🎯 In 30 seconds: Momentum traders buy the stocks that have gone up the most compared with others, and avoid or short the ones that have done worst. Decades of research find that past winners over the last 3 to 12 months have tended to keep beating past losers for a few more months. The catch is that momentum occasionally crashes, hard and fast, usually right when a beaten-down market rebounds. Managing that crash risk is most of the job.
What momentum means
Momentum is the tendency of an asset that has been rising to keep rising, and one that has been falling to keep falling, for some period. There are two versions, and it helps to keep them apart.
- Relative (cross-sectional) momentum compares stocks with each other. You rank a group by past return and buy the top of the list. A stock can have strong relative momentum even if it fell, as long as it fell less than the others.
- Absolute (time-series) momentum compares an asset with its own past. Is it higher than 12 months ago? That is the basis of trend following.
This chapter is mostly about relative momentum, the version most associated with stock picking.
Measuring relative strength
The standard academic measure is the return over the past 12 months, skipping the most recent month. The Fama and French momentum factor, for example, ranks stocks on their return from 12 months ago to 2 months ago. The last month is skipped because very short-term returns tend to partly reverse, which is the opposite effect (see Mean reversion).
12-1 momentum = (price 1 month ago / price 12 months ago) minus 1
Worked example with six hypothetical stocks:
| Stock | Price 12 months ago | Price 1 month ago | 12-1 return | Rank |
|---|---|---|---|---|
| D | $12.00 | $18.60 | +55% | 1 |
| A | $40.00 | $58.00 | +45% | 2 |
| B | $25.00 | $30.00 | +20% | 3 |
| E | $60.00 | $63.00 | +5% | 4 |
| C | $80.00 | $76.00 | -5% | 5 |
| F | $33.00 | $26.40 | -20% | 6 |
If the S&P 500 returned 12% over the same window, D beat it by 43 percentage points and A by 33. A momentum trader would focus on D and A and ignore C and F, however "cheap" they look.
Traders also use a relative strength line: the stock's price divided by an index level, plotted over time. A rising line means the stock is beating the index, even on days both fall. This is not the same thing as the RSI indicator in Momentum indicators, despite the similar name.
Three practical momentum signals
1. Ranking a universe monthly. Pick a list (for example the stocks in a large-cap index, or a set of sector ETFs). Once a month, rank by 12-1 return, hold the top 10% to 20%, and replace names that drop out. This is the closest to what the research tests.
2. New 52-week highs. A stock at a new 52-week high has nobody who bought in the past year sitting on a loss and waiting to "get back to even". Many momentum traders treat a close at a 52-week high on strong volume as confirmation. The stock screener can filter for stocks near their 52-week high.
3. Earnings momentum. Some traders look for companies whose earnings beat expectations and whose analysts are raising estimates. The idea is that one strong report can change the outlook for several quarters, and prices may adjust over weeks rather than all at once. Treat this as a hypothesis to test on your own records. See Earnings season.
Why momentum might exist
Nobody fully agrees. The main explanations are behavioral:
- Underreaction. Investors adjust slowly to good or bad news, so prices drift in the direction of the news for months.
- Herding and overreaction later. Rising prices attract buyers, which pushes prices further. Over several years, part of that move tends to reverse.
- Risk. Some argue momentum is compensation for crash risk, described below.
Whatever the cause, the pattern is time-scale dependent: very short-term (about a month) returns tend to reverse, 3 to 12 month returns tend to continue, and multi-year returns tend to partly reverse again.
Momentum crashes
Momentum has a dangerous habit. It does well for long stretches, then gives back years of gains in a few months. Daniel and Moskowitz studied US stocks from 1927 to 2013 using a strategy that buys the top 10% of past 12-month winners and shorts the bottom 10% of losers.
- In July and August 1932, the past losers rose 232% while the past winners gained only 32%. Those were the strategy's two worst months.
- From March to May 2009, the past losers rose 163% while the past winners gained only 8%.
The pattern behind both: the market had fallen hard, volatility was high, and then the market snapped back. The most beaten-down stocks rebounded the most. The authors found these crashes are partly predictable: they cluster in "panic" states after market declines and when volatility is high. A version of the strategy that cut exposure in those conditions roughly doubled its risk-adjusted performance in their tests.
Managing momentum risk
- Diversify. Hold 5 to 20 names across sectors. Momentum lists often fill up with one hot industry, so cap each sector, for example at 30% of the momentum book.
- Size by volatility. High-momentum stocks are usually volatile. Use risk-based sizing with an ATR stop so each position risks the same dollar amount (Position sizing).
- Use exits. Sell when a stock falls out of the top of your ranking at the monthly review, or hits its stop, whichever comes first.
- Watch the regime. When the broad market is below its 200-day average and volatility is high, reduce the size of new momentum positions.
- Count costs and taxes. Monthly rebalancing creates turnover. Gains held a year or less are taxed as ordinary income (Taxes for traders).
Worked sizing example (hypothetical): $20,000 account, 0.75% risk per position = $150. Stock D trades at $19.00 with an ATR of $0.80. A 2 x ATR stop sits $1.60 below entry, at $17.40. Shares = $150 / $1.60 = 93. Position value = 93 x $19.00 = $1,767, about 9% of the account. Five such positions would put about 45% of the account in momentum stocks while risking 3.75% in total if every stop hit at once.
Try it
Pick 10 sector ETFs and add them to your watchlist. On the first trading day of the month, compute each one's 12-1 return and rank them. "Buy" the top three in the paper portfolio in equal dollar amounts. Repeat the ranking monthly for six months, swapping out any ETF that drops out of the top three. Compare your result with holding an S&P 500 fund over the same months, and note the worst month for each.
- What is the difference between relative and absolute momentum? Answer: Relative momentum ranks stocks against each other by past return. Absolute momentum compares an asset with its own past, the basis of trend following.
- A stock was $40.00 twelve months ago and $58.00 one month ago. What is its 12-1 momentum? Answer: +45% ($58.00 / $40.00 minus 1).
- Why does the 12-1 measure skip the most recent month? Answer: Very short-term returns tend to partly reverse, the opposite effect.
- In what conditions have momentum crashes happened? Answer: After a hard market fall with high volatility, when the market snaps back and the most beaten-down stocks rebound the most, as in 1932 and 2009.