Trend following: how the strategy works, the evidence, and where it fails

Trend following is one of the oldest and best-documented approaches in investing. The idea fits in one sentence: own markets that are rising, step aside from markets that are falling. It doesn't forecast. It reacts to what prices have already done, with rules that are fixed in advance and applied the same way every time.

This guide explains why trends exist, what a century of research says, how the approach behaved in the big crashes, where it struggles, and how investors use it today, including trend following ETFs.

What is a trend following strategy?

A trend following strategy holds an asset while its price trend is up and reduces or exits the position when the trend turns down. The money that leaves goes into something defensive, usually cash or short-term government bonds (T-bills). Professional versions can also go short, betting on falling prices, and spread their risk across dozens of markets.

Three features define it:

• Rules, not opinions. The signal is calculated from prices, so two people following the same rules make the same decision.
• Reaction, not prediction. A trend follower is always late at the turning points: it sells after the top and buys after the bottom. The aim is to capture the large middle part of big moves.
• Many small losses, a few big wins. When a market moves sideways, the strategy gets in and out repeatedly at a small cost. The pay-off comes from the long trends.

Why do price trends exist?

If markets were perfectly efficient, past prices would say nothing about future prices. Research points to several reasons why trends persist in practice:

• Information spreads slowly. Investors react to news in stages, so prices adjust over months rather than instantly.
• Herding and momentum chasing. Rising prices attract buyers, falling prices trigger more selling.
• Slow-moving big money. Central banks, pension funds and governments change positions gradually, sometimes for non-economic reasons.
• Forced selling. Margin calls and risk limits make declines self-reinforcing in crises.

A century of evidence

Trend following is not a backtest found last year. Some of the most cited studies:

• Moskowitz, Ooi and Pedersen (2012), "Time Series Momentum", Journal of Financial Economics: across 58 futures markets (stocks, bonds, currencies, commodities), an asset's own past return predicted its future return for up to about a year, and a diversified trend strategy earned positive returns with little correlation to stock markets.
• Hurst, Ooi and Pedersen (2017), "A Century of Evidence on Trend-Following Investing": going back to 1880, a diversified trend strategy was profitable in every decade they studied and tended to do well in the worst stock market periods.
• Managed-futures funds (CTAs) have run trend strategies for professional money since the 1970s and 1980s, the best known being the "Turtle" traders trained by Richard Dennis in 1983.

The evidence is strongest for diversified trend following across many markets. On a single market the results vary much more.

How trend following behaves in crashes

The main reason investors use trend following is what happens in long bear markets. Because most crashes unfold over months, a trend strategy usually steps aside before the deepest part of the decline. Here is the S&P 500 since 1993, comparing buy & hold with the InvestWealth systematic strategy from our Research Terminal:

The difference is mostly in the falls. Buy & hold lost about 55% from 2007 to 2009; the strategy's worst fall over the whole period was about 30% (backtested, after estimated costs, hypothetical). Across the 30 markets in the Terminal, the pattern is similar: the worst fall was smaller in almost every market, while the return per year was higher in most but not all of them. The Terminal shows every market side by side.

The same logic explains the well-known "crisis alpha" of managed-futures funds: in 2008 most of them gained while stock markets fell, because they had moved out of equities and into falling or defensive markets as the trends developed.

Where trend following fails

No strategy works all the time, and trend following has clear weak spots:

• V-shaped crashes. In February–March 2020 the S&P 500 fell about 32% in five weeks and recovered almost as fast. A trend strategy that waits for confirmation tends to sell near the bottom and buy back after much of the rebound.
• Sideways markets. When prices move back and forth without direction, the strategy collects many small losses ("whipsaws").
• Long periods behind buy & hold. In a steady bull market a trend strategy is almost always invested anyway, and the occasional false exit costs return. Managed-futures funds went through a weak decade in the 2010s.
• Discipline. The rules only work if they are followed, including after a run of small losses. Most investors abandon a strategy at exactly the wrong moment.
• Costs and taxes. Each switch has trading costs and, in a taxable account, can realise gains.

This is why trend following is best judged on its worst fall and its return per unit of risk (see the Sharpe ratio), not on whether it beat buy & hold in a given year.

Trend following vs momentum vs market timing

• Trend following (also called *time-series momentum*) compares a market with its own past. It can be fully out of the market.
• Momentum investing (*cross-sectional momentum*) compares assets with each other and buys the recent winners, for example the strongest stocks in an index. It is usually always invested.
• Market timing usually means forecasting the next move from economic data or opinion. Trend following doesn't forecast; it reacts. Our guide Does market timing work? shows why that distinction matters.

Trend following ETFs and managed futures

Private investors can access trend following in three ways:

• Managed-futures ETFs and funds. These run diversified trend strategies across futures on stocks, bonds, currencies and commodities, long and short. In the US, examples include DBMF, KMLM, CTA and WTMF; mutual funds from AQR and others follow the same approach. Fees are higher than index funds, and results differ widely between managers.
• Trend-based ETF strategies. Some funds switch a single index between equities and Treasuries based on its trend.
• Doing it yourself with index ETFs and a fixed set of rules, reviewed on a fixed schedule. It is cheap, but it needs the discipline to follow the rules for years.

Whichever route you choose, check three things before investing: the worst fall in backtests and live results, how long the strategy has been live, and the total cost including trading.

How investors use trend following

• As a diversifier next to a core of stocks and bonds, because its returns tend to be independent of both, especially in long bear markets.
• As a risk overlay on an equity portfolio, to limit the depth of the worst falls.
• As a process: a monthly review with fixed rules, which replaces reacting to headlines.

The free Research Terminal shows the InvestWealth systematic strategy next to buy & hold for 30 markets, including the worst fall, the Sharpe ratio and the equity curve for each. The rules themselves are not published.

Frequently asked questions