Trending Fast and Slow (volatility-switched momentum)

Trending Fast and Slow is a stock market timing rule from Eddie Cheng, Nazar Kostyuchyk, Wai Lee, Pai Liu, and Chenfei Ma of Allspring Global Investments, published in the Journal of Portfolio Management in 2022. It lets market volatility choose the lookback: when the S&P 500 has been volatile it follows the 1-month trend, and otherwise the 12-month trend, holding stocks in an uptrend and Treasury bills in a downtrend.

The idea#

Trend followers face a trade-off. A slow signal, such as the 12-month return, ignores short dips and keeps you invested through long bull markets, but it reacts late to a crash. A fast signal, such as the 1-month return, reacts quickly but gets faked out by ordinary pullbacks.

The authors found that which one works better depends on the volatility regime. In calm markets slow momentum did better; in turbulent markets fast momentum did. They used a decision tree to find the dividing line, about 17% annualized volatility over the past month, and switched signals at that level. They found the same pattern in most developed and emerging stock markets.

How it works#

At the close of the last trading day of each month:

1. Measure the annualized volatility of the S&P 500's (SPY) daily returns over the past month (21 sessions).
2. If it is 17% or higher, use the 1-month (21-session) return; otherwise use the 12-month (252-session) return.
3. If that return is positive, hold 100% SPY; otherwise hold 100% Treasury bills (BIL).

The paper goes short the S&P 500 when the trend is down. This long-or-cash version is the one Allocate Smartly tests.

What the backtest shows#

The ETF-era test starts in mid-2007, when the Treasury bill fund begins. Since then the strategy returned about 9% a year, against 11% for the S&P 500 and 8% for a 60/40 stock and bond mix. Its worst drawdown was about 25%, less than half the S&P 500's and smaller than the 60/40's, and its risk-adjusted return was above the S&P 500's but just below the 60/40's.

It limited the 2008 crash to a 9% loss against 37% for the S&P 500; its worst drawdown ran from July 2007 to January 2009 and was recovered by December 2009. In many calm years it simply matched the market. But in 2022 it lost 22%, more than the S&P 500's 18%, in a volatile year when a single up month could switch the fast signal back into stocks.

Since publication at the start of 2022 it has returned about 6% a year, and at its worst it fell about a quarter from its peak.

With simulated history the test starts in 1960 and returns about 10% a year, a little below the S&P 500, with a better risk-adjusted return and a worst drawdown of about 27% against 55%, from January 1973 to January 1974. It was roughly flat in 1974 while stocks lost 27%, and it gained 37% in 1987 while the market made 5%.

When it struggles#

• Choppy bear markets. High volatility triggers the fast signal, and a single up month can pull it back into stocks just before the next leg down.
• Slow declines in calm markets. When volatility stays low, the 12-month signal reacts late.
• All-or-nothing. It is always 100% stocks or 100% cash, with no diversification.

Using it on Tactfolio#

The live strategy above runs the long-or-cash rules on SPY, measuring volatility and trend on SPY's dividend-adjusted daily returns rather than the paper's futures prices and log returns. Copy it to try another threshold or lookbacks, or compare it with the single slow signal of Faber's 10-month moving average.

Year by year

* Partial year.

The rules as implemented

This is the exact tree Tactfolio runs, rebalanced monthly with signals and trades at the close. Open it to inspect or copy it.

Sources and caveats

• Eddie Cheng, Nazar Kostyuchyk, Wai Lee, Pai Liu, and Chenfei Ma, Trending Fast and Slow, Journal of Portfolio Management (2022)

• The paper goes short S&P 500 futures when the chosen trend is down; this version holds Treasury bills instead, as Allocate Smartly tests it.
• Volatility is the annualized standard deviation of SPY's simple daily returns over 21 sessions, and trends are SPY's 21- and 252-session total returns; the source uses log returns and futures or price-index changes.
• The paper sits behind a paywall and its exact test period could not be confirmed; the development window covers the S&P 500 futures era up to its release.
• Signals and trades use the close of the last trading day of each month, as in the source.

Common questions#

What is Trending Fast and Slow?#

It is a timing rule from Allspring researchers that uses the S&P 500's 1-month volatility to choose between a fast (1-month) and a slow (12-month) momentum signal, then holds stocks when the chosen signal is positive.

Why 17% volatility?#

The authors chose the threshold with a decision tree that split past months into low and high volatility regimes. Monthly volatility sits below it most of the time.

What ETFs does Trending Fast and Slow use?#

SPY for the S&P 500 and BIL for cash. The paper traded S&P 500 futures and went short instead of to cash.

Does Trending Fast and Slow still work?#

Since publication in 2022 it has earned about 6% a year and fell more than the S&P 500 in 2022. The live record on Tactfolio shows how it is doing now.

Related strategies

• Faber’s 10-month moving average timing model
• 12-Month High Switch (Meb Faber)
• Gold Cross-Asset Momentum (Quantpedia gold and Treasury rule)
• Leverage Rotation Strategy (LRS): UPRO above the 200-day average