The Momentum Factor
Securities that have performed well over the past several months have tended to continue over the following months. Robust, widely replicated, and prone to sudden severe reversals.
MadStockAlerts Research · Updated August 29, 2026
What to take away
- The standard construction ranks by trailing return over 3 to 12 months, skipping the most recent month.
- It is among the most replicated anomalies, across markets and asset classes.
- It survived publication better than most documented effects.
- Momentum crashes are severe, and they occur at market turning points.
- It contradicts the weak form of market efficiency directly.
MAD Academy Training Video · 0:46
The Effect That Should Not Exist
Momentum is the best-documented and least comfortable anomaly: what has gone up keeps going up, until it crashes.
This lesson is part of a Stock Alerts + Tools plan.
The construction
The conventional academic construction ranks a universe by return over the previous twelve months excluding the most recent one, holds the top group and shorts the bottom, and rebalances monthly. The exclusion of the last month is deliberate.
That most recent month tends to reverse rather than continue, a separate documented effect attributed partly to microstructure and partly to short-term liquidity provision. Including it weakens the result, which is why the skip is standard.
The distinction between horizons matters throughout. Very short horizons reverse, intermediate horizons continue, and very long horizons reverse again, which is the long-run reversal effect.
Why it is taken seriously
- It has been documented in equity markets across dozens of countries.
- It appears in bonds, currencies, commodities and across asset classes as a group.
- It has been found in data going back to the nineteenth century.
- Unlike most published anomalies, it persisted after publication, though with a reduced premium.
- It has no widely accepted risk-based explanation, which makes it a genuine challenge to efficiency.
The fifth point is why this effect appears in a library that is generally sceptical of price-based claims. Momentum is a price-based effect that has survived the tests most such claims fail.
Momentum crashes
The factor's returns are strongly negatively skewed. It produces steady gains for extended periods punctuated by sudden, severe reversals, and the reversals occur at market turning points.
The mechanism is structural. After a sharp decline, the losers a momentum portfolio is short are heavily beaten-down, high-beta securities. When the market turns, those rebound hardest, and the short side loses violently at the same moment the long side of defensive winners lags.
Documented episodes have produced losses of a substantial fraction of a momentum portfolio's value within weeks. Any evaluation of the factor that excludes those episodes is describing the strategy without its principal risk.
Scroll the chart sideways to see all of it.
The explanations
| Explanation | Mechanism |
|---|---|
| Underreaction | Information is incorporated gradually rather than instantly |
| Delayed overreaction | Trends attract attention, extending them past fundamentals |
| Disposition effect | Reluctance to realise losses slows the adjustment downward |
| Crash risk premium | The return compensates for the severe tail, which is a risk explanation |
The fourth row is the only risk-based candidate and it has some support: an effect with a severe negative skew may simply pay for that skew. That would make momentum a compensated risk rather than a market failure, and it would explain the persistence after publication.
What implementing it costs
Momentum is the highest-turnover of the common factors by construction: the ranking changes as prices change, so the portfolio has to change with it.
- Annual turnover in a monthly-rebalanced momentum portfolio is frequently well above one hundred percent.
- That turnover incurs spread and impact on every rebalance, in securities that have recently moved.
- Academic returns are typically gross of costs, so published premiums overstate what is capturable.
- In a taxable account, the turnover produces short-term gains, which are taxed at ordinary rates.
The fourth item is the one that most affects an individual. A factor with a documented premium of a few percent, implemented at high turnover in a taxable account, can deliver nothing after tax while the pre-tax record continues to look good.
Practical implementations reduce turnover deliberately, through wider rebalancing bands and less frequent reconstitution, at the cost of tracking the documented factor less closely.