Track 2 · Lesson 6 · 7 min
Momentum: own what's rising
Time-series versus cross-sectional momentum, relative strength, dual momentum, and the indicators built on rate of change.
Time-series momentum
Momentum is the tendency of rising prices to keep rising for a while. The simplest way to use it is to compare an asset with its own past: has it gone up over the last year, or at least beaten cash? If yes, own it; if not, sit in T-bills. That is time-series momentum, the rule behind Moskowitz, Ooi and Pedersen's 2012 paper and Be The Puck's 12-month momentum play.
Cross-sectional momentum
The older academic version compares stocks with each other. Robert Levy tested it in 1967 by buying stocks trading well above their 27-week average. Narasimhan Jegadeesh and Sheridan Titman's 1993 paper is the classic: ranking U.S. stocks by their past 3 to 12 months' return, recent winners kept beating recent losers over 1965–1989. Later researchers usually skip the most recent month, because very short-term returns tend to reverse; that's the "12-1" measure.
Be The Puck tests one stock at a time, so it can't rank a whole market. Its versions either ask whether the stock's own momentum is positive, or compare it with SPY as a stand-in for the ranking.
Relative strength and dual momentum
Relative strength means beating the market. Divide a stock's price by the S&P 500's and you get a line that rises when the stock is outperforming. Stan Weinstein reads it from Mansfield charts and wants it positive before buying. Robert Levy's 27-week ratio is another version.
Gary Antonacci combined the two kinds into dual momentum: an asset must beat T-bills (absolute momentum) and beat the alternatives (relative momentum). In his book he applies it to whole markets, U.S. versus international stocks, with bonds as the safe asset. Andreas Clenow measures momentum as the slope of a 90-day trend line, multiplied by how well the line fits, so smooth climbers score higher than jumpy ones.
Nvidia with Dual Momentum, last two years
The panel shows the three 12-month returns the play compares each month-end: the stock, SPY and T-bills.
12-month returns
Indicators built on rate of change
A rate of change is just the percentage move over a set period. Several famous indicators smooth and combine them:
- MACD (Gerald Appel): the 12-day EMA minus the 26-day EMA. Above zero, the fast average is on top; crossing its 9-day signal line is the most common signal.
- KST (Martin Pring): four smoothed rates of change over different look-backs, weighted and added together.
- Coppock curve (Edwin Coppock): a monthly blend of the 11- and 14-month rates of change, designed to spot the end of bear markets. Coppock reportedly took those periods from how long Episcopal bishops said mourning normally lasts.
- CCI (Donald Lambert): how far the price has run from its average, scaled so ±100 marks unusual moves.
When momentum breaks
Momentum's worst moments tend to come right after a bear market, when the beaten-down stocks rebound hardest and the recent leaders lag; Kent Daniel and Tobias Moskowitz documented these "momentum crashes" in 2016. Slow, monthly momentum plays are also late to get back in after a sharp V-shaped recovery.
Every momentum play on the S&P 500
On SPY since 2005, after trading costs. On SPY itself the relative tests compare SPY with itself, so only their absolute parts matter.
| Play | Checks | Switches / yr | Time in | Annual return | Worst drawdown | Calls right |
|---|---|---|---|---|---|---|
| 12-Month MomentumMoskowitz, Ooi & Pedersen | Month-end | 0.6 | 83% | +9.7% | −33.7% | 43% |
| 12-1 MomentumNarasimhan Jegadeesh & Sheridan Titman | Month-end | 0.8 | 85% | +11.3% | −33.7% | 56% |
| Dual MomentumGary Antonacci | Month-end | 0.6 | 83% | +9.7% | −33.7% | 43% |
| Stocks on the MoveAndreas F. Clenow | Daily | 6.9 | 67% | +6.7% | −23.0% | 23% |
| 52-Week HighThomas J. George & Chuan-Yang Hwang | Month-end | 0.4 | 87% | +9.2% | −33.7% | 62% |
| Levy Relative StrengthRobert A. Levy | Weekly | 4.0 | 77% | +8.3% | −20.3% | 30% |
| MACD Signal CrossGerald Appel | Daily | 21.3 | 51% | +5.4% | −23.1% | 35% |
| Elder ImpulseAlexander Elder | Daily | 34.3 | 66% | +1.5% | −40.6% | 30% |
| Know Sure ThingMartin J. Pring | Daily | 14.2 | 48% | +4.3% | −26.3% | 38% |
| Coppock CurveEdwin S. C. Coppock | Month-end | 0.1 | 82% | +12.4% | −33.7% | 67% |
| CCI +100Donald R. Lambert | Daily | 35.6 | 32% | −0.9% | −46.5% | 28% |
| TRIX Signal CrossJack K. Hutson | Daily | 12.8 | 50% | +3.5% | −34.4% | 38% |
| TSI Zero LineWilliam Blau | Daily | 8.0 | 70% | +6.8% | −25.5% | 31% |
| 13612W MomentumWouter J. Keller & Jan Willem Keuning | Month-end | 2.7 | 79% | +8.6% | −21.0% | 40% |
| Market-State MomentumMichael J. Cooper, Roberto C. Gutierrez Jr. & Allaudeen Hameed | Month-end | 0.9 | 77% | +9.8% | −33.7% | 39% |
| Frog in the PanZhi Da, Umit G. Gurun & Mitch Warachka | Daily | 0.6 | 82% | +11.2% | −33.7% | 64% |
| PMO Signal CrossCarl Swenlin | Daily | 17.3 | 52% | +4.1% | −32.1% | 32% |
| Buy & holdbenchmark | Never | 0 | 100% | +11.0% | −55.2% | – |
Check yourself