Post-earnings announcement drift
MomentumAdvanced
Post-earnings announcement drift is the tendency for a stock to keep moving, for weeks, in the direction of an earnings surprise. It is an event strategy, not a chart pattern. The usual explanation in the research is that investors underreact to the new information, though that is a theory, not something a backtest can prove.
Not investment advice. Nothing here is a recommendation to buy or sell anything. Backtests and historical studies are not guarantees, and a rule that worked in one sample can fail in the next. Trading can lose money. Links are for education, not endorsements. Do your own research.
- Rules
- Define a surprise, such as earnings versus the consensus estimate, or the stock's own return over the announcement window. Go long the strongest positive surprises and, if the book is long-short, short the strongest negative ones. Hold for several weeks to a few months. Most published versions find more of the effect in smaller, less liquid stocks, so a tradable version has to cap those names.
- Where AI fits
- NLP can read the release and the call transcript and score tone, guidance, and uncertainty on top of the numeric surprise. A model can also decide position size from liquidity so the strategy does not pretend small-cap prints are free to trade. The announcement timestamp has to be lagged correctly or the backtest looks into the future.
- Markets
- Individual stocks with a reliable earnings calendar and analyst estimates. Large caps are easier to trade and may show a smaller effect.
- Timeframes
- Hold for weeks. The entry is around the announcement, not a clock time.
- Risks
- The drift is not guaranteed on a single name, and one gap can erase a quarter of small signals. Estimates get revised, so the surprise definition matters. Crowding and costs are larger than a frictionless academic portfolio implies.