Stop Losses: Types and Placement
A stop is a decision made in advance about when a position is wrong. Where it sits should follow from the structure of the chart rather than from a preferred dollar amount.
MadStockAlerts Research · Updated August 28, 2026
What to take away
- A stop loss defines what would make the reason for the position no longer true.
- Placement should follow structure, and size should then follow placement.
- A stop-market fills in a price gap; a stop-limit may not fill at all.
- Obvious levels attract clustered stops, which is why marginal breaks happen.
- Widening a stop as price approaches it converts a defined loss into an undefined one.
- A thesis stop usually triggers while a position is profitable, which is why it is the one most often skipped.
MAD Academy Training Video · 0:45
A Stop Is a Statement, Not a Wish
A stop marks the price at which your reason for the trade is gone — which is a chart decision, not a tolerance for pain.
This lesson is part of a Stock Alerts + Tools plan.
What a stop is for
Every position rests on a reason. A stop loss is the price at which that reason no longer holds, decided while the decision is still unemotional, which is before the position exists.
The order of operations matters. Choose the level at which the idea is wrong, then size the position so that reaching it costs an acceptable amount. Choosing a stop distance to fit a position size already decided is the reverse, and it puts the exit somewhere the chart does not justify.
A useful test: if the stop level is reached, would the original reason for the position still be intact? If yes, the stop is in the wrong place. If no, it is a genuine invalidation point rather than a tolerance for loss.
Placement approaches
| Method | Basis | Weakness |
|---|---|---|
| Structural | Below a support level, a base or a swing low | Distance varies, so size must vary with it |
| Volatility | A multiple of ATR | Ignores where the chart's structure actually sits |
| Percentage | A fixed percentage from entry | Arbitrary; ignores both structure and volatility |
| Time | Exit if nothing happens within a period | Not a loss limit at all, but a separate discipline |
The first two are frequently combined: find the structural level, then check it is at least some multiple of ATR away, so that a stop is never placed inside the security's ordinary daily noise.
The clustering problem
Stops congregate just below obvious lows and round numbers, because that is where everyone reading the same chart places them. A marginal break through such a level triggers them all, and the resulting supply can carry price further than the break itself justified before it reverses.
Placing a stop slightly beyond the obvious level, and accepting the wider distance with a correspondingly smaller position, is the usual response. It is a trade-off rather than a solution: a wider stop is triggered less often and costs more when it is.
The arithmetic of that trade-off is worth doing rather than guessing. Widening a stop by half means halving the position for the same risk, and whether that is worth it depends on how often the marginal break actually happens in that security.
Scroll the chart sideways to see all of it.
What a stop cannot do
A stop is not a guarantee. A stock closing at $40 and opening at $28 after news fills a stop-market order near $28, not at the trigger. A stop-limit avoids the poor fill by not filling at all, leaving the position open. Overnight gap risk cannot be removed by any order type, only by position size.
A trading halt has the same effect. A stop cannot execute in a market that is not trading, and it will execute into whatever price the reopening auction produces.
Moving a stop
Moving a stop further away as price approaches it converts a defined loss into an undefined one and is the mechanism behind most account-ending losses. It is almost always accompanied by a reason, and the reason is almost always constructed after the price arrived.
Moving one closer as a position advances is a different act with a different rationale, and the distinction is worth keeping explicit. One reduces risk that has already been accepted; the other increases risk that was already decided against.
Stop orders, stop-limits, and mental stops
The three common forms fail in different ways, and the differences matter most exactly when the stop is reached, which is when the market is moving quickly.
| Form | What it guarantees | How it fails |
|---|---|---|
| Stop-market | Execution once triggered | The fill can be far below the trigger in a fast market or a gap |
| Stop-limit | A price floor once triggered | It can fail to fill at all, leaving the full position through the whole move |
| Mental stop | Nothing mechanical | It requires a decision at the moment decisions are hardest, and it is often not taken |
A stop-limit is often chosen after a bad fill on a stop-market, which trades a known cost for an unknown one. A stop-market on a gap fills badly; a stop-limit on the same gap does not fill, and the position is still open with the price below the limit and falling.
A resting stop order is visible to the market in aggregate. It is not visible as your order, but the cluster of orders it belongs to is inferable from the chart, which is the mechanism behind the observation that obvious levels get run before the move continues.
Time stops and thesis stops
A price stop answers one question: has the price gone somewhere that makes the idea wrong. Two other conditions can make an idea wrong without the price doing anything at all, and neither is covered by a level.
A time stop closes a position that has not done what it was supposed to do within the window in which it was supposed to do it. A setup expected to resolve in a week that has sat still for a month has not been proven wrong, and it has failed to be proven right, and the capital has been committed the whole time. The convention exists because a position that is going nowhere costs opportunity rather than money, which makes it invisible to a price stop.
A thesis stop closes a position when the reason for holding it stops being true, whatever the price is doing. If a position was opened because a company was expected to win a contract and it loses it, the reason is gone; the price may not have moved yet, and the position is now being held for no stated reason at all.
The thesis stop is the one most often skipped, because it usually triggers while the position is profitable. A position that is up on a thesis that has failed is a position held for a reason that has been replaced by the fact that it is up.