Trend following is a simple idea that is hard to practise: buy what is already rising, keep it while it keeps rising, and sell quickly when it stops. It makes no forecast about where a stock should go. It reacts to where the stock is going, with rules fixed in advance. This course applies the idea to US stocks, long side only, on daily bars. This first lesson explains why the approach can work at all, what its results look like trade by trade, and why the hardest part is psychological rather than technical.
What trend following is
A trend-following system has three parts. An entry rule that fires only after price has already moved in the direction of the trade, such as a close above the highest high of the last 20 sessions. A stop that limits the loss when the move fails. An exit rule that stays out of the way while the trend continues and closes the trade once it clearly ends. There is usually no profit target: the market decides how large a winner becomes.
This is the opposite of buying a dip. A dip buyer expects prices to return to a norm; a trend follower expects that a stock already moving strongly is more likely than average to continue. Both can work, but they produce very different trade statistics and demand a different temperament, as the rest of this lesson shows.
Why trends persist: the evidence
Trend following is one of the best-documented effects in finance. A few landmarks:
- Stock momentum. Narasimhan Jegadeesh and Sheridan Titman (1993) showed that US stocks with the highest returns over the past 3 to 12 months tended to outperform the weakest stocks over the following 3 to 12 months.
- Time-series momentum. Tobias Moskowitz, Yao Hua Ooi and Lasse Pedersen (2012) found that across 58 liquid futures and forward markets, an asset's own return over the past year helped predict its return over the next month, with the effect fading and partly reversing beyond a year.
- A century of data. Brian Hurst, Ooi and Pedersen (2017) extended a diversified trend-following strategy back to 1880 and found positive returns in every decade, with some of its best results during long equity bear markets.
- The 52-week high. Thomas George and Chuan-Yang Hwang (2004) showed that stocks trading close to their 52-week high tended to keep outperforming, a result covered in lesson 3.
The usual explanations are behavioural and structural. Investors tend to underreact to news at first: they anchor on old prices and old estimates, and many sell winners too early to lock in a gain (the disposition effect), which slows the price adjustment. Large institutions build positions over weeks, and good news about a company often arrives in a sequence of earnings beats and estimate revisions rather than all at once. Later, attention and buying pressure can carry a move further than fundamentals justify.
The evidence has limits worth stating plainly. The strongest results come from portfolios diversified across many markets, not from a single stock. Momentum strategies can suffer sudden crashes when beaten-down stocks rebound sharply, as Kent Daniel and Tobias Moskowitz documented, for example in 2009. And research returns rarely include the costs, slippage and taxes a retail trader pays. Trend following has an edge; it is not a guarantee.
Positive skew: many small losses, a few large wins
To compare trades of different sizes, measure each result in R: the profit or loss divided by the initial risk (entry minus stop). A trade that risked $2 per share and made $6 is +3R; a trade stopped out as planned is −1R. The figure shows what 100 trend-following trades typically look like in R.
The shape is the point. Losses are capped by the stop and pile up near −1R. Winners are not capped, so they spread out in a long right tail. Statisticians call this positive skew. In this example the 35 winners average about +2.7R and the 65 losers about −0.9R, so the expectancy is 0.35 × 2.7 − 0.65 × 0.9 ≈ +0.36R per trade. With 1% of the account at risk on each trade, that is roughly 0.36% per trade on average, before costs.
Notice also that the five largest winners earn more than the entire net result. Remove them and the system loses money. This is normal for trend following, and it has a practical consequence: you cannot know in advance which signals will become those trades, so skipping signals by feel is the surest way to miss them.
Win rate versus payoff
Whether a system makes money depends on win rate and payoff ratio together. A system breaks even before costs when win rate × average win equals loss rate × average loss. Rearranged, the break-even payoff ratio is (1 − win rate) ÷ win rate: a 25% win rate needs winners three times the size of the losers, a 50% win rate needs them equal.
Trend-following systems usually sit in the upper left of this chart: win rates of roughly 30% to 45% with payoff ratios of 2:1 or more. That low win rate is not a flaw to be fixed; it is the price of letting winners run. Most breakouts fail or stall, and the stop takes a small loss. A system that wins more often by taking profits sooner usually pays for it with a smaller payoff, and can easily slip below the curve once costs are included. For the arithmetic of targets and break-even win rates, see risk-reward ratio.
The psychology of a low-win-rate system
A system that loses on most trades produces long losing streaks, even when it has a real edge. With a 40% win rate, the chance of at least six losses in a row somewhere in 100 trades is about 87%, and of eight in a row about 49%. The equity curve rarely rises smoothly. It tends to drift lower for months and then jump when a few trends arrive.
This is where most trend followers fail, not in the rules. Three urges are especially costly:
- Taking profits early. Selling a winner at +1R feels prudent, but it cuts off the right tail that pays for every loss.
- Skipping signals after a losing streak. The streak says nothing about the next signal, and the next one may be the trade of the year.
- Changing the rules during a drawdown. A drawdown that the backtest already showed is the system working as designed, not evidence that it is broken.
The remedy is preparation: know the expected win rate, streak length and drawdown before the first trade, and keep the risk per trade small enough that a normal streak is survivable.
How to apply it
Every lesson in this course ends with a rule set. For this lesson, the rules describe the framework that later lessons fill in.
Entry
Enter only on a defined, objective signal that confirms strength: a breakout above a prior high, a stock newly meeting a trend template, a breakout from a base. No signal, no trade.
Stop
Decide the stop before entering and size the position so that hitting it costs a fixed, small share of the account, typically 0.5% to 1%. That loss is 1R.
Target
For the core position, none. Let the exit rule decide when the trend is over. Any fixed target must be weighed against the large winners it removes.
Invalidation
The trade is wrong when the stop is hit or the exit rule fires, not when it feels wrong. Record every result in R so that the system, not a single trade, is judged.
In Clear Tape
The Trading systems page lists the preset systems this course draws on, grouped as trend, breakout, momentum and volume systems. Each preset shows its complete rule book, its source and its known weakness. The 20-Day Breakout, for example, states openly that often fewer than 40% of its trades win and that a few large trends pay for many small losses. Every member can read the rules, switch systems on and keep a watchlist of up to 50 stocks; the daily screening of that watchlist after each close and the backtests, which report win rate, average R, expectancy after costs and maximum drawdown, are available to Tools and Pro members.
Common mistakes
- Judging a trend-following system by its win rate rather than its expectancy after costs.
- Risking so much per trade that an ordinary losing streak becomes a deep drawdown.
- Expecting research results from diversified futures portfolios to transfer one for one to a handful of stocks.
- Adding discretionary profit-taking that quietly removes the right tail.
- Abandoning the system in the middle of a drawdown that was within its tested range.
Summary
Trend following buys strength and lets the market decide the size of each win. The edge is supported by decades of research and explained by underreaction, gradual positioning and later herding, but it is not guaranteed. Its trade results are positively skewed: many small losses near −1R and a few large winners that carry the account. Success therefore depends on expectancy, small risk per trade and the discipline to take every signal through long losing streaks. The next lesson defines what a trend is and which stocks qualify as leaders.
For educational purposes only. Not investment advice.