How to place a stop-loss orderA stop is a trigger, not a promise. It says where you want to start getting out - the price you actually get is a separate question, and the difference between the two is where most of the surprises live.How trailing stops work
A stop that follows the market up and never comes back down. It is the only exit that can capture a move nobody predicted - and it guarantees you will give some of that move back.
A fixed stop sits where you put it. A trailing stop moves with the market when the market goes your way, and stays put when it goes against you - a ratchet that tightens and never loosens.
For a long position: new high, stop rises. Price falls, stop holds. That is the whole mechanism, and everything difficult about trailing stops is a consequence of it.
The ratchet, step by step
A share bought at $100 with a $5 trailing distance. Watch the stop follow, then stop following.
| Highest price reached | Where the stop sits |
|---|---|
| $100: entry | $95 |
| $103 | $98 |
| $105 | $100: now at break even |
| $110 | $105: now protecting a gain |
| $120 | $115 |
| Falls back to $118 | $115: does not follow down |
| Falls to $115 | Triggered |
Exit around $115 on a $100 entry: roughly $15 a share, from a move that peaked at $20. The $5 given back is the fee for not having to guess where the top was.
Against a fixed target
These solve different problems, and choosing between them is really a choice about where your strategy's profit comes from.
Fixed take profit
You decide the exit in advance and the trade closes there. Certainty, and a known reward on every winner.
Nothing can produce more than the target, a trade that would have run to +8R closes at +2R like all the others.
Suits strategies with a decent hit rate and a defined move in view.
Trailing stop
No ceiling. The position stays open while the market keeps going, and the exit is decided by the reversal rather than by you.
Every winner surrenders the trailing distance from its peak, and some trades that were briefly profitable close at a loss.
Suits strategies that need the occasional very large winner to pay for a lot of small losses.
They combine: close part of the position at a fixed target and trail the rest. That banks something and keeps a claim on a larger move.
Choosing the distance
This is the only real decision, and there is no correct answer - only a trade-off with two bad ends.
Too tight and ordinary movement closes the trade. If a pair routinely swings 40 pips inside a session, a 15-pip trail will be triggered constantly by noise, generating exits, costs and no participation in the move you were waiting for.
Too wide and the position gives back most of what it made. A trade that reached $150 from a $100 entry with a $30 trail closes near $120, keeping a fifth of what was briefly on the table.
Three ways to set it
A fixed distance - $5, 50 pips, 100 index points - is simple and blind. As the price rises the trail becomes proportionally tighter: $10 behind a $100 share is 10%, and behind a $200 share it is 5%, so a large trend quietly squeezes the stop.
A percentage keeps the proportion constant. Ten per cent behind the high is $90 at $100 and $180 at $200, which suits instruments that move in large multiples.
Average True Range ties the distance to how much the instrument is actually moving now, widening in volatile conditions and tightening in quiet ones. The Chandelier Exit is the best-known version: the highest high since entry, minus a multiple of ATR.
Trailing structure instead of distance
The stop does not have to follow by a fixed number. Many trend traders move it beneath each successive higher low instead, so the exit follows the shape of the trend rather than a measurement of it.
Entry at $100 with the first low at $98; the market reaches $110 and forms a low at $105, then $120 with a low at $114. The stop moves $98, $105, $114, and the trade ends when the trend stops making higher lows - which is a definition of the trend ending, rather than a proxy for it.
It does not have to start immediately
Trailing from the moment of entry gives a trade no room to breathe through its first pullback. Many strategies leave the original stop in place until the position has reached some milestone - a fixed profit, +1R, +2R, a technical break - and only then begin the ratchet.
That single parameter changes the behaviour completely. Trailing from entry produces many small wins and few large ones; delaying it produces more full losses and keeps the large winners intact. They are different strategies wearing the same name, and they should be tested as such.
Two things a trailing stop does not do
It does not guarantee a profit. Until the ratchet has carried the stop past your entry the trade can still close at a loss, and a position that was briefly up $3 a share with a $10 trail is nowhere near safe.
It does not guarantee its price. A trailing stop is still a stop: when triggered it becomes an instruction to close at the next available price. If the market gaps from $125 to $110, a stop sitting at $120 executes around $110, and the protection it appeared to have built is not what you receive. That mechanism is covered in slippage.
Where the trailing logic actually runs
This is the practical detail that catches people out. A server-side trailing stop is managed by the broker and keeps working whether or not your computer is on. A platform-side one is calculated by the software in front of you, and if that closes, disconnects or loses power, the trailing stops updating.
The stop level already sent to the broker usually remains in place - it simply stops moving. So you can return to find a position closed at a level from hours earlier, having stopped following the market at the moment your laptop went to sleep. This is the classic issue with terminal-managed trailing on desktop MetaTrader, and it is why traders running trailing strategies often use a VPS.
Eighty of the hundred brokers we rate record trailing stops among their order types, which means twenty do not. Whether the trail runs on the server or on your machine is a separate question again, and worth asking before relying on it.
Before relying on one
- How much does this instrument normally move in the timeframe I trade?
- What trailing distance sits outside that ordinary movement?
- Fixed distance, percentage, ATR or market structure?
- Does trailing start at entry, or after a milestone?
- Can my rule ever move the stop backwards? It should not.
- Is the trailing managed by my broker or by my platform?
- Does my platform need to stay connected for it to keep working?
- Am I combining it with a partial fixed exit?
Questions people ask about trailing stops
Does a trailing stop guarantee a profit?
No. Until it has moved above your entry it is still a loss-limiting stop, and even above entry the fill can differ from the level after a gap.
Can a trailing stop turn a winner into a loser?
Yes. A position briefly in profit whose trail has not yet reached the entry price can still close below it.
Can it move down?
For a long position, no. That is the point of the ratchet. For a short it moves down with the market and never back up.
How many pips should a trailing stop be?
There is no universal figure. It has to sit outside the instrument's ordinary movement on your timeframe, which is why volatility-based distances adapt better than fixed ones.
Is a tighter trailing stop safer?
It gives back less from the peak and gets triggered by noise far more often. Neither end of that trade-off is safe in itself.
Does it work when my computer is off?
Only if the broker manages it server-side. Platform-side trailing stops updating when the software is closed or disconnected, though the last stop level normally remains with the broker.
Can I use a take profit and a trailing stop together?
Yes, and it is a common arrangement: close part of the position at a fixed target and trail the remainder.
Why did my trailing stop exit below the highest price?
Because that is how it works. The stop only triggers after the market reverses by the trailing distance, so the exit is always below the peak by at least that much.
Ready to put this to work?
Our questionnaire ranks all 100 brokers against your own answers in about a minute - including, if it matters to you, filtering out the ones whose leverage and protections do not suit how you intend to trade.
Find my brokerBrowse all 100 reviews
Nothing here is financial advice. Leveraged products can lose more than they make.
Other relevant articles
How to place a stop-loss orderA stop is a trigger, not a promise. It says where you want to start getting out - the price you actually get is a separate question, and the difference between the two is where most of the surprises live.
How to use a take-profit orderSetting a target is easy. Leaving it alone is not, and a trader who habitually closes winners early has quietly replaced their strategy with a worse one.
Risk-to-reward, win rate and expectancyA 1:3 ratio is not better than 1:1. It is a different bet, requiring a different hit rate, and the ratio on its own cannot tell you whether a strategy makes money.