Direct Answer
A momentum screen filters or ranks securities by raw trailing price return over a defined lookback period, such as the trailing 3, 6, or 12 months. It rests on the premise that recent relative winners tend, on average, to keep outperforming over the near term more often than chance would suggest, though results vary and no screen guarantees continued outperformance. A pure momentum screen can also concentrate results in one sector or theme and can reverse sharply once a trend breaks, which is why it is commonly combined with liquidity, volatility, or quality filters rather than used alone.
Key Takeaways
- A momentum screen ranks or filters securities by raw trailing price return over a defined lookback period, such as 3, 6, or 12 months.
- The underlying premise is that recent relative winners tend, on average, to keep outperforming over the near term more often than chance alone would predict.
- Momentum screening is related to but distinct from momentum-oscillator screens like RSI, which filter on a calculated indicator value rather than raw price return.
- Different lookback windows capture different kinds of momentum, shorter windows react faster, longer windows are smoother but slower to turn.
- Momentum screens are commonly combined with liquidity, volatility, or quality filters rather than used as a standalone selection rule.
- A pure momentum screen can concentrate results in one sector or theme and can reverse sharply once a trend breaks.
Momentum Screen: How Price-Based Momentum Screening Works
A momentum screen filters for securities showing strong recent price performance over a defined lookback period, such as the trailing 3, 6, or 12 months. It's built on the premise that securities which have recently outperformed tend, on average, to continue outperforming over the near term more often than chance would suggest, though results vary and no screen guarantees continued outperformance.
What Does a Momentum Screen Measure?
At its core, a momentum screen ranks a universe of securities by their raw trailing price return, the percentage change in price from the start of a lookback window to today. Securities with the highest trailing returns pass the screen or rise to the top of the ranking; those with weak or negative returns are filtered out or ranked lower.
This is a relative measure by construction: a momentum screen typically cares less about whether a security's price rose in absolute terms and more about how it performed relative to the rest of the universe being screened, over the same window.
Choosing a Lookback Period
The lookback period is the window over which trailing return is measured, commonly 3, 6, or 12 months, though any window can be used. The choice of window changes what the screen is actually capturing.
A shorter lookback, like 3 months, reacts quickly to recent shifts in leadership but is more sensitive to short-term noise and can churn, flagging securities today that fall out of favor a few weeks later. A longer lookback, like 12 months, smooths out short-term noise and tends to capture more durable trends, but reacts more slowly when a trend actually changes direction. Because of this tradeoff, some momentum screens combine multiple lookback windows, for example requiring strength across both a 6-month and a 12-month window, rather than relying on a single period.
Momentum Screen vs. Momentum-Oscillator Screens
Momentum screening is related to, but distinct from, momentum-oscillator-based screens like RSI (Relative Strength Index). A momentum screen typically measures raw price return over a lookback period, a straightforward percentage change. An oscillator-based screen instead filters on the calculated value of an indicator, which is itself derived mathematically from price data (RSI, for instance, is derived from the average size of a security's up moves versus down moves over a lookback period, then bounded between 0 and 100).
Because the two approaches process price data differently, they can rank or filter the same universe differently. A security can show strong raw price momentum while its RSI sits in a more moderate range, or vice versa. Traders who want both signals often run a raw-return momentum screen alongside a separate oscillator-based filter rather than treating one as a substitute for the other.
A Simple Momentum Screen in Practice
Consider a hypothetical screen applied to a universe of securities: rank every security in the universe by its trailing 6-month price return, then keep only the top decile. A security that has climbed steadily and sits near the top of that ranking passes the screen; a security that has been flat or declining over the same 6 months does not, regardless of how it performed further in the past. The screen says nothing on its own about valuation, fundamentals, or why the price moved, only that recent trailing price performance was strong relative to the rest of the universe over the chosen window.
Limitations and Common Mistakes
- Treating a screen as a forecast, passing a momentum screen describes recent price behavior, not a guarantee of continued outperformance; past returns don't predict future ones.
- Ignoring concentration risk, a pure momentum screen run during a strong sector rally can return a list dominated by one sector or theme, reducing diversification without the screen flagging that fact.
- Not planning for reversals, momentum-based rankings can shift quickly once a trend breaks, sometimes called a momentum crash; a screen alone doesn't include an exit rule.
- Confusing raw return with oscillator value, assuming a security's RSI reading and its trailing price return will always agree, when the two measures can diverge.
- Using only one lookback window, relying on a single period can overweight short-term noise or underweight a genuine recent shift, depending on which window is chosen.
A List That Can Quietly Become One Sector
Run a pure momentum screen during a strong sector move and the output will be dominated by that sector, because the screen ranks on trailing return and has no awareness of what the names have in common. Building a position across the top ten looks like diversification and can be a single concentrated bet, and nothing in the ranking flags it. Checking the sector composition of the output is a step the screen will never do for you.
The lookback is the other decision the screen hides. Three-month, six-month and twelve-month rankings surface different names from the same universe, with shorter windows turning over faster and longer ones responding slowly to a change in leadership. Neither is correct, and a ranking quoted without its window is not reproducible.
Momentum rankings can also reorder quickly. When a trend breaks, the names at the top of the list are frequently the ones that fall hardest, and the screen contains no exit rule, so a process built entirely on periodic rescreening will be selling after the reordering rather than during it.
Finally, keep this separate from oscillator screens. A momentum screen ranks raw trailing price return; an RSI screen filters on a calculated indicator value. Both use the word momentum and they answer different questions, and treating a high RSI reading as equivalent to strong trailing performance conflates the two.
Momentum Screen FAQs
What lookback period should a momentum screen use?
There's no single correct lookback period. Trailing 3, 6, and 12 months are common choices, and each captures a different kind of momentum, shorter windows react faster to recent shifts, longer windows smooth out noise but react more slowly. Many screens combine more than one window rather than relying on just one.
Is a momentum screen the same as an RSI screen?
No. A momentum screen typically ranks or filters securities by raw trailing price return over a lookback period. An RSI screen filters by the calculated value of the Relative Strength Index, a bounded oscillator derived from the size and frequency of up and down closes. They're related in spirit, both try to capture the strength of a recent move, but they measure different things and can produce different rankings.
Does momentum screening guarantee continued outperformance?
No. The premise behind momentum screening is that recent outperformers tend to keep outperforming more often than chance would suggest over the near term, not that every screened security will continue rising. Individual results vary widely, and past price performance does not predict future returns.
Can a momentum screen be combined with other filters?
Yes. Momentum screens are commonly layered with liquidity, volatility, quality, or valuation filters so the resulting list isn't just the securities with the largest recent price swings, but also meets other criteria a trader or investor cares about.
What's a common risk of relying only on a momentum screen?
A pure momentum screen can concentrate results in a single sector or theme during a strong trend, and momentum-based rankings can reverse sharply and quickly once a trend breaks, which is sometimes called a momentum crash.
Should the most recent period be excluded from the lookback?
Skipping the most recent month is a long-standing convention in the academic momentum literature, adopted because short-term returns behave differently from the intermediate horizon the effect is measured over. A screen that includes the most recent weeks is therefore measuring something slightly different from the published research. Whether to skip is a design choice, and the convention exists for a reason worth knowing.
How should a momentum screen treat a stock under a takeover offer?
The return is real and the mechanism producing it will not continue, since the price is now anchored to a deal rather than to trading. Such names can dominate the top of a momentum ranking after the announcement gap. Screens run for research purposes usually exclude securities with pending deals, and doing so is a filter that has to be applied deliberately because the price data gives no indication.
Does a momentum screen need a volatility adjustment?
Ranking on raw return systematically favours the most volatile securities, because they produce the largest moves in either direction. Dividing the return by a volatility estimate produces a risk-adjusted ranking, which returns a noticeably different list weighted toward steadier advances. Neither is the correct version; they answer different questions and should not be compared as though they were the same screen.
How often should a momentum screen be re-run?
Less often than intuition suggests, for long lookbacks. A ranking based on twelve months of return barely changes from one day to the next, so daily re-running produces the appearance of new information without any. It also generates turnover if the list drives positions. Matching the re-run frequency to the lookback keeps the output meaningful and the trading it implies proportionate.
References
Disclaimer
This page is for educational purposes only and does not constitute personalized investment advice, a recommendation to buy or sell any security, or a guarantee of any particular result. Momentum screening describes historical price behavior and does not predict future performance. Consult a licensed financial professional before making investment decisions.