Short and Long-Term Reversal
Reversal effects describe recent losers beating recent winners over very short and very long horizons. Learn the evidence, the causes and the link to momentum.
Momentum says winners keep winning over the medium term. But at the very short horizon, up to about a month, and at the very long horizon, three to five years, research has found the opposite: past losers tend to outperform past winners. These reversal effects bracket momentum in time, and they explain why momentum strategies skip the most recent month and why value investing has long horizon roots. Understanding all three horizons helps traders avoid applying the right idea at the wrong timescale.
The return horizon map#
| Horizon | Typical pattern in stocks | Lesson |
|---|---|---|
| Days to about one month | Short term reversal | This lesson |
| 3 to 12 months | Momentum | Momentum Factor |
| 3 to 5 years | Long term reversal | This lesson |
Short term reversal#
- Jegadeesh (1990) and Lehmann (1990) found that stocks with the lowest returns over the past week or month tended to outperform the following week or month, and the best performers tended to underperform.
- The effect is strongest in small, illiquid stocks and around non news price moves.
Why it happens#
| Explanation | Idea |
|---|---|
| Liquidity provision | Large trades push prices away from fair value; prices bounce back once the pressure passes, rewarding traders who provide liquidity. See Liquidity Factor |
| Bid ask bounce | Trades alternate between bid and ask prices, creating artificial negative autocorrelation. See Autocorrelation and Partial Autocorrelation |
| Overreaction | Investors overreact to short term noise |
Long term reversal#
- De Bondt and Thaler (1985), in "Does the Stock Market Overreact?", found that stocks with the worst returns over the previous three to five years outperformed the best performers over the following three to five years.
- The effect overlaps heavily with the value factor: long term losers tend to become cheap. See Value Factor.
Why it happens#
| Explanation | Idea |
|---|---|
| Overreaction | Investors extrapolate past performance too far, then correct |
| Risk | Long term losers may be riskier, distressed firms |
| Value overlap | Long term losers become cheap, capturing the value premium |
Reversal and momentum together#
The three horizon pattern helps reconcile the evidence:
- Short term: prices overshoot due to trading pressure and noise, then partly reverse.
- Medium term: news and fundamentals diffuse slowly, so trends persist.
- Long term: extrapolation goes too far, and valuations correct.
This is why the standard momentum signal uses returns from 12 months ago to 1 month ago, skipping the reversal month. See Momentum Factor.
Reversal in other markets#
| Market | Pattern |
|---|---|
| Futures and currencies | Short term reversal is weaker; time series momentum dominates medium horizons. See Trend Following |
| Crypto | Very short horizon reversals and momentum both appear, varying by period |
| Intraday | Mean reversion around VWAP and after large moves without news. See Mean Reversion |
Trading considerations#
- Costs dominate short term reversal; only low cost, fast execution can capture it.
- Distinguish news from noise: reversal works better after moves without fundamental news.
- Liquidity matters: the effect is biggest where liquidity is thin, which raises costs and capacity limits. See Alpha Capacity and Crowding.
Frequently asked questions#
What is short term reversal?#
The tendency for stocks with the worst returns over the past week or month to outperform, and recent winners to underperform, in the following weeks.
What is long term reversal?#
The tendency for stocks with poor returns over three to five years to outperform past winners over the next several years, overlapping with value.
Why does momentum skip the most recent month?#
Because short term reversal over the latest month works against momentum, so excluding it improves the momentum signal.
Next, learn what happens when too many investors chase the same factors in Factor Timing, Crowding and Crashes.
3 quick questions on this lesson. Get them all right to finish it.
Turn on JavaScript to take the quiz.
Mentioned in
- Signal and Alpha DecayResearch and Backtesting
- Growth and Dividend FactorsResearch and Backtesting
- Momentum TradingStrategies and Styles
- White Noise and Random WalksMath and Statistics
- Securities Lending and Stock LoanThe Trading Industry