice. It shows how ideas held up in past data, including the ones that failed. It does not tell anyone what to buy or sell, and it does not know anyone's goals, taxes or risk tolerance. Talk to a licensed professional before making investment decisions.

Where can I see the SageFusion research log?

A summary of the research and the waitlist for full access are at sagefusion.co/research.html. The log lists each study with its label, including Verified, Verified (no edge), and Failed or retracted. New studies are added as they are finished, and labels change when a recheck changes the answer.

Sources: SageFusion, Research log (labels as of October 7, 2026), https://www.sagefusion.co/research.html; Michael Kelly, The Wealth Almanac, Amazon Kindle, https://www.amazon.com/dp/B0FW7V8PJC

About the author: Michael Kelly is the founder of Investment Science and SageFusion, and the author of The Wealth Almanac, a #1 Amazon Best Seller in Wealth Management, Economic History and Investing Portfolio Management (Kindle, October 2025). He has more than a decade in banking and financial technology, with roles at Citigroup, Barclays, UBS, and Mizuho Securities. He is an Adjunct Instructor at Hofstra University's Frank G. Zarb School of Business, and holds a Stonier Diploma from the ABA Stonier Graduate School of Banking and a Wharton Leadership Certificate.

Research tool, not investment advice. Studies use daily prices of today's listed stocks unless stated.

0 Comments

    Author

    Michael Kelly is the founder of Investment Science and SageFusion and the author of The Wealth Almanac, a #1 Amazon Best Seller in Wealth Management, Economic History and Investing Portfolio Management (Kindle, October 2025). He has more than a decade in banking and financial technology, with roles at Citigroup, Barclays, UBS, and Mizuho Securities, and is an Adjunct Instructor at Hofstra University. His experience spans algorithmic trading, quantitative finance, hedge funds, private equity, and machine learning. This page is intended to educate others on the capabilities of SageFusion.

    Archives

    October 2026
    January 2025
    July 2024

    Categories

    All
    Data Analysis
    Economic Insights
    Financial Analysis
    Investment Strategies
    Investment Tools
    Market Crises
    Portfolio Management
    Stress Testing

    RSS Feed

  • Research access
  • Firms
  • Insights
  • Podcast
  • The Wealth Almanac
  • About
  • Contact Us
  • Login