Ready to go further
Next: size the trade around the stop
The position size calculator works out how many units a stop distance allows for your risk.
What is included
- What it is, in plain words
- The formula or mechanism, with exact arithmetic
- How it hits grid, martingale and DCA bots
- A check you can run before you switch a bot on
Who it is for
- People who run or plan to run a trading bot
- Anyone reading a backtest or a bot sales page
- Readers who want the numbers, not the pitch
- Not advice, and not a signal service
Key facts
- A trailing stop sits a fixed distance below the highest price since you entered, and moves up as the price makes new highs. It never moves down. If the price falls to it, the trade closes.
- The distance is a trade-off. A tight trail exits on small pullbacks and keeps little of the move, and a wide trail stays in longer and gives back more of the move before it exits.
- On 2,000 made-up random paths, a trail of 1 point averaged +0.54, 2 averaged +1.07, 4 averaged +2.92 and 8 averaged +6.84, against +9.67 for holding to the end (calculated by us on 28 Oct 2026).
- A trailing stop is not a prediction. On a path with no drift it adds nothing, and on a real market the best distance changes with the volatility.
- Real fills can be worse than the stop price when the price gaps. See the guide on stop loss orders.
How the stop moves
Trailing stop = highest price - trail distance
Exit = the first time the price is at or below the trailing stop
Buy at 100 with a trail of 2. The stop starts at 98. The price rises to 103, so the stop rises to 101. If the price then falls to 102 the stop stays at 101, because it never moves down. If the price falls to 101 the trade closes there, for a gain of 1 on the entry. The distance can be a fixed number of points, a percent of the price or a multiple of the average true range, which adapts to volatility. See Supertrend for an ATR based version.
What 2,000 random paths show
To see the trade-off we generated 2,000 made-up price paths of 80 bars with a small drift of 0.12 a bar and noise of 1.3, and applied four trail distances to each. The table shows the average result in points, how often the trade made a profit, how often it closed within the first 10 bars, and the average bar of the exit.
| Trail (points) | Average result | Profitable | Out in first 10 bars | Average exit bar | Bad-luck result (5th percentile) |
|---|---|---|---|---|---|
| 1 | +0.54 | 44% | 90% | 4 | -2.60 |
| 2 | +1.07 | 47% | 68% | 9 | -3.27 |
| 4 | +2.92 | 51% | 25% | 24 | -5.00 |
| 8 | +6.84 | 63% | 2% | 56 | -8.32 |
Holding to the end averaged +9.67 on the same paths, because the drift is positive. Every trail averaged less than holding, and the wider the trail the closer it came: the best was 8 points at +6.84. The price of that is the bad-luck column. The worst 5% of paths ended at -9.59 or worse if held to the end, against -8.32 with the widest trail and -2.60 with the tightest. So in this simulation the stop did not add return, it narrowed the bad outcomes, and the tighter it was the more of the return it gave up for that protection. A very tight stop is hit by noise in the first few bars on 90% of paths and so keeps few of the gains, while a wide stop stays in longer and so behaves more like holding. These paths have a drift and no trends, no gaps and no costs, so they say what the arithmetic does, not what any market does.
One path, four trails
On the single path in the picture the exits were: a trail of 1 left at bar 10 for +5.60, a trail of 2 left at bar 19 for +6.50, a trail of 4 left at bar 22 for +5.24, a trail of 8 left at bar 33 for +3.73. The same path gives four different results, which is why a single backtest of a stop distance tells you little. Test across many paths and many market periods.
Where bots go wrong
- Tuning the distance to the past. The best trail in a backtest is usually the one that fit that history. See overfitting.
- Ignoring costs. A tight trail exits and re-enters often, and each exit pays the spread.
- Trusting the stop price. A gap can fill below the stop. See stop loss orders and slippage.
- A fixed distance in changing volatility. A trail that suits a calm market is too tight in a wild one.
- Test fills. A backtest that exits at exactly the stop price is too kind, as covered in backtesting.
How to check a trailing stop rule
Test several distances over several market periods, with full costs and realistic fills, and look for a range of distances that works rather than a single peak. Keep data back for one final test. Check the max drawdown, and size the position with the position size calculator so the worst plausible exit fits your risk.
What this page does not tell you
It does not say which distance to use, or that a trailing stop adds profit. The paths are random and made up, and this is education, not advice.
Keep Going
Related Tools and Guides
Keep exploring
FAQ
Frequently Asked Questions
What is a trailing stop?
A stop order that follows the highest price since entry by a fixed distance or percent. It moves up with the price, never down, and closes the trade if the price falls to it.
How does a trailing stop work?
The stop is the highest price minus the trail distance. When the price makes a new high the stop rises. When the price falls to the stop, the trade closes.
How far should a trailing stop be?
There is no single answer. A tight trail exits on noise and a wide trail gives back more profit, so test several distances across several market periods with full costs.
What is the difference between a trailing stop and a trailing stop limit?
A trailing stop sends a market order when hit and nearly always fills. A trailing stop limit sends a limit order, which controls the price but may not fill in a fast fall.
Does a trailing stop make a bot more profitable?
Not by itself. It changes the shape of the results, and in the simulation on this page the average depends on the drift and the distance. Test it with realistic fills.
Are the numbers on this page real?
Example numbers are made up to show how the maths works, and the page says so. A figure only counts as exact when the page says it was calculated on a stated date. Nothing you enter in a calculator leaves your browser.
Can I use this to decide whether to run a bot?
It can show whether a bot's numbers hold together and how much an account can lose. It cannot say what the market will do, and it is not advice. Use it to ask better questions, then read the terms of your own broker or exchange.
Is this financial advice?
No. It is general education about an order type. The paths are random and made up and it does not consider your situation.
Know the Risk Before You Run It
Free calculators and guides, added every week. Leave your details and we will email you when new ones go up.
Trading carries a risk of loss