Every stop-loss answers one of two questions. Either it asks where the market would have to go for the reason you entered to be wrong, or it asks how much you are willing to lose. These produce different levels, and the difference between traders who are stopped out constantly and traders who are not is usually which question they answered.
The four common approaches
- Structural: beyond the level whose breach invalidates the trade.
- Volatility-based: a multiple of the instrument's recent average range.
- Time-based: exit if the move has not happened within a defined period.
- Account-based: a fixed currency amount, which is where the trouble starts.
Structural stops
A structural stop sits beyond the price level whose breach would mean the trade thesis is wrong — under the swing low for a long, above the range high for a short, beyond the level being tested. It is placed by the chart rather than by the account, and it is the only method that gives the stop an actual meaning.
Its strength is that being stopped out is informative: the market did the thing that would make the trade wrong. Its cost is that the distance is whatever the structure requires, which on a wide setup may be more than the account can accommodate at a normal position size.
That constraint is the point rather than a flaw. When the structural stop is too far away for a sensible position size, the correct response is a smaller position — or no trade — rather than a closer stop.
Volatility-based stops
A volatility stop sets the distance as a multiple of the instrument's recent average true range: two ranges, three ranges, whatever testing supports. It answers the question of how far the instrument routinely moves without meaning anything, and places the stop beyond that.
This is the correct method for anyone trading several instruments, because a fixed pip distance across a quiet major and a volatile cross is implicitly taking several times more risk on the second. Sizing then follows from the stop, so position size becomes inversely proportional to volatility automatically.
Its weakness is that it ignores structure entirely. A volatility stop can sit just inside an obvious level that the market is likely to reach, which is the worst of both approaches. Many traders use both: the volatility measure sets the minimum distance, and structure sets where within that the stop actually goes.
Time-based exits
A time stop closes the position if the expected move has not begun within a defined period. It does not replace a price stop; it sits alongside one and addresses a different failure, which is a trade that is not wrong but is not working.
This suits methods whose edge is time-bounded — a reaction to an event, a session-based pattern, a continuation that should resolve quickly. It suits nothing whose thesis is measured in weeks. The main benefit is capital release: money committed to a position that has done nothing for three days is money not available for a setup that is working.
A stop-loss is a market order when triggered
Under every execution model, a triggered stop becomes a market order and fills at the best available price. In a gap that can be far beyond the level. Only a guaranteed stop, offered separately and usually for a fee, fixes the exit price — and it is one of the few things worth paying for on an instrument prone to gapping.
Account-based stops, and why they fail
The fourth approach places the stop at a fixed currency loss — twenty units, fifty units — regardless of what the chart says. The stop distance is then whatever that amount buys at the position size already chosen.
This inverts the correct order of operations. Position size should follow from the stop distance and the risk limit; here the stop follows from the position size. The result is a stop placed where the market has no reason to respect it, and it is hit by ordinary noise at a rate that has nothing to do with whether the trade idea was sound.
The version of this that does the most damage is tightening a stop specifically to permit a larger position. The arithmetic works — a closer stop does allow more lots at the same risk — and the trading does not, because the closer stop is inside the range the instrument moves in without meaning anything.
The correct sequence
Placing a stop and sizing from it
- Identify the level whose breach invalidates the trade.
- Check that level against the instrument's average true range — if it is inside one range, it is too close.
- Place the stop beyond both, with a small buffer for spread widening at the exit.
- Measure the distance in ticks and multiply by the tick value to get risk per lot.
- Divide your risk limit by that number to get position size.
- If the resulting size is below the minimum the broker permits, do not take the trade.
The last step is the one people skip, and it is the discipline the whole sequence exists to enforce. A setup whose structural stop is too wide for your account is a setup for a larger account, not an invitation to move the stop.
Moving a stop
Moving a stop further away is almost never justified: it converts a defined loss into an undefined one at the moment when the thesis is already under question. Moving it closer as the trade works is a separate decision with its own trade-off — it protects gains and it also removes the room the position needs to breathe, which is why trailing stops so often exit just before the move continues.
Whichever you do, decide the rule before the trade rather than during it. A stop moved in reaction to watching the price is being moved by the same impulse the stop exists to contain.
Common questions
How far should a stop be from entry?
Far enough to sit beyond both the level that invalidates the trade and the instrument's normal noise, measured by average true range. The distance is an output of those two things, not a number chosen in advance.
Should I use a mental stop instead?
No. A mental stop requires you to act correctly at the moment you are least able to, and it fails silently when you are away from the screen. A resting order does not.
Do brokers hunt stops?
Prices reaching obvious levels and reversing is a market phenomenon, not usually a broker one — those levels are where resting orders cluster. Placing stops beyond obvious round numbers and swing points addresses most of it.
Is a guaranteed stop worth paying for?
On instruments prone to gapping, and on positions held over weekends or events, often yes. It is the only instrument that fixes the exit price, since an ordinary stop becomes a market order when triggered.
