Trailing Stop Orders Explained: Percent, ATR and the Trade-Off Nobody Mentions

A trailing stop is a stop loss that moves in your favour as price moves in your favour, and never moves back. On a long position it rises behind price as new highs are made; on a short position it falls behind price as new lows are made. The appeal is obvious: it lets a winning trade run without a fixed target while locking in part of the gain. The cost is less obvious, and it is the reason most trailing stops are set badly: every trailing stop gives back part of the move by design. The only question is how much, and in exchange for what.

How a trailing stop order works

Most exchanges and brokers offer a trailing stop as a built-in order. You set a distance, either in percent or in price, and the platform does the rest.

• On crypto exchanges this distance is often called the callback rate, for example 5%. Some also let you set an activation price: the trail only starts once price reaches that level.
• On forex platforms such as MT4 and MT5 the distance is set in points, and on MT4 the trailing stop runs in the terminal on your computer, so it stops moving when the terminal is closed. The stop already in place stays put, but it no longer follows price.

The rule is the same everywhere: the stop sits a fixed distance below the highest price reached since the order became active. When price falls back by that distance, the stop triggers as a market order.

A worked example

You buy at 60,000 with a 5% trailing stop.

| Price action | Highest price so far | Stop level |

|---|---|---|

| Entry | 60,000 | 57,000 |

| Rises to 64,000 | 64,000 | 60,800 |

| Dips to 62,000 | 64,000 | 60,800 (unchanged) |

| Rises to 70,000 | 70,000 | 66,500 |

| Falls to 66,500 | 70,000 | Triggered |

The trade made 10,000 at its peak and closed with 6,500. The 3,500 given back is not a failure of the trailing stop. It is the price of not knowing in advance where the top would be.

Why a fixed percentage trail is usually the wrong distance

A percentage trail treats every market and every period the same. A 5% trail on Bitcoin during a calm month and during a volatile month are two different orders: in the first, normal fluctuation rarely reaches 5%; in the second, it can reach it several times a day. A trail that is too tight for the current volatility does not protect profit. It turns ordinary noise into exits, and you watch the trade continue without you.

The more robust approach is to set the distance as a multiple of current volatility, usually the [average true range](/learn/technical/atr-and-volatility). The classic version is the chandelier exit:

stop = highest high since entry − k × ATR

With k around 3 on a daily chart, the stop sits outside normal daily swings and only triggers when the move against you is large compared with what the market has recently been doing. When volatility rises, the distance widens; when it falls, it tightens. That is what a percentage trail cannot do.

Example: Bitcoin's 14-day ATR is 1,800 and the highest high since your entry is 70,000. With k = 3 the stop is 70,000 − 5,400 = 64,600. If the ATR rises to 2,500 during a volatile week, the same rule gives 62,500, and the stop does not move up until price makes new highs again.

Trail on the close, or intraday?

A trailing stop order on an exchange reacts to every tick, so a single wick can trigger it. An alternative is to trail on the close: update the stop level once per candle, and exit only if a candle closes below it, with a separate hard stop further away to catch a crash. This avoids being stopped by short spikes, at the cost of exiting later on a real reversal. Neither is free. Choose one, write it down, and stick to it.

What we found when we tested exits

We tested ten different exit rules on trend trades in large-cap crypto, including tighter trails, trails that tighten after a profit target, momentum-based exits and Fibonacci-based exits, all against a plain volatility-based trailing stop and with trading costs included. None of the alternatives beat the plain version over the full set of test periods. The tighter rules looked better on the trades where they caught the top and worse on the trades where they cut a large winner short, and the second group cost more than the first group saved.

That matches the arithmetic. In trend following, a few large winners pay for many small losses. An exit rule that trims the large winners has to save a lot elsewhere to be worth it, and in our tests none did.

What a trailing stop cannot do

• It cannot guarantee the locked-in profit. Once triggered it is a market order. A gap or a crash can fill it well below the stop level.
• It does not choose the trade. A trailing stop manages a position; it does not make a bad entry good.
• It does not remove the urge to interfere. The hardest moment is when price is just above the stop and you are tempted to move it lower "to give it room". Moving a trailing stop backwards turns it into an ordinary stop with a worse level. If you need more room, the distance was wrong from the start, and the [where to put a stop](/learn/risk/where-to-put-a-stop) lesson is the place to fix it.

Frequently asked questions

What is a good trailing stop percentage? There is no good fixed percentage, because the right distance depends on how volatile the market is right now. A multiple of the ATR adapts automatically. If you must use a percentage, check it against recent daily ranges: a trail smaller than a normal day's move will be hit by normal days.

Is a trailing stop better than a take profit? They answer different questions. A take profit decides in advance where you get out. A trailing stop lets the market decide. In trending markets trailing usually captures more of the large moves; in ranging markets a fixed target usually does better.

What is the callback rate on a crypto exchange? It is the trailing distance in percent: how far price must fall from its highest point since activation before the order triggers.

Should I use a trailing stop from the moment I enter? Many traders start with a fixed initial stop and switch to trailing only once the trade has moved in their favour by a set amount, for example one times the initial risk. That avoids the trail tightening the stop before the trade has had room to work.

Indikora is an AI-powered trading coach for crypto, forex, gold and indices. It records the exit plan you set before each trade, and its journal shows how much of each move you kept and how often you changed a stop after entry. Try it free: https://indikora.com

This article is for educational purposes only and is not financial advice.