Momentum Trading Explained
Momentum is one of the few anomalies that has survived decades of academic scrutiny across almost every market tested. It also fails spectacularly at turns.
Momentum is the observation that assets which have performed well recently tend to continue performing well over the medium term. It contradicts market efficiency, and it has been documented across equities, currencies, commodities and bonds over more than a century.
Why it persists
- Under-reaction to news. Information is absorbed gradually rather than instantly, so prices adjust over weeks.
- Institutional accumulation. Large positions cannot be built in one order. Sustained buying pressure follows the initial decision.
- Anchoring. Participants anchor on previous price levels and revise their views too slowly.
- Flows follow performance. Money moves toward what has recently done well, which mechanically reinforces the move.
Notably these are behavioural and structural rather than informational, which is why the effect has not been arbitraged away despite being extensively published.
How it is measured
The classic academic formulation ranks assets by their return over the past twelve months, excluding the most recent month. That exclusion matters: very short-term returns tend to reverse rather than continue, so including them dilutes the signal.
Simpler implementations use price relative to a long moving average, or the rate of change over a fixed lookback.
The crash risk
Momentum earns steadily and gives back violently.
Momentum strategies experience infrequent but severe drawdowns, typically at sharp market turns. By construction the strategy is most exposed to whatever has run furthest, which is exactly what unwinds fastest when sentiment reverses. Momentum crashes are well documented and are the price of the long-run return.
Timeframe sensitivity
The effect is horizon-dependent in a way that catches people out. Very short horizons show reversal. Medium horizons — roughly three to twelve months — show continuation. Multi-year horizons show reversal again, which is the value effect. Using the wrong lookback does not weaken the signal; it inverts it.
Practical notes
Momentum works best applied across a diversified set rather than a single instrument, because the effect is statistical and any individual position may not cooperate. Volatility-scaled position sizing meaningfully reduces the crash exposure, since momentum drawdowns typically coincide with volatility spikes.
Educational content only. Not financial advice. Past performance does not guarantee future results.
See the models in action
Reading about forecasting is one thing. Create a free account and get 5 AI predictions a month — no card required.
Start free