Stop-Loss Orders: What They Do, Where They Fail

A stop-loss order rests inactive until price trades at or through your stop price, then converts to a market order. You get filled — almost always — but at an unknown price. A stop-limit converts to a limit order instead: you control the price but accept that in a fast move you may not get filled at all, and an unfilled stop is no stop.

That is the entire trade-off. A plain stop protects you from being trapped and exposes you to slippage. A stop-limit protects you from slippage and exposes you to being trapped. Neither is universally correct; the choice depends on whether the greater risk in this name is a bad fill or no fill.

A trailing stop moves your level up as price advances by a fixed amount or percentage, and never moves back down. It is a profit-management tool, not a risk-definition tool — useful once a trade is working, useless for deciding where a thesis is wrong.

Gaps are the big one. A stop at $48 on a stock that closes at $50 and opens at $39 fills near $39. Overnight risk is not something a stop can cover, which is why position size, not stop placement, is the real defense against gap risk.

Halts behave the same way. When a name halts on news and reopens materially lower, your stop triggers into the reopening auction and takes the print. Low-liquidity names produce the same effect without any news at all — a thin book means your market order walks several levels down.

Then there is the self-inflicted failure: clustering your stop at the obvious round number or just under the visible swing low, where every other stop sits. Liquidity gets sought where stops are dense. That is not a conspiracy, it is simply where the resting orders are.

Set the stop where your idea is wrong, then size the position so that distance equals your intended dollar risk. Doing it in the other order — picking the dollar loss first and putting the stop wherever that lands — guarantees you get stopped out of correct trades by ordinary noise.

Practical anchors: below the structural low that defined the setup, beyond a volatility measure such as a multiple of average true range, or beyond the session level (opening range, VWAP, prior day's high or low) that your thesis depends on. Give it enough room to be a level rather than a tick.

Then respect it. Moving a stop further away is not risk management, it is converting a defined loss into an undefined one. If you find yourself wanting more room, the honest read is that the position was too large, and the fix is to reduce size rather than widen the level.

Resting stops are visible in the aggregate and can be run. Mental stops are invisible and require you to actually execute under stress, which most people do not reliably do. Experienced discretionary traders often use mental levels on liquid names and resting stops on anything volatile or thin — and hard stops always when they step away from the screen.

Bracket orders attach a stop and a target to the entry so risk is defined the moment you are filled, which removes the window where a distracted trader is naked in a position. On most active-trading platforms this is a one-click template, and it is worth building.

The discipline that matters is not which mechanism you choose. It is that the level exists before the entry, the size derives from it, and you do not renegotiate with it afterward.