Track 2 · Lesson 3 · 6 min
Trend following: why it works and when it doesn't
The idea behind the biggest family of plays, the research that supports it, and the price you pay for using it.
The idea
Trend following means owning something while its price is rising and stepping aside once it starts falling. It doesn't try to predict anything. It reacts. A trend follower doesn't ask "is this stock cheap?" or "will the economy slow down?" It asks one question: is the price above or below its trend line?
It's the biggest family on Be The Puck: 30 plays, from Paul Tudor Jones's 200-day rule to the Ichimoku cloud. Breakout and momentum plays are close cousins. Mean-reversion and calendar plays work differently and get their own lessons.
Why it can work
Prices tend to keep going in the direction they've been going, for a while. Researchers call this momentum. In 2012, Tobias Moskowitz, Yao Hua Ooi and Lasse Pedersen showed that an asset's return over the past year helped predict its return over the next month, across stocks, bonds, currencies and commodities, going back decades.
The usual explanations are human. News sinks in slowly, so prices adjust in steps rather than all at once. People pile into what's working and dump what isn't. And big investors take weeks to move money, which keeps a trend going.
When it doesn't
Trend following has two built-in weaknesses.
Whipsaws. In a choppy, sideways market the price keeps crossing back and forth over the line. The play buys, sells a little lower, buys again a little higher, and loses a bit each time. Faster plays suffer most: on SPY, the daily PB EMA band switches about 29 times a year, while the monthly 10-month play switches about 1.5 times.
Lag. The play always gets out after the top and back in after the bottom. When a market drops sharply and recovers quickly, as in early 2020, the play can sell near the low and buy back higher, missing both sides.
The result is that timing plays spend long stretches trailing buy-and-hold. Meb Faber's own ten-year review of his 10-month rule found it shone in the 2008 crash, then lagged stocks in six of the next eight years. Professor Valeriy Zakamulin, testing 155 years of US data, found the rules' advantage came mainly from a handful of severe bear markets.
Every trend play on the S&P 500
On SPY since 2005, after trading costs. Click a column to sort.Sort with the menu above the list.
| Play | Checks | Switches / yr | Time in | Annual return | Worst drawdown | Calls right |
|---|---|---|---|---|---|---|
| PB EMABitcoin Playboy | Daily | 28.9 | 67% | +3.4% | −48.9% | 28% |
| Bull Market Support BandBenjamin Cowen | Weekly | 4.0 | 77% | +8.0% | −21.7% | 31% |
| Ripster EMA CloudRipster | Daily | 15.4 | 72% | +6.3% | −29.4% | 26% |
| Trend TemplateMark Minervini | Daily | 4.0 | 21% | +2.6% | −17.4% | 24% |
| Stage AnalysisStan Weinstein | Weekly | 2.9 | 74% | +8.9% | −17.6% | 29% |
| Stage 2 + Mansfield RSStan Weinstein | Weekly | 3.9 | 79% | +8.2% | −21.4% | 27% |
| 10-Month SMAMeb Faber | Month-end | 1.5 | 81% | +8.7% | −24.3% | 38% |
| 200-Day RulePaul Tudor Jones | Daily | 6.2 | 80% | +8.3% | −20.9% | 17% |
| Golden CrossMarket tradition | Daily | 0.9 | 80% | +9.4% | −33.7% | 50% |
| Granville 200-DayJoseph E. Granville | Daily | 5.0 | 77% | +8.2% | −20.9% | 17% |
| Hull Moving AverageAlan Hull | Weekly | 6.1 | 67% | +6.2% | −24.4% | 39% |
| Kaufman Adaptive MAPerry J. Kaufman | Daily | 38.4 | 65% | +0.6% | −47.7% | 24% |
| TEMA CrossoverPatrick G. Mulloy | Daily | 12.8 | 49% | +4.3% | −34.6% | 34% |
| Seykota 15/150Ed Seykota | Daily | 0.8 | 85% | +8.6% | −30.0% | 39% |
| Guppy Multiple MAsDaryl Guppy | Daily | 6.3 | 69% | +5.6% | −21.2% | 29% |
| Ichimoku CloudGoichi Hosoda | Daily | 4.6 | 74% | +8.0% | −20.5% | 35% |
| SupertrendOlivier Seban | Daily | 8.2 | 64% | +5.5% | −23.7% | 37% |
| Williams AlligatorBill Williams | Daily | 14.7 | 61% | +5.1% | −17.3% | 38% |
| Vortex IndicatorEtienne Botes & Douglas Siepman | Daily | 23.5 | 65% | +2.8% | −34.3% | 33% |
| AroonTushar S. Chande | Daily | 8.2 | 65% | +7.6% | −15.8% | 42% |
| Parabolic SARJ. Welles Wilder Jr. | Daily | 25.1 | 62% | +2.0% | −48.3% | 36% |
| ADX TrendJ. Welles Wilder Jr. | Daily | 6.4 | 23% | +2.6% | −11.4% | 35% |
| MACD Zero LineGerald Appel | Daily | 8.4 | 71% | +6.5% | −26.4% | 30% |
| Zweig 4% ModelMartin Zweig | Weekly | 5.4 | 78% | +6.3% | −39.0% | 36% |
| MAMA/FAMA CrossJohn F. Ehlers | Daily | 10.4 | 68% | +5.0% | −23.3% | 32% |
| Schaff Trend CycleDoug Schaff | Daily | 12.3 | 63% | +4.4% | −43.7% | 37% |
| Zero Lag EC CrossJohn F. Ehlers & Ric Way | Daily | 6.2 | 63% | +6.0% | −23.6% | 34% |
| DP Trend ModelCarl Swenlin & Erin Swenlin | Daily | 3.4 | 76% | +8.8% | −25.0% | 39% |
| Elder Triple ScreenAlexander Elder | Daily | 10.1 | 30% | +4.4% | −28.1% | 36% |
| Holy GrailLinda Bradford Raschke & Larry Connors | Daily | 0.6 | 0% | +1.7% | −2.5% | 33% |
| Buy & holdbenchmark | Never | 0 | 100% | +11.0% | −55.2% | – |
What it's really for
Across all 11,242 trend-play-and-stock pairs on Be The Puck, the play cut the worst drawdown in 9,432 cases but beat buy-and-hold's annual return in only 1,949. That's the honest summary: trend plays are mainly a way to limit how much you can lose in a crash, and you usually pay for it with lower growth in good years.
Try it
Sort the table above by "Switches / yr", then by "Worst drawdown". Then read the skeptic's scoreboard to see the same pattern across every play.
Check yourself