Momentum - Why Recent Winners Keep Winning, Until They Crash

By EC Assets Research Team, Portfolio Research · Published 29 October 2026

Momentum: Momentum is the tendency of assets that have risen most over the past six to twelve months to keep outperforming those that have fallen most over the following months. It is one of the best-documented return premia across stocks, bonds, currencies and commodities, and the one with the most violent crashes.

What Momentum Is

Momentum is the tendency of recent winners to keep outperforming recent losers. Rank a universe of stocks by their return over the past year, buy the top group and sell the bottom group, and the long-short portfolio has historically earned a substantial premium over the following months. Jegadeesh and Titman (1993) documented the effect in US stocks; Asness, Moskowitz and Pedersen (2013) found it in country indices, government bonds, currencies and commodities as well, which makes it one of the most pervasive patterns in factor investing.

The premium is large and its explanation contested. Behavioural accounts point to investors underreacting to news and then overreacting as trends become visible; risk-based accounts point to the factor's crash exposure, which this entry quantifies below. Both can be true at once, and neither has made the effect disappear since it was published.

How It Works

The standard construction, used in Kenneth French's research factor and in Carhart's (1997) four-factor model, ranks stocks on their return from twelve months ago to one month ago. The most recent month is skipped because one-month returns tend to reverse, an effect that would otherwise contaminate the signal. Large and small stocks are sorted separately and averaged, so that the factor is not simply a bet on size:

$$\text{MOM}_t = \tfrac{1}{2}\big(\text{Small High} + \text{Big High}\big) - \tfrac{1}{2}\big(\text{Small Low} + \text{Big Low}\big)$$

Time-series momentum applies the same idea to each asset's own history. Moskowitz, Ooi and Pedersen (2012) show that a futures contract's past twelve-month excess return predicts its next month's return across equity indices, bonds, currencies and commodities. Going long assets in uptrends and short those in downtrends is the core of managed futures and trend following. Unlike the cross-sectional version, it can be net short in a falling market, which is why trend followers have tended to do well in prolonged sell-offs.

Both versions trade frequently. Portfolios formed on a twelve-month window replace a large share of their positions every year, so momentum is the factor most exposed to trading costs and capacity limits.

Worked Example

In French's data the momentum factor averaged 7.16 percent a year with 14.57 percent volatility from July 1963 to August 2026, a ratio of 0.49, comparable with the equity market itself. The average hides where the risk sits. Following Daniel and Moskowitz (2016), split the months since 1929 by the state of the market: a bear state when the cumulative market return over the previous 24 months is negative, and a normal state otherwise.

January 1929 to September 2026; Kenneth French data library, monthly Mom, with the market return taken as Mkt-RF plus RF.

Bear states cover only 16.3 percent of the months, but they contain the disasters. Nine of the ten worst momentum months came in a bear state with the market rising. The worst, August 1932, lost 52.61 percent while the market gained 37.15 percent; in April 2009 momentum lost 34.36 percent in a month the market rose 10.18 percent, part of a 49.4 percent loss from March to May.

The mechanism is mechanical rather than mysterious. After a long decline the past winners are defensive, low-beta stocks that fell least, and the past losers are the high-beta stocks that fell most. The factor is therefore short the market's most volatile names exactly when a rebound would lift them most. In those states momentum behaves like a short call option on the market: modest gains if the decline continues, heavy losses if it reverses.

When It Applies (and Limitations)

The premium is not steady. In French's US data the factor returned about 7.9 percent a year from 1927 to 2008 and about -2.2 percent a year from 2009 to August 2026. The long-run evidence across markets remains strong, but more than a decade of weak results in the most studied market is a real cost of holding it.

Costs take a larger share than for other factors. Paper returns ignore turnover, bid-ask spreads, market impact and borrowing costs on the short side, and momentum turns over far more than value or quality. Long-only implementations capture only the long half of the spread.

Crash risk is concentrated, not random. The table shows that the bad outcomes cluster in a recognisable state. That makes them partly manageable, but it also means momentum tends to lose at the same time as other positions that are short the market's rebound.

Crowding magnifies reversals. When many managers hold the same winners and short the same losers, forced selling by one moves prices against all of them, which turns ordinary reversals into sharp ones.

Why It Matters for Institutional Investors

It is in most portfolios already. Many active equity managers, smart-beta products and multi-factor funds carry momentum exposure, sometimes unintentionally, through rules that favour stocks with rising prices or improving estimates. Measuring that exposure is part of evaluating any long/short equity or quantitative manager.

It diversifies value. Momentum and value had a correlation of about -0.19 from 1963 in French's data, so a blend has been smoother than either factor alone, and the two rarely suffer their worst periods together.

Its risk can be scaled. Momentum's volatility is predictable and spikes in the crash states. Barroso and Santa-Clara (2015) show that scaling the factor to a constant level of risk, measured from the realised variance of its daily returns, virtually eliminates crashes and nearly doubles its Sharpe ratio, the same logic as volatility targeting applied to a whole portfolio.

References

1. Jegadeesh, N., & Titman, S. (1993). Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency. Journal of Finance, 48(1), 65-91.
2. Carhart, M. M. (1997). On Persistence in Mutual Fund Performance. Journal of Finance, 52(1), 57-82.
3. Moskowitz, T. J., Ooi, Y. H., & Pedersen, L. H. (2012). Time Series Momentum. Journal of Financial Economics, 104(2), 228-250.
4. Asness, C. S., Moskowitz, T. J., & Pedersen, L. H. (2013). Value and Momentum Everywhere. Journal of Finance, 68(3), 929-985.
5. Barroso, P., & Santa-Clara, P. (2015). Momentum Has Its Moments. Journal of Financial Economics, 116(1), 111-120.
6. Daniel, K., & Moskowitz, T. J. (2016). Momentum Crashes. Journal of Financial Economics, 122(2), 221-247.