What Is Momentum Trading?
Momentum trading means buying what has already risen the most, usually ranked against other instruments rather than judged in isolation. It differs from trend following in that the entry is a ranking rather than a signal, and the holding period is fixed rather than open-ended.
Buy strength. It sounds identical to trend following and is a different thing in two specific ways that change how it behaves.
How it works
Measure how much each instrument has moved over a fixed lookback. Rank them. Buy the top of the list. Re-rank periodically and rotate.
That is a portfolio procedure, not a chart signal, and it is where the approach came from — academic work on cross-sectional momentum in equity returns.
It is not trend following
Two differences, both consequential.
The entry is a ranking, not a condition. Trend following asks “is this trending?” and can answer no for everything. Momentum asks “which of these has moved most?” and always has an answer — including in a market where every candidate is falling and the winner is merely falling least.
The exit is usually a date. Rebalance monthly, or quarterly. Trend following exits when the trend ends, which could be tomorrow or in a year.
So momentum is always in the market and trend following is not, and that difference is larger than anything in how either one enters.
The premise, measured
Momentum needs a move to be followed by more of the same — the same persistence trend following needs, measured the same way.
Across the full 574 bars: +0.043, with a standard error of 0.042. One standard error from zero, which on this series is no relationship at all.
And the single strongest bar was followed by one less than half its size. Which is what “no relationship” looks like in one instance: not a reversal, just no reason to expect a repeat.
One caveat that matters. The published evidence for momentum is cross-sectional — comparing many instruments — and this is one series. A single chart cannot test a ranking strategy, and this page does not claim it has.
The lookback is the strategy
Twelve bars and three bars rank the same instruments differently, most days.
There is no correct lookback, and the choice is not a sensitivity setting — it defines what you are buying. A twelve-month ranking and a one-week ranking are different strategies with the same name.
The published work mostly used 3 to 12 months, with the most recent month skipped — a detail worth knowing, because the very recent past tends to reverse and including it works against you.
Absolute and relative are different strategies
The word covers two procedures and they behave very differently, which the shared name hides.
Relative momentum ranks instruments against each other and buys the top of the list. Everything on this page so far has described that.
Absolute momentum asks whether an instrument has risen over the lookback at all, and holds cash if it has not. It is sometimes called time-series momentum.
The second has an off switch and the first does not. In a market where every candidate is falling, relative momentum is fully invested in the least bad one and absolute momentum is flat.
That single difference changes the risk profile more than any parameter. The failure described below — a ranking that always returns a winner — belongs to the relative version and is precisely what the absolute version was designed to remove.
Most people running “momentum” are running the relative kind without having chosen it, because it is the one the academic literature is usually summarised as, and the summary rarely mentions that there is a second.
A worked example
Fix the universe first. Momentum is a comparison, so what you are comparing decides everything.
Fix the lookback and the rebalance date, and write both down. These are the strategy.
Rank, buy the top, and rotate on schedule — not when it feels right, because the discipline is the only thing holding the procedure together.
Size each position from risk management, and note that a ranking gives you no invalidation price. You will need to supply one from the chart, which is the point at which this stops being a pure momentum strategy and becomes a better one.
The original data
Across our study of 24,971 trading videos, 106 cover momentum trading. The median one gets 5,394 views, 86% never pass 50,000, and the median length is 11.1 minutes.
The corpus carries description text for only 6 of those 106, far too few to say anything about how the topic is written.
106 videos is a small field for an idea with more published academic support than almost anything else in this glossary — which is itself worth noticing.
When it fails
The reversal after the run
Buying the strongest bar and looking six bars later shows most of the move gone.
Momentum’s characteristic failure is a sharp reversal in what was leading, and because a ranking concentrates you in whatever has run furthest, you are maximally exposed at exactly that moment.
Sideways there is still a winner
A ranking always returns a top of the list, even when nothing is moving. The strategy has no way to say “not today” — which is exactly the output the average directional index (ADX) page argues is the most valuable one an indicator can give.
Relative strength is not strength
Everything can be falling. The best-ranked instrument in a declining market is a long position in something losing money more slowly, and the ranking reports it as a buy.
You looked once the run had finished
Every momentum trade that worked looked exactly like this one, and so did every one that bought the last bar of a move.
Related
Trend following is the approach this is most often confused with, and the two differences above are what separate them.
Swing trading is the holding period momentum usually operates over.
And average true range (ATR) is how you size a position that a ranking gave you no stop for.
The distinction I had to learn is that momentum is a comparison and a trend is a description. Momentum says this is moving more than those; a trend says this is moving. They sound the same and they behave completely differently, because a ranking always has a winner - even in a market where everything is falling and the winner is simply falling least.
— Michael Whitman, from this video
This page is educational, not financial advice. Test every idea on your own charts before risking money.