Momentum investing means buying what has already gone up and avoiding or shorting what has already gone down, then repeating the process on a schedule. The idea is old, widely documented in academic research, and still uncomfortable to hold, because the same force that produces long stretches of outperformance produces sudden, violent reversals.
If you want to use momentum, the useful question is not whether it works. It is when it stops working, what that costs you, and whether you can stay in the position long enough to find out. This article separates what the evidence shows from what remains assumption, and treats traders and long-term investors separately, because their time horizons change the answer.
What momentum actually is
Momentum is the tendency of assets that performed well over the past several months to keep performing well over the next several months, and for past losers to keep losing. Researchers usually measure it by ranking a broad universe of stocks on returns over a lookback window, often around six to twelve months, skipping the most recent month, then buying the top decile and shorting the bottom decile.
The skip matters, because short-term reversal is a separate, well-documented effect: stocks that jumped in the last few weeks often give some of it back. Excluding the most recent month therefore reduces the overlap between the two effects.
Momentum is not the same as trend following, though they are cousins. Trend following typically uses moving averages or breakout rules on futures, currencies, and commodities, and it often works on longer horizons. Momentum as a factor is usually a cross-sectional ranking: it tells you which stocks to hold relative to each other, not just whether to be long or flat.
What you are capturing, mechanically, is a mix of slow information diffusion, investor underreaction to news, and institutional behavior such as fund flows and window dressing. None of these mechanisms is guaranteed to persist. That is the honest starting point.
The evidence, and what it does not promise
Momentum is one of the most replicated findings in empirical finance. It shows up in US equities, and has been studied in various markets and asset classes. That breadth is why it is usually described as a genuine anomaly rather than a data artifact.
The evidence also has limits that retail investors should take seriously. Published factor returns are gross of transaction costs, and momentum is a high-turnover strategy. Bid-ask spreads, market impact, and taxes eat into the premium, sometimes substantially. Studies that adjust for realistic trading costs find smaller but still positive net returns in liquid large-cap universes, and much weaker results in small, illiquid names.
There is also a publication and replication problem across factor research generally. Some factors that looked strong in earlier samples weakened or disappeared out of sample. Momentum has held up better than most, but past robustness is not a promise about your holding period.
Treat the premium as compensation for bearing a specific risk, not as free money. The risk is real, it is concentrated in time, and it is the reason the premium exists at all.
Why momentum crashes
Momentum crashes have a recognizable shape. They tend to happen after a market bottom, during a sharp rebound, when the previous losers rally hardest and the previous winners lag. If you are short the losers and long the winners, both legs move against you at once.
The mechanism is straightforward. After a severe bear market, the loser portfolio is often full of small, highly levered, distressed companies. When the market turns, these stocks can rise multiples of the index in a matter of weeks. Meanwhile, the winners are frequently defensive, low-beta names that lag in a recovery. The spread collapses.
Crashes are also amplified by positioning, because when many systematic funds run similar momentum signals, they tend to de-risk at the same time. This pushes prices further and forces more selling, while leverage makes the dynamic worse. A strategy that looks diversified across hundreds of stocks can still be one crowded trade.
Momentum’s return distribution is the key fact here, and it is not symmetric. Long quiet periods of modest gains are interrupted by rare, large losses, which is why momentum is often described as picking up pennies in front of a steamroller. That shape also explains why the strategy is psychologically hard to stick with, even when the long-run numbers look attractive.
A practical implication: the biggest risk in momentum is not the average drawdown. It is the tail event that arrives when you are most confident and most leveraged.
Momentum for traders (weeks to about six months)
If your horizon is weeks to a few months, momentum is a tactical tool, and your main job is risk control rather than signal generation.
Start with the universe, because liquid large-caps and liquid ETFs are where momentum signals survive transaction costs. Illiquid small-caps, by contrast, can show spectacular backtests yet disappointing live results, since you cannot get filled at the prices the model assumes.
Define your signal before you trade. A common approach is a ranking on medium-term returns with a short-term reversal filter, rebalanced periodically, and the exact parameters matter less than consistency. Changing the lookback every time the strategy underperforms is how you turn a factor exposure into discretionary noise.
Size positions so that a momentum crash does not end your account. That means limiting gross exposure, avoiding leverage in the strategy itself, and setting a hard stop on the portfolio, not just on individual names. A stop on each stock does not protect you when the whole factor unwinds at once.
Watch the regime, since momentum tends to do well in trending markets and poorly around sharp reversals, especially after bear market bottoms. You do not need to predict turns, but you should know that your strategy has a known weak spot; therefore, reduce exposure when the market environment matches it.
Finally, accept that you will underperform for stretches. If you cannot tolerate a year of lagging the index, a tactical momentum overlay is probably the wrong tool for you.
Momentum for investors (six months and longer)
For investors with a multi-year horizon, momentum is best treated as one factor among several rather than a standalone strategy.
A long-only momentum tilt, implemented through a rules-based ETF or a screened portfolio, avoids the short leg that drives most of the crash risk. That changes the payoff: although you give up some of the premium, you avoid the scenario where a short squeeze in junk stocks wipes out a year of gains.
Rebalancing discipline does most of the work, because momentum portfolios need to be refreshed as the signal decays. Rebalancing frequencies vary; some momentum strategies may rebalance quarterly or semi-annually, though each rebalance creates turnover and taxes, so use tax-advantaged accounts where possible. Also prefer strategies with buffer rules that reduce unnecessary trading.
Combine momentum with value if you can tolerate the tracking error. The two factors have historically been negatively correlated at times, and a portfolio that holds both tends to have a smoother ride than either alone. That is not a guarantee, but it is a reasonable structural argument for diversification across factors.
Be honest about the behavioral cost. Long-only momentum can lag a broad index for years, and if you sell during those periods, you lock in the underperformance and miss the recovery. The strategy only works for people who can hold it through the boring and the painful parts.
Common implementation mistakes
Overfitting the lookback window is a common trap. Testing dozens of parameter combinations until one looks great is a recipe for disappointment live, so pick parameters with an economic rationale and stick to them.
Ignoring costs is equally dangerous, because momentum trades frequently. If your backtest assumes zero costs and no slippage, you should therefore subtract a realistic amount before you believe the result.
Using too much leverage is another critical error. Leverage turns a manageable drawdown into a forced liquidation, and momentum’s tail risk is exactly where leverage does the most damage.
Confusing momentum with a stock pick is a category mistake. A single stock that has gone up a lot is not a momentum strategy, because momentum is a relative ranking across a universe, applied consistently.
Abandoning the strategy at the worst moment is perhaps the most costly error. The periods when momentum looks broken are often precisely the periods when forward returns are highest. That is uncomfortable. It is the point.
Momentum persists because it is risky, not despite it. The evidence is strong enough to take seriously, and the crash risk is real enough to plan for. Traders should focus on liquidity, position sizing, and regime awareness. Investors should use long-only tilts, rebalance on a schedule, and diversify across factors. In both cases, the outcome depends less on the signal and more on whether you can hold it when it hurts.
None of this is personalized advice; instead, it is a framework for thinking about a strategy that rewards discipline and punishes conviction without risk control.
- Sector Rotation: Following Money Through the Business Cycle
- Trading Around Earnings: Before, During and After the Report
- Pullback Trading in Uptrends: Entries at Moving Averages
Frequently Asked Questions
What is momentum investing in simple terms?
Momentum investing means buying assets that have recently outperformed and avoiding or shorting those that have underperformed, based on the tendency of those trends to persist over the following months. It is usually implemented as a ranking across a broad universe of stocks, rebalanced on a fixed schedule.
Why do momentum strategies crash?
Momentum crashes typically occur after market bottoms, when beaten-down stocks rally sharply and previously strong stocks lag. If the strategy is long winners and short losers, both legs lose at the same time, and crowded positioning plus leverage can amplify the move.
Is momentum investing suitable for long-term investors?
It can be, but usually as a long-only tilt within a diversified portfolio rather than a standalone strategy. Long-only versions avoid the short leg that drives most crash risk, though they still require regular rebalancing and tolerance for periods of underperformance.
How often should a momentum portfolio be rebalanced?
Many implementations rebalance monthly, quarterly, or semi-annually, depending on the lookback window and the universe. More frequent rebalancing captures the signal faster but increases transaction costs and taxes, so the trade-off should be tested with realistic cost assumptions.
What is the difference between momentum and trend following?
Momentum is usually a cross-sectional ranking that tells you which assets to hold relative to others, while trend following uses rules like moving averages to decide whether to be long, flat, or short a given market. The two overlap, but they are implemented differently and often on different asset classes.
AI Notice: This article was created wholly or predominantly with the assistance of artificial intelligence and was published without human editorial review.
This article is for general information only and does not constitute investment advice. Always do your own research before making trading decisions.