Did copying Congress and insiders beat the market?
The average return of buying after each disclosure, against the S&P 500, at 30, 90 and 180 days. Sample sizes and a plain reading of the result sit next to every number.
Key Facts
- Congress: -0.7% against the S&P 500 over 90 days across 12,732 scored buys, 46% of them ahead, which is modestly below the market, but could still be chance.
- Insiders: -3.3% against the S&P 500 over 90 days across 2,393 scored buys, 41% of them ahead, which is based on too few months of entries (4) to judge.
- Cluster buys: -4.2% against the S&P 500 over 90 days across 146 scored buys, 43% of them ahead, which is based on too few months of entries (4) to judge.
- No Congress filings are recorded for January 2024 to April 2025 (16 months) and July 2025 to November 2025 (5 months), so results do not cover that period.
- Insider filings begin on 9 March 2026, so insider results cover a short window.
- The benchmark is the S&P 500, measured through SPY. Prices are split and dividend adjusted and current to 2026-09-28 (28 September 2026).
The short answer. Congress buys trailed the S&P 500 by 1.9% on average over 180 days, a gap larger than luck would usually explain. That is a measured historical average, not a forecast, and it can change as new trades mature. The insider results cover too few months of entries to judge either way.
Trailing the S&P 500 does not prove poor stock picking. The index is weighted by company size, so a basket of typical stocks can lag it when a few very large companies lead. It shows what buying after a disclosure earned compared with simply owning the index.
Average excess return after disclosure
prices as of 2026-09-28| Days after entry | Congress | Insiders | Cluster buys |
|---|---|---|---|
| 0d | 0.00% | 0.00% | 0.00% |
| 7d | +0.19% | -0.65% | -0.47% |
| 14d | +0.02% | -1.00% | -0.18% |
| 21d | +0.14% | -1.53% | -0.45% |
| 28d | +0.08% | -1.90% | -1.12% |
| 35d | +0.17% | -1.95% | -1.67% |
| 42d | -0.01% | -2.68% | -2.51% |
| 49d | -0.02% | -3.73% | -3.41% |
| 56d | +0.03% | -4.21% | -3.56% |
| 63d | -0.35% | -3.93% | -2.77% |
| 70d | -0.42% | -3.84% | -3.05% |
| 77d | -0.52% | -4.42% | -4.92% |
| 84d | -0.63% | -3.24% | -4.29% |
| 91d | -0.69% | -3.26% | -4.52% |
| 98d | -0.79% | -3.90% | -4.84% |
| 105d | -0.81% | -4.80% | -5.01% |
| 112d | -0.83% | -5.04% | -5.44% |
| 119d | -1.08% | -6.45% | -7.32% |
| 126d | -1.17% | -7.64% | -8.25% |
| 133d | -1.27% | -8.05% | -8.70% |
| 140d | -1.45% | -8.59% | -7.96% |
| 147d | -1.51% | -9.81% | -8.72% |
| 154d | -1.81% | -9.90% | -8.42% |
| 161d | -1.86% | -9.69% | -7.39% |
| 168d | -1.91% | -9.86% | -7.12% |
| 175d | -1.90% | -9.62% | -6.54% |
| 180d | -1.91% | -9.58% | -9.86% |
Congress
| Days after entry | Scored buys | Avg vs S&P | Beat S&P | Entry months | t statistic | Reading |
|---|---|---|---|---|---|---|
| 30 days | 13,839 | +0.1% | 49% | 47 | 0.42 | not distinguishable from luck |
| 90 days | 12,732 | -0.7% | 46% | 45 | -1.92 | modestly below the market, but could still be chance |
| 180 days | 11,680 | -1.9% | 44% | 42 | -4.43 | clearly below the market |
| Days after entry | Avg return | S&P 500 return | Avg vs S&P | Median vs S&P |
|---|---|---|---|---|
| 30 days | +1.4% | +1.3% | +0.1% | -0.1% |
| 90 days | +2.7% | +3.4% | -0.7% | -1.3% |
| 180 days | +3.6% | +5.5% | -1.9% | -2.5% |
| Year | Scored buys | Avg vs S&P, 90d | Beat S&P |
|---|---|---|---|
| 2021 | 3,375 | -1.4% | 45% |
| 2022 | 3,804 | +0.7% | 52% |
| 2023 | 3,069 | -1.8% | 42% |
| 2025 | 639 | -1.3% | 42% |
| 2026 | 1,845 | -0.5% | 43% |
20,125 buys scored, 1,574 skipped because no usable price history exists.
Insiders
| Days after entry | Scored buys | Avg vs S&P | Beat S&P | Entry months | t statistic | Reading |
|---|---|---|---|---|---|---|
| 30 days | 3,826 | -1.9% | 41% | 6 | n/a | based on too few months of entries (6) to judge |
| 90 days | 2,393 | -3.3% | 41% | 4 | n/a | based on too few months of entries (4) to judge |
| 180 days | 527 | -9.6% | 34% | 2 | n/a | based on too few months of entries (2) to judge |
| Days after entry | Avg return | S&P 500 return | Avg vs S&P | Median vs S&P |
|---|---|---|---|---|
| 30 days | +0.5% | +2.4% | -1.9% | -1.8% |
| 90 days | +3.3% | +6.5% | -3.3% | -3.8% |
| 180 days | +7.7% | +17.3% | -9.6% | -12.1% |
| Year | Scored buys | Avg vs S&P, 90d | Beat S&P |
|---|---|---|---|
| 2026 | 2,393 | -3.3% | 41% |
5,942 buys scored, 166 skipped because no usable price history exists.
Cluster buys
| Days after entry | Scored buys | Avg vs S&P | Beat S&P | Entry months | t statistic | Reading |
|---|---|---|---|---|---|---|
| 30 days | 225 | -1.0% | 42% | 6 | n/a | based on too few months of entries (6) to judge |
| 90 days | 146 | -4.2% | 43% | 4 | n/a | based on too few months of entries (4) to judge |
| 180 days | 25 | -9.9% | 40% | 1 | n/a | too few trades to tell |
| Days after entry | Avg return | S&P 500 return | Avg vs S&P | Median vs S&P |
|---|---|---|---|---|
| 30 days | +1.6% | +2.6% | -1.0% | -1.7% |
| 90 days | +2.4% | +6.6% | -4.2% | -2.1% |
| 180 days | +7.7% | +17.6% | -9.9% | -13.6% |
261 clusters scored.
Related pages
Frequently asked questions
Did copying Congress trades beat the market?
Congress: -0.7% against the S&P 500 over 90 days across 12,732 scored buys, 46% of them ahead, which is modestly below the market, but could still be chance. Returns start on the first trading day after each filing.
Did copying insider buys beat the market?
Insiders: -3.3% against the S&P 500 over 90 days across 2,393 scored buys, 41% of them ahead, which is based on too few months of entries (4) to judge. Only open market buys are scored.
What does "not distinguishable from luck" mean?
The t statistic compares the average result with how much individual results vary. Below about 1.65 in size, a gap this large would appear by chance fairly often, so it says nothing reliable about skill.
Does trailing the S&P 500 mean these traders pick bad stocks?
Not necessarily. The S&P 500 is weighted by company size, so a handful of very large companies drive it. A basket of typical stocks can trail it in years when the largest companies lead, even if the picks were reasonable. The comparison answers whether buying after a disclosure beat simply owning the index, not whether the picks were good in isolation.
Why not use the date of the trade?
Because nobody outside can act on a trade until it is filed. Starting on the trade date would credit the follower with a head start they never had.
Sources: Congressional trades: STOCK Act disclosures, collected from CapitolTrades. Insider trades: SEC Form 4 filings, collected from Finviz. Prices: Yahoo Finance daily adjusted closes. Data as of 2026-09-29 (29 September 2026). About these sources.
Not investment advice. Past disclosed trades do not predict future returns. Disclosures are published late, so the trades shown are ones you could only have copied after the fact. Talk to a licensed adviser before you invest.