Following Congress Trades: Can You Build an Edge From STOCK Act Filings?
Following Congress trades means mirroring the stock transactions that U.S. lawmakers are legally required to disclose under the STOCK Act. The appeal is obvious — these are well-connected people — and the data is free and public. But there is a structural catch: you only learn about a trade after a reporting delay, so you are never trading on the same information at the same time. Whether the disclosed flow still predicts anything is a testable question, and TradeBricks lets you test it for free.
What following Congress trades is
The STOCK Act requires members of Congress to publicly file their securities transactions, generally within 45 days. These Periodic Transaction Reports (PTRs) are published by the Clerk of the House and the Senate, listing the ticker, direction (buy/sell), and a dollar range.
A Congress-follow strategy turns those filings into trade signals: when a member (or any member) discloses a purchase in a stock, go long; on disclosed sells, exit or go short. You can filter by specific members, chamber, or a minimum dollar size. It is an alternative-data strategy — the signal comes from disclosures, not price.
- Source: STOCK Act Periodic Transaction Reports (PTRs), public and free
- Signal: a disclosed buy or sell in a ticker (with a dollar range)
- Filters: by member, chamber (house/senate), and minimum trade size
- Key constraint: filings appear after a reporting lag of up to ~45 days
- Family: alternative-data / event-driven
When it tends to work
The case for following Congress is informational: some members may have access to, or simply good intuition about, sectors they oversee, and concentrated disclosed buying in a name can occasionally precede a sustained move. Following a small set of historically active, concentrated traders rather than the whole body can sharpen the signal.
It works best as a slower, position-style overlay — holding for weeks to months — because the disclosure lag makes it useless for short-term timing. Combining a disclosed-buy signal with a price or trend confirmation (only act if price is also holding up) tends to filter out stale or already-played-out filings.
When it fails
The fundamental limitation is the lag. By the time a trade is disclosed, the news that motivated it may be weeks old and largely priced in; you are following a cold trail. Many disclosed trades are also routine portfolio management, index funds, or spouse-directed accounts with no informational content.
The dollar amounts are reported only in wide ranges, so you cannot size precisely, and aggregate Congressional trading has not reliably beaten the market once the lag is accounted for. As with any signal, it is easy to find a member or filter that looked great in hindsight and overfit to it. Out-of-sample testing is the check.
How to build it in TradeBricks bricks
TradeBricks has a native, free Congress feed — no broker or API key. The brick reads the official House Clerk PTR filings and aligns each to its public filing date, so there is no look-ahead.
A basic follow-the-buys strategy:
- data_source — the stock you want to trade (e.g. a daily feed); sets the traded symbol and outputs price/ATR.
- congress_trades — leave watch blank to track disclosures for the stock you are trading, or set a specific ticker; choose chamber, an optional member or list of members, and a min_amount. It outputs bought, sold, and net (signed dollar flow).
- enter_when — wire the bought signal in, direction long, to buy when a disclosed purchase posts.
- barrier_exit / trailing_stop_exit / time_exit — pick a hold style; disclosed-flow ideas suit longer holds, so a trailing or time exit fits.
- Add a price confirmation: AND the bought signal with a simple trend gate (price above an sma) so you only act on filings where price still agrees.
Honest backtest note
Unlike the options bricks, the Congress feed is genuinely backtestable once the public filing history is ingested, and crucially it is aligned to each filing's public date — so the backtest cannot cheat by trading on information before it was actually disclosed. That look-ahead discipline is the single most important thing about testing this strategy.
Run it through the locked DEV/HOLDOUT split: a member or filter that shines in-sample but is flat out-of-sample is overfit to a few lucky filings. The verdict (EDGE holdout +EV vs NO EDGE net of costs) and Monte Carlo robustness label tell you whether the disclosed flow carried real predictive value after the lag, or just looked good in hindsight. Build and test a Congress-follow strategy free at /lab.
Put it into practice — no code, free, with an honest out-of-sample check.
Build this strategy free →FAQ
Yes. The transactions are disclosed publicly under the STOCK Act specifically so the public can see them, and acting on already-public information is legal. TradeBricks uses the official, free House Clerk filings. This is not legal or financial advice — it is an educational backtesting tool.
It is unproven as a reliable edge. The disclosure lag of up to ~45 days means the motivating news is often already priced in, and much of the activity is routine. Some concentrated members or filters may show signal in backtests, but it must be confirmed out-of-sample, not assumed from headlines.
The STOCK Act generally requires reporting within 45 days of a transaction, so you can be over a month behind. TradeBricks aligns every signal to the filing's public date to avoid look-ahead, which means your backtest honestly reflects that you only learned about the trade when it was actually disclosed.
Yes. The congress_trades brick lets you filter by a single member, a list of members, chamber, and a minimum dollar size. Be careful: picking the member who happened to perform best in the past is a classic overfit, so always validate any member-specific filter on the out-of-sample holdout.