Josh Gottheimer Files More Stock Trades Than Almost Anyone in Congress. Copying All 943 Lost to the Index

Nancy Pelosi gets the attention. On volume, Josh Gottheimer's filings dwarf hers: 943 disclosed buys since 2017, against her 75 in eleven years.

That volume makes his record unusually well-suited to a backtest. A handful of trades can be luck. Nine hundred is a process — and the process trailed a index fund.

The result

Copying these filings more than doubled the money over eight years — and still finished 64.9 points behind simply buying the index. He ranks 12th of 153.

+141.9%Copying the filings+206.8%S&P 500837Positions opened27dMedian disclosure lag

Exit disciplineMirrorPosition size5%Add membersUp to 10

Run this on any member, with your own exit rules

A frozen run, not a live one: $10k start, every disclosed buy sized at 5% of equity and entered at the first open after it became public, mirroring the member's own disclosed exits. Past results, not a performance claim.

This article previously said the opposite

An earlier version of this page reported +514.2% and called it "one of only 13 congressional records that beat the market". That was wrong, and the cause is worth stating plainly because it is specific to this record.

One of his disclosed tickers was renamed years ago — Priceline to Booking Holdings — leaving a dead stub price series. Our engine consumed its fill queue in filing order, so that single unfillable entry blocked every entry queued behind it. His simulated account froze at its starting balance for five years, then executed the ~880-entry backlog at original planned prices using current cash, and promptly exited at current prices. Years of appreciation, booked with no market exposure in between.

Fixed, the queue drains in entry-date order and stub entries are dropped. His measured exposure goes from 1.6% to 97.5%, the drawdown from near-zero to 33.6%, and the return from +514% to 141.9%.

The correction is the most useful thing on this page: a backtest that reports a huge return with almost no drawdown is not describing a great strategy, it is describing a broken simulation.

What the filings show

Microsoft alone accounts for more buys than Pelosi's entire disclosed history. The pattern — repeated, incremental purchases of the same large-cap names — looks far more like a managed account on a regular schedule than like discrete conviction bets.

That's worth sitting with, because it changes the interpretation. A high-frequency drip into mega-cap tech between 2018 and 2026 was an extremely good thing to be doing, and it does not require anybody to have known anything.

Only 4.5% of these buys were options, so unlike Pelosi's record — where 30.7% were — this simulation is a fair mechanical copy of what was filed.

Why it lost, on the largest sample in Congress

The disclosure lag eats the move. A median 27-day gap means the first part of every price move happened before the filing was public. On a book that drips into fast-moving large-cap technology, that first part is often the significant part.

The win rate counts only closed trades. A position scores win or lose when the member sells it and we mirror that exit. Anything never sold stays open and never scores. People sell winners more readily than losers, so the win rate describes the trades he chose to close, not his judgement.

Volume is not edge. 837 filled entries keep the account almost fully invested — 97.5% on average — which means it tracks the market closely and pays the lag on every entry. Being in the market more does not beat being in the market; it just costs more to get there.

Could you have copied it in practice?

This is where it gets less appealing. 943 buys over eight years is roughly two entries a week, each requiring you to read a filing, size a position and act — plus the mirrored sells. Commissions are zero at most brokers now, but attention isn't, and the simulation charges nothing for either.

And you'd still be acting a median 27 days late — for a result that trailed the index.

The honest summary

The largest congressional disclosure record we can test, copied mechanically from the date each filing became public, trailed the S&P 500 by 64.9 points over eight years — with a modest options share, a 97.5% average exposure and a high fill rate, so there is no coverage excuse in it.

If following any single member were going to work, this is the record where it had the best chance: the most trades, the cleanest data, a market that rose throughout. It still lost. The full board — all 153 rankable members, with fill counts beside every return.

The fine print

Simulation enters at the first open after each filing became public, sizes each position at 5% of account equity, and mirrors the member's disclosed exits. Amounts are disclosed as bands, so equal weighting is an assumption the filings force. Runs on adjusted closes, so dividends are reinvested; taxes, market impact and commission are not modelled. This is an analysis of public filings, not a claim about anyone's conduct, and it is educational content rather than investment advice.