Trend Following and Crisis Alpha: The Hedge That Worked in 2008 and 2022

In 2022 the S&P 500 fell 19.4% on a price basis and, unusually, bonds fell with it. The classic 60/40 portfolio had one of its worst years in decades. Yet the SG Trend Index, which tracks the largest trend-following managers, gained 27.35%, its best year on record (Société Générale, via Hedgeweek). In 2008, when the S&P 500 fell 38.5%, the Barclay CTA Index gained 13.9%. That pattern has a name: crisis alpha. This guide explains where it comes from, when it fails, and how it fits next to the option hedges in our tail risk hedging pillar.

What is trend following?

Trend following is a strategy that goes long markets whose prices have been rising and short or out of markets whose prices have been falling, using nothing but past prices. It does not forecast; it follows.

Academics call it time-series momentum. In their 2012 paper in the Journal of Financial Economics, Moskowitz, Ooi and Pedersen studied 58 liquid futures and forwards across equity indices, currencies, commodities and government bonds and found that each market’s own past 12-month excess return predicted its return over the following months. That is different from the better-known cross-sectional momentum, which ranks stocks against each other: time-series momentum asks only whether a market is in an uptrend or a downtrend versus its own history.

Professional trend followers are usually commodity trading advisors (CTAs) running managed-futures programs. They trade dozens to hundreds of futures markets, size each position to a volatility target, and can be short as easily as long. Individual investors can access the strategy through managed-futures mutual funds and ETFs, or apply a simplified version to their own stock allocation with a moving-average rule.

What is crisis alpha, and why does trend following earn it?

Crisis alpha is the tendency of trend-following strategies to make money during prolonged market crises. It arises because most bear markets unfold over months, which gives trend signals time to turn short equities and long the assets that benefit.

The deepest evidence is AQR’s A Century of Evidence on Trend-Following Investing by Hurst, Ooi and Pedersen (Journal of Portfolio Management, 2017). Using 67 markets from 1880 to 2016, with simulated transaction costs and hedge-fund-style 2-and-20 fees, they found that a time-series momentum strategy had positive returns in 8 of the 10 largest drawdowns of a traditional 60/40 portfolio. Their explanation is about speed: the average peak-to-trough length of those ten drawdowns was about 15 months, long enough for trend followers to reposition. They also note that the strategy “may not perform well in bear markets that occur very rapidly, such as the 1987 stock market crash.”

There is a second source of crisis returns. In many crises, falling stocks come with other strong trends — falling oil, rising bonds, a rising dollar — and a trend follower can be on the right side of all of them. In 2022 the profits came less from shorting stocks than from being short bonds and long commodities and the dollar as inflation forced rates up. That is exactly the environment in which the stock-bond correlation turned positive and bonds stopped hedging stocks.

How did trend following perform in 2008 and 2022?

It made money in both years. The Barclay CTA Index, an average of managed-futures programs, gained 13.9% in 2008; in 2022 the SG Trend Index gained 27.35% and the broader SG CTA Index 20.07%, its best year since the index began in 2000.

Both years were slow bears. The S&P 500 took about 17 months to fall 56.8% from its October 2007 peak to its March 2009 low, and about nine months to fall 25.4% in 2022 (closing basis, FRED). In both, the trend turned early and stayed turned. These are published index returns, net of fees, so they are what a diversified basket of managers actually delivered — individual funds varied widely around them.

Why does trend following struggle in sudden crashes?

Because the signal arrives after the damage. A trend rule needs the decline to persist before it flips; in a crash that happens in days, the strategy is still long, or only just turning, when the bottom arrives.

The February–March 2020 crash is the modern example. The S&P 500 fell 33.9% from its February 19 closing peak to its March 23 low in 33 days. A month-end moving-average rule was still invested at the end of January and only switched to cash at the end of February, after a fall of about 13%; it then sat in cash through March and April, missing much of the rebound, and re-entered at the end of May. Real CTA programs, which trade daily and use faster signals, adapt more quickly but still face the same problem in principle.

The reversal is also dangerous. After a V-shaped bottom, a trend follower that has turned short can lose on the way back up. This is whipsaw, and it is the main running cost of the strategy.

How does a simple 10-month moving-average rule work?

At each month end, hold stocks if the index closes above its average of the last 10 month-end closes; otherwise hold Treasury bills. The rule was popularized by Mebane Faber’s 2007 paper on tactical asset allocation, and it is the simplest trend hedge an individual investor can run.

Our backtest applies the rule to the S&P 500 from December 1990 to October 2026, using month-end closes and 3-month T-bill yields for the cash periods (FRED TB3MS). It is illustrative: price index only, no dividends, taxes or trading costs.

The trade-off is clear. The rule gave up about half a point a year of return, and in exchange cut the worst drawdown by more than half. Given the recovery math — a 52.6% loss needs a 111% gain — that is a trade many investors would take. But note 2022: the rule went out at the end of February, back in at the end of March and out again at the end of April, and still lost 16.3% at month ends.

Is trend following a better hedge than options?

Neither is better in every crash. Puts pay on a contract in any decline past the strike and do best in sudden crashes; trend following costs no premium and does best in slow bears and rate shocks. Their weaknesses are almost mirror images.

The matrix shows why many institutions hold both. A put bought before February 2020 paid out within weeks, when trend was still long. In 2022, a 10% out-of-the-money put paid only once the index fell far enough, and the VIX never closed above 36.45, so VIX calls did little, while trend funds had a record year. And in the long bull years, puts cost a steady premium — about 1–3% a year for index protection at today’s prices, as our guide to the cost of hedging shows — while trend costs fees and occasional whipsaws.

For concentrated stock, neither replaces a position-level hedge. A trend rule on the S&P 500 does nothing about your employer’s stock falling alone. A trend rule on the stock itself is possible but can force taxable sales; for most holders a collar is cleaner.

How do you add trend following to a portfolio?

There are three practical routes, each with different costs:

• Managed-futures funds and ETFs. The most diversified route: dozens of markets, long and short. Check the expense ratio, how closely the fund tracks a trend benchmark, and whether gains are distributed annually. Futures are complex leveraged instruments; the NFA’s Opportunity and Risk guide explains the risks.
• A moving-average rule on your index allocation. Cheap and transparent, but it is a single-market trend signal, so it has none of the cross-asset diversification that drove 2022 returns.
• A trend overlay that triggers hedges. When the index closes below its 10-month average, add index put spreads instead of selling stock. This keeps you invested and avoids realizing gains, at the cost of premium.

On sizing, AQR’s study simulated moving 20% of a 60/40 portfolio into trend following and found it reduced maximum drawdown and volatility over 1880–2016 while raising returns. Real-world allocations tend to run 5–20%, because a trend sleeve that lags stocks for several years tests anyone’s patience. The crash-protection ladder shows where it sits alongside cash and options.

What should you check before using trend following as a hedge?

1. Decide which crash you are hedging: slow bear (trend fits) or sudden crash (options fit). If both, plan for both.
2. Choose the route: diversified managed-futures fund, a simple rule, or a rule that triggers option hedges.
3. For funds, compare fees, the 2008, 2020 and 2022 track record, and tax distributions.
4. For rules, decide the signal (10-month average), check frequency (month end only) and account type (tax-deferred if possible).
5. Size it so you can hold it through three flat years: typically 5–20% of the portfolio.
6. Keep position-level protection, such as collars, for any concentrated stock; trend does not cover it.

Frequently asked questions

Key takeaways

1. Trend following goes long rising markets and short falling ones, across stocks, bonds, currencies and commodities.
2. Crisis alpha is real but conditional: managed futures gained in 2008 (+13.9%, Barclay CTA) and 2022 (+27.35%, SG Trend).
3. AQR’s 1880–2016 study found trend positive in 8 of the 10 worst 60/40 drawdowns.
4. Trend lags crashes that happen in days; puts and VIX calls cover that gap, so the two pair well.
5. A 10-month moving-average rule cut the S&P 500’s worst month-end drawdown from about 53% to 22% in a 1990–2026 backtest, at a small cost in return.
6. Costs are whipsaws, fund fees and, in taxable accounts, realized gains: size it as a complement, not a replacement.

Sources and method

1. Hurst, Ooi and Pedersen, A Century of Evidence on Trend-Following Investing, Journal of Portfolio Management (2017) — performance since 1880 and in the 10 worst 60/40 drawdowns
2. Moskowitz, Ooi and Pedersen, Time Series Momentum, Journal of Financial Economics (2012) — the academic definition of time-series momentum across 58 futures
3. Hedgeweek, Flat December sees CTAs complete record year (Jan 2023) — SG Trend Index +27.35% and SG CTA Index +20.07% in 2022, Societe Generale data
4. Hedgeweek, Barclay CTA Index gained 13.9 per cent in 2008 — managed futures return in 2008
5. FRED, S&P 500 (SP500) — S&P 500 closes for calendar-year returns and the backtest (data © S&P Dow Jones Indices)
6. FRED, 3-Month Treasury Bill Secondary Market Rate (TB3MS) — cash return in the backtest
7. National Futures Association, Opportunity and Risk — how futures and managed-futures accounts work and their risks

Managed-futures returns are published index figures, net of fees, as reported by the index providers. S&P 500 calendar-year returns are price returns from daily closes. The 10-month rule is a Proflex backtest from December 1990 to October 2026 on month-end S&P 500 closes (price index, dividends excluded), switching to 3-month Treasury bills (FRED TB3MS) when the close is below its 10-month average, with no taxes, trading costs or slippage. Month-end data understates intra-month lows. Backtests are not forecasts and not a recommendation of any fund.