Trend following buys markets that are already rising, sells markets that are already falling, and stays in until the trend ends. It makes no forecast of how far a move will go. Most trades end as small losses and a few large winners pay for them, so the exit and the risk per trade decide the result more than the entry does.
What you will learn
- Two ways to enter a trend: a channel breakout and a pullback
- How to trail a stop with an ATR multiple, a channel or swing lows
- Why the win rate is low, and how the expectancy can still be positive
- How long losing runs fit inside the daily limit and the floor
Enter on a breakout or a pullback
A channel breakout uses a Donchian channel, which marks the high and the low of the last N bars. A common version buys when the daily close is above the high of the last 20 days and sells when it is below the low of the last 20 days. It gets you into every sustained trend at some point, and into many moves that fade after a few days.
A pullback entry waits for a trend to prove itself first. Price is above a rising 50-day moving average and making higher lows. You wait for a dip towards the average or towards an old breakout level, then buy when a daily candle closes back up. You get fewer false starts, but you miss the trends that never pull back. The indicators lesson explains moving averages if you need a refresher.
Both versions place the first stop with the average true range (ATR), which is the average size of a bar's full range, gaps included, usually over 14 bars. A stop placed a multiple of ATR away scales itself to how much the market is moving.
Take a breakout on AUD/USD. The 20-day high is 0.6600 and the daily candle closes at 0.6610, so you buy at 0.6610. The 14-day ATR is 50 pips. With the stop 2.5 ATR away, it goes 125 pips below, at 0.6485. On a $25,000 Direct account at 0.5% risk, 1R is $125, which is $1 a pip, or 0.10 lots.
Trail the stop so winners can run
A trend follower usually sets no profit target. The trailing stop is the exit, and it only ever moves in the direction of the trade. Three common ways to trail it:
- Keep it a fixed number of ATRs below the highest close since entry.
- Move it to the low of the last 10 days, the lower line of a shorter Donchian channel.
- Raise it under each new higher low once price has bounced from it.
Follow the AUD/USD trade with the first method, assuming the ATR stays at 50 pips. Weeks later the highest close is 0.7135, so the stop sits at 0.7010. Price turns and hits it. The trade made 400 pips, $400, which is 3.2R. At the peak it was 525 pips up, or 4.2R, so the trail gave back exactly 1R. A tighter trail gives back less on each trade but stops you out of more trends during normal pullbacks. You can move the stop by hand after each daily close, or use a trailing-stop order if the platform offers one.
Why most trades lose and the method can still work
Many breakouts from a 20-day channel stall and fall back, so a trend follower usually loses more often than it wins. Suppose your journal shows 60 trades: 21 winners averaging 2.8R and 39 losers averaging 1R. That is a 35% win rate. Expectancy is (0.35 × 2.8) minus (0.65 × 1), which is 0.98 minus 0.65, or +0.33R a trade. At $125 per R, that averages about $41.
The break-even win rate at 2.8R is 1 ÷ 3.8, about 26%, as the expectancy lesson showed. The method depends on the size of its winners. Close every winner at 1.5R to make it feel safer and, at the same 35% win rate, expectancy becomes 0.525 minus 0.65, or -0.125R. Even if the earlier exit lifted the win rate to 40%, you would get 0.60 minus 0.60, which is zero.
Size for long losing runs
A low win rate brings long strings of losses. If each trade wins 35% of the time, independently of the others, the chance that any eight given trades all lose is 0.65 to the power of 8, about 3.2%. Over 100 trades, the chance of at least one run of eight or more losses is about 69%, and of ten or more about 37%.
On a $25,000 Direct account the floor is $23,500, which is $1,500 below the start. At 0.5% risk that is 12R of room, and a run of ten losses costs $1,250 and leaves $250. At 1% risk the floor is only 6R away, and the chance of a run of six or more losses in 100 trades is about 96%. The lower your win rate, the more R you need between your equity and the floor.
The daily limit matters when several trades share a theme. Long breakouts in AUD/USD, NZD/USD and EUR/USD can all fail on the same strong-dollar day. At $125 each that is $375, more than a third of the $1,000 daily limit. The Direct target suits a slow method: 10%, or $2,500, is 20R, about 61 trades at +0.33R, and there is no time limit.
Check your understanding
You buy at 1.2500. The ATR is 60 pips and you trail the stop 2.5 ATR below the highest close. The highest close is 1.2980. Where is the stop, and what is the trade worth in R if it is hit?
2.5 × 60 is 150 pips, so the stop is at 1.2830. That is 330 pips above the entry. The first stop was 150 pips away, so the trade makes 2.2R.
A method wins 30% of its trades, with an average win of 3R and an average loss of 1R. What is its expectancy, and what happens if the winners are cut to 2R?
0.9 minus 0.7 gives +0.2R a trade. With 2R winners, 0.6 minus 0.7 gives -0.1R, so the method loses over time.
Why does a trend follower need more room above the floor than a strategy that wins 60% of the time?
With a 35% win rate, runs of eight or ten losses are likely over 100 trades. The risk per trade has to be small enough for such a run to fit inside the floor.
Key points
- Enter on a close beyond a 20-day channel or on a pullback inside an established trend, with the first stop set by ATR.
- The trailing stop is the exit. It only moves with the trade and always gives back part of the move.
- Expect a win rate well under 50%. Large winners carry the expectancy, so do not cut them short.
- At a 35% win rate, a run of eight or more losses is likely over 100 trades, so size for it.
- Count trades on a shared theme, such as several US dollar pairs, as one risk against the daily limit.
Next lesson: Trade ranges and stretched prices back to the average
All trading is simulated. Rewards are based on performance and are not guaranteed.