Post-Earnings Drift: What 50 Years of Research Does and Doesn’t Show
Stocks that report earnings well above or below expectations have historically tended to keep drifting in the same direction for weeks. Here is where that finding comes from, what the studies actually measured, and how some traders try to use it.
By PYE Markets • Published 2026-10-19 • 6 min read
Key points
- Post-earnings-announcement drift (PEAD) was first documented by Ball & Brown (1968): share prices kept moving in the direction of an earnings surprise after the announcement.
- Bernard & Thomas (1989, 1990) studied it in depth. They sorted companies by standardized unexpected earnings (SUE)—how far reported earnings differed from a simple forecast based on past earnings—and found the drift persisted for about 60 trading days after announcements, with part of it showing up around the following quarters’ announcements.
- The studies measured earnings surprises, not revenue or guidance. Checklists that add revenue beats, guidance raises or volume filters are practitioner extensions that the original research did not test.
- First Documented: Ball & Brown (1968)
- Key Extension: Bernard & Thomas (1989, 1990)
- Surprise Measure: Standardized Unexpected Earnings
Holding a stock or options through an earnings release is a bet on an unknown number. An alternative some traders prefer is to wait for the report, see the surprise, and then consider whether the post-earnings drift documented in academic research might apply. That research is real and long-standing, but it is narrower than many trading checklists imply.
1. What the Research Found
Ball & Brown (1968) showed that stock prices continued to adjust after earnings were announced. Bernard & Thomas (1989) sorted firms into groups by standardized unexpected earnings and found that the most positive-surprise group kept outperforming, and the most negative-surprise group kept underperforming, over roughly the 60 trading days after the announcement. Their 1990 paper argued that prices did not fully reflect what current earnings implied for future earnings. Explanations debated since include slow investor reaction and limits to arbitrage; some later studies report that the effect is concentrated in smaller, less liquid stocks and has weakened in recent decades.
- What was measured: the size of the earnings surprise relative to a statistical forecast.
- What was not measured: revenue beats, management guidance, day-one volume or chart patterns.
- Historical tendency, not a guarantee: results vary by period, company size and costs.
2. How Some Traders Try to Apply It
A common practitioner approach is to wait for the report and look for a clear positive surprise, often alongside revenue growth and raised guidance, strong trading volume on the day after the report and a close near the day’s high. Entry methods vary—some use a break above the first 15–30 minutes’ range, others wait several days for a tight consolidation. These filters are heuristics; they have not been tested in the studies cited above.
3. Defining the Risk
Because the drift is a statistical tendency across many stocks, any single trade can fail. Traders who act on it typically decide in advance where the idea is wrong—for example, below the post-earnings day’s low—and size the position so that a loss at that level is a fixed, small share of their account.
Research Finding vs. Practitioner Add-Ons
| Element | Supported by the Cited Research | Practitioner Heuristic (Not Tested There) |
|---|---|---|
| Earnings surprise | Yes — standardized unexpected earnings (SUE) | — |
| Revenue beat / guidance raise | No | Often used as an extra filter |
| Day-one volume and close | No | Often used as confirmation |
| Holding period | Drift measured over about 60 trading days | Exits often based on moving averages or chart levels |
Post-earnings drift is one of the best-documented patterns in academic finance, but the evidence concerns earnings surprises measured across many stocks. Revenue, guidance and volume filters may be sensible, but they are untested additions.
Editorial Disclosure: Research cited: Ray Ball & Philip Brown, “An Empirical Evaluation of Accounting Income Numbers” (Journal of Accounting Research, 1968); Victor L. Bernard & Jacob K. Thomas, “Post-Earnings-Announcement Drift: Delayed Price Response or Risk Premium?” (Journal of Accounting Research, 1989) and “Evidence That Stock Prices Do Not Fully Reflect the Implications of Current Earnings for Future Earnings” (Journal of Accounting and Economics, 1990). The trading checklist is a practitioner approach, not a finding of that research. General information, not personal trading advice.