Why Price Spikes Through Your Stop, Then Reverses
You bought well, put the stop where the book said, and got taken out within a nickel of the low. The explanation isn't that somebody came for you. It's less flattering than that: you were predictable.
By Pavel Penev, MScFounder, TradeWize · 10+ years trading the marketsThe short answer
Your stop was hit because it was in the obvious place, and so was everyone else's. Stops pile up just beyond the prices everybody can see — under the swing low, past the round number, outside the range — and a stop order turns into a market order the instant it's touched. A pile of stops is therefore a pile of market orders waiting to fire at one price, which is why price so often reaches exactly that far and then turns around. Nobody had to be hunting you. And the fix is not a wider stop, and definitely not no stop: it's a smaller position, which is what buys you the room to put the stop somewhere less crowded.
You did it properly. Price pulled back, held, and you bought 200 shares at $52. You put the stop at $49.40 — just under the swing low, where every book and every video says to put it. Two days later price slid to $49.20, took you out, closed the day back at $51.80, and then spent the next fortnight climbing to $58 without you.
Being stopped out of a trade that turns out to be right is one of the most reliably infuriating experiences in this business, and the internet has a ready explanation waiting for you: they came for your stop. Your broker saw it. The market makers went and got it.
That explanation is wrong in its details and right in its instincts, which is the worst possible combination — because believing it sends you to the wrong fix.
It wasn't personal. That's the bad news
Start with the comforting half. Your stop was not visible to anybody. A stop order isn't a live order sitting in the public book where people can look at it — it's a dormant instruction that does nothing at all until price touches the trigger, at which point, in the SEC's own words, it becomes a market order. Until that moment there is nothing to see. Nobody picked your $49.40 out of a crowd, because your $49.40 was never on display.
Now the uncomfortable half. It didn't need to be visible. You put your stop just under the swing low for precisely the same reason everyone else did — because that is where the chart says the idea stops working. Thousands of people looking at one chart and applying one sensible rule will independently arrive at roughly one price. Your order was invisible and entirely predictable at the same time, and predictable is the half that costs money.
The mechanism underneath this is worth knowing once and then not worrying about again. Research on a large currency dealer's order book found that stop-losses really do bunch up just beyond round numbers, and that price moves unusually fast once it reaches one of those piles — because every stop that fires becomes a market order shoving in the same direction. Our guide to break of structure walks through that research and what a sweep does to a chart's levels. For today, one sentence covers it: the pile is real, it sits where you'd guess, and reaching it accelerates price.
Your stop was invisible and completely predictable at the same time. Only one of those two things was in your favour.
The three places everyone puts a stop
Ask a hundred traders where the stop goes on a long position and you will get three answers, not a hundred.
- Just under the last swing low — the price that would prove the pullback wasn't a pullback.
- Just past the round number — under $50, under $100, under whatever the price is nearest to. Humans like round numbers and it shows up in the order book.
- Just outside the range — below the floor of the sideways patch price has been stuck in.
Now look at those three again, because here is the part that matters: on any given chart they are frequently the same place. The swing low forms near a round number because the round number is part of why buyers showed up there. The range floor is built out of those same swing lows. Three different rules, three different traders, one price — and every one of them believes they've made an independent decision.
Which is why a move that goes exactly far enough to clear that price and no further isn't a conspiracy. It's what a crowd looks like from the outside.
This is the short seller's mirror of the trade above — the stop sits above a high rather than below a low — and it's the identical event either way up. Price rallies into $52.85, the high everyone measures from. The stops sit in the band just above it, straddling the round $53. One candle reaches $53.45, clears the crowded stop at $53.10, and closes back down at $52.10. Price then falls for the rest of the chart. A stop at $53.70, beyond the crowd, was never touched — and the only thing it cost was a smaller position.
The wider stop is the wrong lever
So the obvious fix presents itself. If $49.40 got hit, use $48. Give the thing room to breathe.
This works, in the narrow sense that a stop further away is hit less often. It also makes every single loss bigger — and most people apply it without touching anything else. Same 200 shares, wider stop. Which means the size of loss they are now willing to take has gone up without their ever having decided to take it.
Run it on our trade. 200 shares bought at $52 with the stop at $49.40 puts $2.60 a share at risk: $520 if you're wrong. Move the stop to $48 and keep the 200 shares, and you're risking $4 a share — $800. You did not decide to risk another $280. You decided to be stopped out less often, and the $280 came along uninvited.
Do that a few times in a row, each time after a stop-out, each time to a slightly wider level, and you arrive at the thing that actually empties accounts. Not the stop-outs. The four losses in a row that were each 60% bigger than the plan said.
Learn it by doing
Reading about it is one thing — it clicks when you do it. Learn it hands-on with free, interactive lessons on TradeWize.
Try the free lesson →Size is the lever, not distance
There are only two numbers in a stop-loss, and they multiply:
risk = position size × stop distance
Almost everybody fixes the size first — "I'll take 200 shares" — and then finds out what their risk is. That is choosing the number you care least about and letting the chart choose the one that can actually hurt you.
Turn it around. Decide the risk first, in money, before you look at where the stop goes. Say $520 is the most you're prepared to lose on this idea. Now the stop is free to go wherever the chart says it should — including well clear of the crowd — because the share count is what absorbs the difference.
Same $520 on the line. Same wider, less crowded stop. The only thing that gave way was the share count — and the share count is the number nobody's ego is attached to, which is exactly why it's the one to move.
The trade-off, stated honestly
A stop beyond the crowd is not free. You are choosing to be wrong less often and to lose more per share when you are, and to hold a smaller position the whole way up when you're right. That is a real cost and anybody who tells you otherwise is selling something. What you're buying with it is that the losses you do take are the ones you planned for, rather than the ones a predictable pile of orders handed you.
Where to put it instead
Two principles, and they pull in the same direction.
First: beyond the level, not at it. If the swing low is $49.55, a stop at $49.50 is inside the crowd and a stop at $48.90 is behind it. The distance between those two numbers is small in dollars and enormous in how often you get taken out, because it's the difference between sitting in the pile of orders and sitting past it.
Second: let the stock's ordinary noise set the distance. Every instrument has a routine daily wobble, and a stop closer than that wobble will be hit by nothing more meaningful than a Tuesday. Traders usually measure this with average true range — the average distance between a day's high and low over the last couple of weeks — and place the stop some multiple of it beyond the level. The specific multiple matters far less than the habit of checking that the stop clears the noise at all.
Both principles push the stop further away, and both are affordable for exactly one reason: you fixed the risk first, so the position shrank to pay for them.
The part nobody can fix for you
There is no placement that cannot be reached. Move the stop beyond the obvious pile and you have moved it to a less crowded price, not a safe one — and if the move is big enough, or the gap at the open is wide enough, it goes through anyway. FINRA's own guidance on volatile markets makes the point flatly: a stop can be triggered by a short, sharp move and the price may then rebound to where it was.
Which leads to the conclusion people reach at about this moment, and it is a trap: trade without a stop. It solves the problem completely, in the way that removing the batteries solves a smoke alarm. A position with no stop has no defined loss, and a position with no defined loss cannot be sized, because there is no distance to divide into your risk. You have not removed the risk. You have removed the only number that told you how much of it you were holding.
The honest summary is the same one that runs through everything in technical analysis. A stop is not protection from being wrong; it's a decision, made in advance and in writing, about how much being wrong is allowed to cost. Being stopped out is that decision working. Being stopped out at the exact low of the move, repeatedly, is a placement problem — and placement is a sizing problem wearing a different hat.
How to place a stop without volunteering for it
- Decide the money first. Before you look at the chart, name the most you're willing to lose on this idea in dollars, not percent-of-nothing.
- Find the level the trade depends on. The swing low that must hold, the range floor, the price that would mean you were simply wrong.
- Put the stop BEYOND it, not on it. Behind the pile of everyone else's orders, not in the middle of them.
- Check it clears the noise. If the stop is closer than the stock's routine daily range, it will be hit by an ordinary day and tell you nothing.
- Divide, don't fudge. Position size = your risk ÷ the stop distance. Whatever number comes out is the number of shares, even when it's less than you wanted.
- Never widen a live stop. Moving it further away mid-trade is the one adjustment that converts a planned loss into an unplanned one. Decide before, not during.
The one-line version
You can't stop your stop being predictable, but you can stop it being in the same place as everyone else's — and the thing that pays for the move is the share count, not the stop.
What is stop hunting?
Stop hunting is the name given to a move that pushes price just far enough to trigger the stop-loss orders clustered beyond an obvious level, then reverses. The name implies deliberate targeting. What's actually observable is simply price reaching the one price where a lot of orders were resting, setting them off, and running out of fuel once they're exhausted.
Why does my stop loss always get hit?
Almost always because it's in the crowded place. Stops bunch up just beyond the levels everyone can see — under the swing low, past the round number, outside the range — so a fairly small move reaches a lot of them at once. If your stop sits in that band, it gets taken by moves that were never big enough to mean your trade was wrong.
Do brokers hunt your stop loss?
On an exchange-traded, centrally cleared market, no — and the mechanics get in the way of the story. A stop order is dormant until it's triggered, so it isn't displayed for anyone to see. The clustering effect shows up just as reliably in deep, heavily regulated futures markets as anywhere else, which is hard to square with anyone targeting individual retail orders.
Where should I place my stop loss?
Beyond the level your trade depends on, not on it, and far enough out to clear the instrument's ordinary daily range. Then size the position from that distance rather than the other way round: shares = the money you're willing to risk ÷ the distance to the stop. The stop's job is to be the price at which you were wrong, not a price you hope isn't reached.
Is a wider stop loss better?
Only if you shrink the position to match. A wider stop is hit less often, which is the appeal, but with the same number of shares it silently raises the money you lose when it is hit. Widening the stop and keeping the size is how a planned $520 loss becomes an unplanned $800 one.
Should I trade without a stop loss?
No, and the reason is arithmetic rather than discipline. Position size is your risk divided by the stop distance — so with no stop there's no distance, and with no distance there is no way to work out how large the position should be. Removing the stop doesn't remove the risk. It removes the only number that measured it.
What is a liquidity sweep?
A move that trades past an obvious level, triggers the resting orders there, and closes back inside — so the level technically held. It's the same event as a stop hunt described from the chart's side rather than the trader's. Our guide to break of structure covers what a sweep does to the level afterwards, which is the part most explanations skip.
Placement is a practised skill, not a rule you memorise
Knowing the stop goes beyond the crowd is the easy half. The hard half is looking at a live chart and deciding where the crowd actually is — which is a thing you only get good at by doing it repeatedly and being told immediately whether you were right. That's what TradeWize's technical-analysis track drills: mark the level, place the stop, size the position, take the verdict. Educational practice, not signals.