Leakage Index — abnormal pre-disclosure drift

Leakage Index Did the stock move before the public knew?

Every trade has a non-public window — the days between when a member trades and when they finally disclose it. If a stock drifts favorably during that window, beyond what the market did, the position was already paying off before anyone outside could see it. The Leakage Index measures that market-adjusted, abnormal drift in the non-public window, aggregated per member and per committee. Last updated: 2026-11-05 00:32 UTC.

How to read it. For each trade we take the stock's move from the trade date to the disclosure date, subtract the S&P 500's move over the same days, and orient it to the trade's direction (a buy that rose, or a sell that fell, counts as favorable). Under the honest null — no information advantage — a member's average should sit near zero. A consistently positive average is a pattern worth scrutiny. It is not proof of leakage: markets move, samples are small, and coincidence is real — which is exactly why every figure below carries its sample size n and a 95% confidence interval, and why the all-member baseline is shown for reference.

-0.51%

All-member baseline
average abnormal drift · n=10580 trades

This is the reference line. Members and committees are ranked by how far above it they sit — and only those with at least 10 qualifying trades are ranked, so a lucky one-off can't top the list.

Members — highest abnormal pre-disclosure drift

#	Member	Avg abnormal drift	95% CI	n
1	Tim Walberg House	+11.29%	-19.8…+42.3	10
2	Laurel Lee House	+5.27%	-3.4…+13.9	10
3	Sheri Biggs House	+3.23%	-1.4…+7.9	64
4	Carol Devine Miller House	+3.18%	-0.7…+7.0	17
5	Brian Babin House	+2.57%	-3.2…+8.3	10
6	August Lee Pfluger House	+1.11%	-1.7…+3.9	10
7	Kevin Hern House flagged	+1.11%	+0.3…+2.0	129
8	Bruce Westerman House flagged	+1.08%	+0.3…+1.9	201
9	Richard Dean McCormick House	+0.91%	-0.2…+2.0	88
10	Thomas Kean House	+0.77%	-0.5…+2.1	113
11	Michael McCaul House flagged	+0.71%	+0.2…+1.2	813
12	Dwight Evans House	+0.70%	-1.8…+3.2	27
13	Dan Newhouse House	+0.58%	-1.0…+2.2	77
14	Rick Larsen House	+0.54%	-0.3…+1.4	39
15	Scott Franklin House	+0.47%	-1.0…+1.9	52

flagged = the 95% confidence interval is entirely above zero — the drift is statistically distinguishable from chance at this sample size. It still isn't proof of anything; it's where to look.

Committees — highest abnormal pre-disclosure drift

Each trade is attributed to every committee the trading member sits on — a committee is where non-public information could plausibly originate, so this asks whether some committees' members show more favorable pre-disclosure drift than others.

#	Committee	Avg abnormal drift	95% CI	n
1	Transportation and Maritime Security	+2.63%	-1.2…+6.5	78
2	Tax flagged	+1.14%	+0.3…+2.0	150
3	Energy	+1.05%	-1.5…+3.6	133
4	Investigations and Oversight	+0.91%	-0.2…+2.0	88
5	House Committee on Ways and Means flagged	+0.86%	+0.1…+1.6	243
6	Space and Aeronautics	+0.78%	-0.4…+1.9	89
7	Defense	+0.77%	-5.3…+6.8	12
8	East Asia and Pacific flagged	+0.74%	+0.2…+1.2	1082
9	Environment	+0.67%	-0.3…+1.6	149
10	Railroads, Pipelines, and Hazardous Materials flagged	+0.63%	+0.1…+1.2	444
11	Trade	+0.58%	-0.8…+2.0	65
12	Forestry and Horticulture	+0.58%	-1.0…+2.2	77

Not an accusation, and not investment advice. Abnormal drift in the non-public window is a statistical pattern computed on public filings and public prices. It has many innocent explanations — broad market moves, sector news, routine advisor activity, or plain coincidence — and this page makes no claim that any individual acted on non-public information. It's a transparency tool: it shows where the pattern is strongest so the question can be asked in the open. See also whether the oversight pattern is real and our full track record.