A stop loss is not a device for limiting loss. It is the price at which your reason for being in the trade stops being true. Traders who treat it as the former place stops at whatever distance feels affordable, and then wonder why the market keeps hitting them before moving in the expected direction.
What a stop is actually for
Every trade rests on an idea: this level should hold, this trend should continue, this break should follow through.
Each idea has a price at which it is no longer valid. If support at 1.0820 should hold and price closes at 1.0790, support did not hold. The reason for the trade is gone.
That price is where the stop belongs.
Placing it anywhere else means one of two things: either you are exiting while the idea is still valid, or you are staying in after it has failed. Both cost money for different reasons.
Stop placement is not a risk decision. It is an analysis decision — you are marking where your read of the market is proven wrong. The risk decision comes afterwards, in the position size. Traders who conflate the two end up with stops placed by affordability, which produces a level that means nothing to the market.
Structure-based placement
The default approach, and the one that follows from the principle above.
For a long at support: below the swing low that defines the level.
For a short at resistance: above the swing high.
For a trend continuation: beyond the last higher low or lower high.
For a breakout: back inside the range, beyond where the break would be shown false.
In every case the stop sits beyond a level that, if broken, changes the reading of the chart.
Adding a buffer
Placing the stop exactly at the structural level is usually too tight, because price frequently overshoots slightly before reversing.
The buffer should scale with volatility rather than being a fixed pip count.
Swing low: 1.0820
ATR(14) on 1H: 12 pips
Buffer ≈ 0.5 × ATR = 6 pips
Stop at 1.0814
On gold the same logic applies with larger numbers, since the overshoot is proportionally larger — the Gold Trading section covers this specifically.
Volatility-based placement
An alternative for situations where structure is unclear, or as a sanity check on structural placement.
Set the stop at a multiple of ATR from entry — commonly 1.5 to 2 times.
This has one clear advantage: it adapts automatically as conditions change. The same rule produces a tight stop in a quiet market and a wide one in a volatile market, without you having to decide.
The disadvantage is that it ignores structure. An ATR stop can sit in the middle of a support zone, where price is likely to react, which is a poor place to be removed from the trade.
The practical combination: place the stop from structure, then check it against ATR. If it is closer than roughly 1.5 ATR, it is probably too tight for the current environment and the setup may not be worth taking at that location.
Stop placement mistakes
Sizing the stop to fit the position
The inversion described in the position sizing lesson. A trader wants 0.50 lots, calculates that a 20-pip stop keeps risk acceptable at that size, and places the stop 20 pips away regardless of what the chart says.
The stop now marks nothing. It will be hit by ordinary movement.
Round numbers and obvious levels
Stops cluster just beyond round numbers and obvious swing points. Price frequently moves into those areas because that is where liquidity is.
Placing the stop a little further out, beyond the cluster, costs a slightly larger stop distance and a slightly smaller position — and avoids a common way of being removed from a correct trade.
Mental stops
Deciding to exit at a level without placing the order.
This fails at exactly the moment it matters. When price reaches the level, the trader watches, waits for a bounce, and by the time they act the loss is larger than planned.
The only defensible use of a mental stop is when you are actively watching and have a demonstrated history of executing them. Most traders who believe they have that history do not.
Moving the stop wider
The most damaging single action in trading.
The position was sized for the original stop. Widening it converts a planned 1% loss into an unplanned 2% or 3% loss, made at the worst possible moment — while the trade is already losing and judgement is compromised.
It also usually does not help. Price that broke the structural level broke it.
A stop can be moved toward profit. It is never moved further away. This is worth treating as absolute rather than as a strong preference, because the situations that seem to justify an exception are precisely the situations where judgement is least reliable. Every trader who has blown an account can identify the moment they widened a stop.
No stop at all
Usually arrives after a run of being stopped out just before price reversed.
The reasoning feels sound — the stops were the problem, so remove them. What actually happens is that the frequent small losses are replaced by rare enormous ones, and the account lasts until the first trade that does not come back.
The correct response to repeated tight stop-outs is wider stops with smaller positions, not no stops.
Guaranteed stops and slippage
A stop is an instruction to close at the best available price once your level trades, not a guarantee of that price.
During gaps and high-impact news, the fill can be considerably worse. A 1% planned loss can become 2% or more.
Three implications:
Position size matters more than stop placement during volatile periods, because slippage scales with size.
Weekend gaps bypass stops entirely. Reducing or closing before the weekend is the practical defence.
Some brokers offer guaranteed stops for a fee. Worth understanding the cost before relying on them.
Trailing stops
A stop that moves toward profit as price advances, locking in gains.
Useful in genuine trends. Problematic in choppy conditions, where a trailing stop gets hit during ordinary retracement and removes you from a trade that then continues.
Common methods: trailing behind each new swing point as structure develops, or at a fixed ATR multiple from the current price.
Worth noting our own signals do not use trailing stops — one entry, one stop, one target, full close at the target. The reason is that a fixed structure makes expectancy measurable. Trailing introduces a variable exit that makes it considerably harder to evaluate honestly whether an approach works, and that clarity matters more than the occasional extra pip.
A placement checklist
1. Where does the idea fail?
2. Place the stop beyond that level
3. Add a buffer scaled to ATR
4. Check the distance is ≥ 1.5 ATR
5. Confirm it is not at an obvious cluster
6. Calculate position size from it
7. Place the order with the trade
8. Do not touch it except toward profit
What to do next
The next lesson covers risk-reward ratios and R-multiples — how to measure trades in a way that makes results comparable across instruments and position sizes.
Frequently asked questions
Where should I place my stop loss?
Beyond the price level at which your reason for the trade stops being true — below the swing low for a long at support, above the swing high for a short at resistance — with a buffer scaled to current volatility.
How far should a stop loss be in pips?
There is no correct pip figure. The distance comes from the chart, not from what feels affordable. Once the structural level determines the stop, position size is adjusted so the money risk stays at your fixed percentage.
Is it ever okay to move a stop loss wider?
No. The position was sized for the original stop, so widening converts a planned loss into a larger unplanned one, decided while the trade is already losing. Stops move toward profit only.
Should I use mental stops instead?
Generally no. Mental stops fail at the moment they matter — price reaches the level, the trader waits for a bounce, and the eventual loss exceeds the plan. Only defensible with a demonstrated history of executing them.
My stops keep getting hit before price reverses. What should I do?
Widen the stops and reduce position size correspondingly — same money risk, more room. Removing stops entirely replaces frequent small losses with rare enormous ones and ends accounts.
Do stop losses always fill at my price?
No. A stop closes at the best available price once your level trades. During gaps and high-impact news the fill can be considerably worse, which is why position size matters more than stop placement during volatile periods.