// study
Do Insider Buys Outperform? A Study of 45,000 Open-Market Purchases
Updated 2026-06-16. By Theodor Nielsen, founder of Form4API.
// answer
Across 45,046 open-market insider purchases in the Form4API database, the average absolute return was +8.6% three months later (median +3.0%), with 58% of buys positive. Open-market sells over the same window averaged just +2.4% with a 50% hit rate — so the buy-minus-sell spread, which cancels most of the market's drift, is the real insider signal.
Open-market buys vs sells
We measured the post-trade return of every open-market purchase (code P) and sale (code S) in the database, excluding 10b5-1 plan trades and derivatives, at one, three, and six months after the filing date:
| Trade | Count | Avg 1m | Avg 3m | Hit 3m | Avg 6m |
|---|---|---|---|---|---|
| Buys (P) | 45,046 | +3.7% | +8.6% | 58% | +12.5% |
| Sells (S) | 92,935 | +0.7% | +2.4% | 50% | +5.6% |
Absolute, split-adjusted returns anchored at the first close on or after the filing date. Median 3-month return: +3.0% for buys, 0.0% for sells.
The buy/sell asymmetry is the signal
Buys beat sells at every horizon — by roughly six percentage points at three months (+8.6% vs +2.4%) and nearly seven at six. This matches the long-standing intuition behind insider analysis: an insider sells for many reasons (diversification, taxes, a new house), but generally buys for only one — they expect the stock to rise.
Note the honest reading: sells still showed a positive average (+2.4% at three months) because the broad market drifted up over the sample. That is exactly why the difference between buys and sells — not the buy number alone — is the cleaner estimate of the insider effect. Differencing two absolute-return series measured over the same calendar removes most of the shared market drift.
Does trade size matter?
Surprisingly little. Splitting the 45,046 buys by dollar value (shares × price), three-month results were:
- Under $100k — +8.8% average, 58% hit rate (n≈33,000).
- $100k–$1M — +8.7% average, 59% hit rate (n≈8,700).
- Over $1M — +7.3% average, 57% hit rate (n≈3,400).
The mid bucket edged the others, and the largest buys were marginally weaker — so raw trade size is a poor standalone filter. Coordinated buying and the insider's track record do more work; see cluster buy signals and the top-insiders leaderboard.
Methodology
- Population. Every open-market transaction in the Form4API database with a matured post-trade return — code P (buys) and code S (sells).
- Exclusions. 10b5-1 plan trades, derivative transactions, and superseded (amended) rows are removed.
- Return definition. Absolute, split-adjusted price return anchored at the first market close on or after the filing date, measured at 21, 63, and 126 trading days (≈1, 3, 6 months). Not market- or sector-adjusted.
- Hit rate. Share of trades with a positive return over the horizon.
- Maturity. Only trades whose horizon has fully elapsed are counted, so the sample is backward-looking and recent filings are under-represented.
Reproduce it yourself
The same return data is exposed on the API. Returns are fractions (so 0.1 = 10%); filter open-market buys that were up at least 10% at three months with:
curl "https://api.form4api.com/v1/transactions?code=P&exclude_10b5=true&min_return_3m=0.1&has_returns=true" \ -H "X-Api-Key: $FORM4API_KEY"
The full filter grammar — codes, size, return horizons, date range — is in the screener guide, and you can export the matching rows as CSV for your own analysis.
Frequently asked questions
Do insider buys outperform the market?
In this dataset, open-market insider purchases averaged a +8.6% absolute return three months later, with 58% of buys positive — versus +2.4% and 50% for open-market sells. The clean signal is the gap between the two: because both are measured over the same calendar with absolute (not market-adjusted) returns, the roughly 6-percentage-point buy-minus-sell spread at three months largely cancels out general market drift and isolates the insider-buying effect. So buys meaningfully outperformed sells; whether they beat the broad market depends on the period, which absolute returns alone cannot settle.
How long after an insider buys does the stock move?
The effect builds over months rather than days. Across 45,046 open-market buys, the average absolute return was +3.7% at one month, +8.6% at three months, and +12.5% at six months, with the share of positive trades rising from the mid-50s to 59%. That gradual build is why most insider-signal strategies use multi-month holding windows rather than trying to trade the filing itself.
Do bigger insider buys perform better?
Only marginally, and not monotonically. Splitting buys by dollar size, purchases of $100k to $1M had the highest three-month hit rate at 59%, while buys over $1M were slightly lower at 57% and a +7.3% average versus +8.8% for sub-$100k buys. Size is a weak filter on its own; coordinated buying (clusters) and the insider’s own track record are stronger signals than raw trade value.
What are the limits of this study?
Three big ones. Returns are absolute, not benchmark- or sector-adjusted, so a rising market lifts every number. The sample only includes trades whose return window has fully matured, so it is backward-looking by construction. And past performance does not predict future returns — these are population statistics, not a trading strategy. Treat the figures as a baseline for your own research, which you can reproduce with the API.
Population statistics from publicly filed SEC Form 4 data, for research and education. Not investment advice; past performance does not predict future returns.