The best and worst traders in Congress
Most lists of "Congress's best stock traders" are built from somebody's best trade. One spectacular pick, annualised, makes anyone look like a genius.
Ours is built differently: the plain average return of every disclosed buy a member has made in the last five years. Every trade counts once — a $2,000 purchase and a $2,000,000 purchase contribute equally — with no cherry-picked window and no dropped losers. We call it Composite, and it is deliberately unflattering.
The current top of the table
| # | Member | Composite | Priced trades |
|---|---|---|---|
| 1 | Nancy Pelosi (D-CA) | +34.5% | 20 |
| 2 | Steve Cohen (D-TN) | +33.9% | 6 |
| 3 | Markwayne Mullin (R-OK) | +32.9% | 289 |
| 4 | Thomas H. Kean, Jr. (R-NJ) | +30.7% | 71 |
| 5 | Shelley Moore Capito (R-WV) | +28.6% | 203 |
And the bottom:
| # | Member | Composite | Priced trades |
|---|---|---|---|
| 24 | Jerry Moran (R-KS) | +1.1% | 99 |
| 25 | Richard McCormick (R-GA) | −0.7% | 1 |
| 26 | April McClain Delaney (D-MD) | −3.5% | 234 |
| 27 | John J. McGuire III (R-VA) | −4.5% | 13 |
The live rankings update daily.
Read the trade count before the percentage
Look at rank 2 and rank 25. Steve Cohen's +33.9% rests on six trades. Richard McCormick's −0.7% rests on one.
A member with six trades and one good one can top a table that a member with 289 trades cannot, because the average has almost nothing to average. We require ten priced trades before ranking anyone at all, and even that is a low bar — below about fifty, treat the figure as an anecdote with a decimal point.
Markwayne Mullin at rank 3 on 289 trades is the more meaningful entry on that list. Sustaining +32.9% across that many decisions is a different claim from doing it across six.
The bigger problem with the whole table
Every number above is a raw return. It credits the member with whatever their sectors did anyway.
Across all 4,791 scored buys in this dataset, the average disclosed purchase is up 5.15% after three months — and −0.32% measured against each stock's own sector benchmark over the same window. Most of what the rankings show is the market, not the picker. The full comparison.
So the honest reading of the table is: these members' disclosed buys returned the most over this period. Not: these members are the best stock pickers in Congress. The second claim needs the sector stripped out, and when you strip it out most of the difference goes with it.
We publish the raw ranking anyway, because it is the number people come for and hiding it would be its own kind of dishonesty. We just put the caveat next to it rather than in a footnote.
What about party?
The congressional-trading ETFs split by party — NANC for Democrats, KRUZ for Republicans — which implies party is a meaningful predictor.
In this dataset it barely registers. Mean Composite is +14.5% across 11 Democrats and +13.2% across 16 Republicans. A 1.3-point gap on samples that small is noise, and the top ten contains seven Republicans and three Democrats. What those funds actually hold.
What the ranking cannot tell you
Whose decision it was. Spouse and dependent-child trades are filed under the member's name. Why that matters.
What they actually paid. Filings disclose a date and an amount band, never an execution price. Every figure uses that day's closing price as a stand-in.
Whether they still hold it. Sales disclose on the same 30–45 day lag as purchases.
Anything about the other 508 members. Coverage is the most active disclosed traders. Members who trade rarely, or hold only index funds, do not appear — absence from this table is not a finding.
Check it
Every trade behind every score is in the published dataset, and the scoring rule is written out in full on the methodology page. If you think equal-weighting each buy is the wrong rule, the file has what you need to score it your way.