Liquidity Sweeps and Stop Hunts: What Actually Happens to Your Stop

Price runs a few pips past the obvious high, triggers every stop resting above it, and then reverses hard. It happens constantly, it feels personal, and it is not. Understanding why it happens turns the most demoralising pattern in trading into one of the more readable ones.

Liquidity Sweep / Stop Hunt at a glance

What it is and why it works

Start with the mechanics, because they explain everything else. A stop-loss order is not a passive instruction; when it is triggered it becomes a market order. If you are long and your stop is below the recent low, your stop is a sell order that will execute the instant price reaches that level. Now consider that thousands of traders put their stops in the same obvious place. Below that low sits a large, dense cluster of sell orders that will all fire together.

That cluster is what people mean by a liquidity pool. Liquidity, in this context, simply means orders available to trade against. If you need to buy a very large position, your problem is finding sellers, and a dense cluster of stop-loss sell orders is a supply of sellers who are guaranteed to sell the moment price reaches them. So price goes there. Not because anyone is looking at your account, but because that is where the counterparty is.

The terminology follows from this. Buy-side liquidity means clustered buy orders, which sit above highs; the stops of short sellers and the entry orders of breakout buyers. Sell-side liquidity means clustered sell orders, sitting below lows. Equal highs or equal lows, two or more swings that stop at almost exactly the same price, are a particularly attractive target, because the repetition concentrates orders at one level rather than spreading them out. A sweep or liquidity grab is price trading through that level, triggering the cluster, and then failing to continue.

Now the part this page exists for. Your broker is not hunting your individual stop. In a market where hundreds of billions change hands daily, no participant can move the price of EUR/USD to reach one retail trader’s stop, and it would be economically absurd to try. What is actually happening is impersonal and mechanical: your stop is in the same place as everyone else’s, that place is a pool of guaranteed order flow, and large participants transact where the flow is. The feeling that you were targeted comes from the fact that you were, as part of a crowd of thousands, identified by where you placed your order rather than by who you are. Two honest caveats belong alongside that. A market-making broker who takes the other side of your trade does see your order, and while regulated firms operate under rules about execution quality, the incentive question is real, which is why choosing a properly regulated broker matters; see our broker comparison. And in genuinely illiquid instruments (exotic pairs, thin CFDs, out-of-hours trade) price feeds can and do wick in ways that have more to do with a single provider than with the global market. The conspiracy framing is wrong; the vigilance is not.

One more point of intellectual honesty. This is not a new discovery. Richard Wyckoff described the same behaviour a century ago as the spring and the upthrust; Linda Raschke published a systematic version as Turtle Soup in the 1990s, built on the observation that false breakouts of twenty-day extremes were tradeable. Much of the modern vocabulary is a relabelling of a long-understood pattern. That does not make it less real, the mechanics are sound, but it should temper the sense that this is secret knowledge.

How to trade it, step by step

1. Mark the levels where stops must be sitting, before the session. The reliable ones are: the highest high and lowest low of the previous session, the extremes of the Asian range, obvious swing highs and lows on the 1-hour chart, and any pair of equal highs or equal lows. Round numbers reinforce all of these. Draw them and leave them; the whole method depends on having decided what is obvious before price gets there.
2. Prefer equal highs and equal lows above everything else. When price stops twice at almost exactly the same level, orders concentrate in a narrow band instead of being spread across a range. That concentration is what makes the level worth reaching, and it is why double tops and double bottoms are swept so frequently and so precisely.
3. Wait for price to trade through the level. Do not anticipate. There is no trade before the sweep. Attempting to position for the sweep in advance is simply a counter-trend trade at a level, and it puts you on the wrong side of the actual break when the break is real. The setup begins the moment the level is exceeded, not before.
4. Require a fast reclaim, and define fast in advance. The signature of a sweep is that price trades beyond the level and comes back quickly: typically within one to three candles on your execution timeframe, leaving a long wick and closing back inside the prior range. If price stays outside the level and starts building bars there, you are watching acceptance, which means a genuine break. That distinction is the entire trade.
5. Look for a shift in structure to confirm. After a sweep of a high, the confirming evidence is price then breaking the most recent minor low on a smaller timeframe; the first sign the short-term sequence has changed direction. Waiting for that costs you some of the move and removes a large share of the trades where the sweep was really just the start of a breakout.
6. Enter on the reclaim or on the retest of the level from the inside. The aggressive entry is the close of the candle that reclaims the level. The patient entry is a pullback to the swept level from the other side, which often holds precisely because the orders there have now been consumed. Choose one approach and stay with it rather than deciding in the moment.
7. Put the stop beyond the wick of the sweep, not at the level. The extreme of the sweep is the point at which your thesis is wrong, so the stop belongs just past it. Placing it at the swept level itself puts it inside the zone that was just proven to be a magnet for orders. Then derive the position size from that distance with the position size calculator, rather than fitting the stop to a size you had already decided on.
8. Target the opposite pool. If price has just swept the highs and reversed, the natural destination is the liquidity resting below the most recent lows, because that is where the next concentration of orders sits. This gives a target defined by market structure rather than by an arbitrary reward multiple, and it is one of the more genuinely useful contributions of this framework.
9. Record every sweep you take, including the failures. This setup has an unusually deceptive hindsight quality: on a completed chart the sweeps are obvious and the failed ones are invisible because they simply became breakouts. Keeping a log of every level you traded, swept or not, is the only way to know whether you are reading it or fitting it.

Size every one of those entries with the position size calculator and check the trade is worth taking with the risk/reward calculator before you commit.

The conditions it needs

An obvious level that many traders can see

The pool only exists because orders concentrate, and orders concentrate at levels that are visually unmissable: the previous day’s high, a clean double top, a round number, the overnight extreme. Levels you found by drawing an unusual trendline or by using an obscure Fibonacci extension have no crowd behind them and therefore no pool to sweep.

A quick, decisive reclaim

The difference between a sweep and a breakout is entirely about what happens immediately afterwards. A sweep punches through, fails, and returns inside the range within a few bars, leaving a long wick. A breakout goes through and stays out, building bars at the new level. If you cannot state the reclaim condition as a rule before the trade, you will resolve every ambiguous case in favour of the trade you want.

A liquid instrument with a real crowd

This behaviour is clearest in markets with enough participants for stop clusters to be genuinely large: FX majors, major index CFDs, gold. In thin instruments the wicks past levels are frequently just poor liquidity and a wide spread rather than anyone transacting against a pool, and the pattern reads the same while meaning something entirely different.

A session where the sweep has a purpose

Sweeps cluster around session opens and around scheduled news, because those are the moments when large participants need to fill and when resting liquidity is most concentrated. The London open running the Asian range extremes is the archetypal example. A sweep in the middle of a dead session is more likely to be drift than a fill.

When it fails

• It was a real breakout. This is the fundamental problem and it has no clean solution. For the first thirty seconds a genuine breakout and a liquidity sweep are identical, and the trader who assumes every push past a level is a sweep will end up systematically short at the start of every rally. Requiring the reclaim before entering is what separates the method from a habit of fading strength.
• Retrospective level-drawing. On a historical chart, sweeps look flawless, because you can see which level got taken and then reversed. Live, you have many candidate levels and no idea which one matters. Traders read a chart backwards, conclude the pattern is reliable, and are then astonished by how ambiguous it is in real time. Marking levels before the session is the only defence.
• The sweep runs much further than expected. Nothing dictates that a liquidity grab stops just past the level. Stops trigger, that triggering pushes price, which triggers more stops, and a cascade can travel a long way before it exhausts. A stop placed a few pips beyond the level is inside that cascade. This is why the stop belongs beyond the full extent of the wick and why the position must be sized for that wider distance.
• Treating it as proof of manipulation. The conspiracy reading, that the broker or a cartel is targeting you specifically, is both wrong and practically harmful, because it converts a readable market behaviour into a grievance. If price is being drawn to obvious levels, that is information you can use. If you have been personally targeted, there is nothing to do but feel aggrieved. Only one of those framings improves your trading.
• Moving stops away to avoid being swept. The natural response to repeated stop-outs is to widen the stop or remove it. Widening it to sit beyond the structure is correct, provided you reduce the position accordingly. Widening it while keeping the same size, or trading without a stop because “they always hunt it”, converts a survivable pattern of small losses into a single account-threatening one.
• Confusing a wick with a fill. In illiquid conditions (exotic pairs, out-of-hours, thin CFDs) a long wick through a level can simply be a poor price from one provider rather than a real transaction against a pool. The chart looks like a textbook sweep and nothing meaningful happened. Check whether the instrument and the hour can support the interpretation you are placing on them.

Which markets this works best on

• GBP/JPY: Wide ranges and violent wicks make sweeps unusually visible, and unusually expensive to misjudge.
• Gold (XAU/USD): Round numbers attract enormous order concentration, and gold respects them almost theatrically.
• NAS100 (Nasdaq 100): The US cash open regularly sweeps the overnight extremes before setting the day’s direction.
• GBP/USD: Deep liquidity and a heavily watched London open produce clean, well-defined sweeps.
• EUR/USD: The tightest spreads make the reclaim entry viable where wider-spread instruments would not be.

For different levels of experience

If you are brand new

If you have ever been stopped out to the pip and then watched the market go exactly where you thought it would, this page is about that experience. Here is what happened, and it is simpler and less sinister than it feels.

When you set a stop loss, you create an order. If you are long, your stop is a sell order sitting below the market. The catch is that you put it in the obvious place, just below the recent low, and so did thousands of other people. Below that low there is now a big pile of sell orders that will all fire at once if price gets there.

Now think about someone who wants to buy a very large amount. They need people to sell to them. That pile of stop orders is exactly that: guaranteed sellers at a known price. So price gets pushed down to it, all those stops fire, the big buyer gets filled, and then price goes up. You were not targeted. Your order was just standing in the busiest place on the street.

What to do about it as a beginner is mostly defensive. Do not put your stop at the most obvious price on the chart: put it beyond the whole structure, and make your position smaller so the wider stop still risks the same amount of money. And absolutely do not conclude that stops are pointless. Trading without a stop because you keep getting swept is how small, recoverable losses turn into one that ends the account.

If your results are inconsistent

Most intermediate traders can already spot a sweep on a chart after the fact. The two things that turn that into a usable method are pre-marking and a hard reclaim rule.

Pre-marking means writing down, before the session, which levels the crowd can see: previous day high and low, Asian range extremes, any equal highs or equal lows, the round numbers nearby. If you choose the level after price has already reversed, you are not trading a pattern, you are narrating one. This single discipline change reveals how much of your apparent accuracy was hindsight.

The reclaim rule is what stops you fading real breakouts. Define it numerically: price must trade back inside the level and close there within a stated number of candles on your execution timeframe. If it does not, the level was accepted and the break is real, and the correct response is to stand aside or to trade with the break, not to keep waiting for a reversal that has already been refused. Adding a structure-shift confirmation on a lower timeframe filters more still.

Then take the targeting seriously. The most useful practical output of this framework is that it tells you where price is likely to go next: towards the opposite pool. A sweep of the highs that reverses is heading for the liquidity beneath the lows. Setting targets at structural destinations rather than at arbitrary multiples is a genuine improvement and it is available whether or not you buy into the wider vocabulary.

If you are experienced

Strip the branding and this is a stop-cascade phenomenon with a well-documented basis. Clustered resting stops create a locally convex supply of marketable orders; execution against them is efficient for a large participant; and the price impact of the cascade is temporary rather than informative, which is why reversion follows. The same behaviour is described in the microstructure literature as stop-loss triggered momentum and its subsequent reversal, and it was described a century earlier by Wyckoff as the spring and the upthrust.

The practical consequences differ from the retail version in two ways. First, the informative object is not the wick but the failure to gain acceptance afterwards: the absence of volume building at the new level. In an instrument with real volume that is directly observable on a profile; in spot FX it has to be inferred from price behaviour, since tick volume is not transacted size. Second, the cascade length is a function of resting depth, which varies enormously by session and by proximity to scheduled events, so a fixed stop offset beyond the level is mis-specified across regimes.

On the manipulation question, be precise. Coordinated attempts to move a fix have been prosecuted in FX, and spoofing has been prosecuted in futures; these are real, documented, and they concern institutional-scale interactions, not retail stop placement. No participant is moving a major pair to reach an individual order. The correct professional stance is that the drawing of price toward resting liquidity is a mechanical feature to be traded, while execution quality at your specific venue is a separate diligence question about counterparty and regulation.

Risk management for this strategy

The defining risk here is that the stop must be wide. Your invalidation point is beyond the full extent of the sweep, and a stop cascade can run considerably further than the level that triggered it before exhausting. A stop placed a few points past the swept high is sitting inside the very move you are trying to trade. The correct response is a wider stop and a proportionally smaller position, and the incorrect response, which is extremely common, is to keep the position size and widen the stop, which silently multiplies the risk on every trade.

The second risk is asymmetric and specific: when this setup is wrong, it is wrong because you are positioned against a genuine breakout. That means you are counter-trend at the start of the strongest move available, which is the single worst place to be stubborn. Take the stop the first time. Re-entering a fade against an accepted break is how one wrong read becomes several.

Third, sweeps concentrate around news and session opens, which are also the moments when spreads widen and fills degrade. A stop in those conditions is filled at the next available price, not at yours. Assume slippage on this strategy rather than treating it as an exception, and avoid holding a tight stop through a scheduled release.

Finally, resist the two behavioural traps this pattern creates. Do not trade without a stop because you keep being swept, that is trading unlimited downside to avoid a limited one. And do not increase size after a run of sweeps to make the losses back; the frequency of this setup makes revenge sizing unusually easy to act on and unusually damaging.

Where Market Structure Pro fits

The whole difficulty in this strategy compresses into one moment: price has just traded through the level, and you must decide whether it is coming back or whether it has gone. A sweep and a breakout are the same event until they are not, and there is no candle that tells you which is which at the time.

Market Structure Pro is built around precisely that class of judgement. It fuses twenty-seven tools into one verdict (TRADE, TRANSITION or NO TRADE) with a confidence percentage, an A/B/C grade and a plain-English explanation of what is supporting or limiting the read. The TRANSITION state is directly relevant here, because a market in the middle of resolving a level is exactly the condition that a binary buy-or-sell tool has to force into a category it does not belong in.

Two properties matter more than usual on this setup. It is non-repainting: state locks on the closed bar, so the reading that existed when you took the trade is still there afterwards. On a strategy defined entirely by whether price closes back inside a level, an indicator that quietly revises its own history is worse than useless. And it is spread-aware and session-aware, which matters because sweeps cluster at session opens and around news, the same moments when the spread widens enough to change the arithmetic of a tight reclaim entry.

The ranging filter contributes too, by returning NO TRADE when the market is simply chopping around a level rather than sweeping it, which is the condition that generates the most false sweep signals. MSP does not place trades, is not a signal service, and guarantees nothing; it gives you a consistent second reading at the one moment where hindsight is unavailable and hesitation is expensive.

Frequently asked questions

Is my broker hunting my stop loss?

Almost certainly not in the way it feels. No participant can move a market like EUR/USD, where enormous volumes trade daily, to reach one retail trader’s order. What happens is that your stop sits in the same obvious place as thousands of others, that cluster is a supply of guaranteed orders, and large participants transact where that supply is. The caveats are that a market-making broker does see your order, and that thin instruments can produce feed-specific wicks, which is why regulation and broker choice matter.

What is a liquidity sweep?

A liquidity sweep is price trading through an obvious level (a swing high, a double bottom, a round number) triggering the stop orders clustered there, and then reversing rather than continuing. The stops become market orders when triggered, which provides the counterparty a large participant needs in order to fill. The reversal is what distinguishes a sweep from a genuine breakout.

What is the difference between a liquidity sweep and a breakout?

Only what happens afterwards. A sweep pushes past the level, fails, and returns inside the prior range quickly, usually within one to three candles and leaving a long wick. A breakout pushes past and stays, building bars and accepting the new price. In the first thirty seconds they look identical, which is why the method requires waiting for the reclaim rather than anticipating it.

What does buy-side and sell-side liquidity mean?

Buy-side liquidity means clustered buy orders, which sit above highs: the stop losses of short sellers plus the entry orders of breakout buyers. Sell-side liquidity means clustered sell orders sitting below lows. The names describe what the resting orders will do when triggered, not who placed them, which is a frequent source of confusion for people meeting the terms for the first time.

Why do equal highs and equal lows get swept so often?

Because repetition concentrates orders. When price stops twice at almost exactly the same level, traders read it as a double top or bottom and place their stops in the same narrow band rather than spread across a range. That concentration makes the level an unusually efficient place for a large participant to fill, which is why clean double tops and bottoms are broken so precisely and so often.

Where should I put my stop loss to avoid being swept?

Beyond the structure rather than at the obvious level, and then reduce your position size so the wider stop still risks the same amount of money. If you are long, that means below the whole swing area rather than a few pips under the low everyone can see. The mistake to avoid is widening the stop while keeping the same position size, which quietly multiplies your risk on every trade.

Is the liquidity sweep a new concept?

No. Richard Wyckoff described the same behaviour around a century ago as the spring and the upthrust, and Linda Raschke published a systematic version called Turtle Soup in the 1990s based on false breaks of twenty-day extremes. The mechanics are genuine and long understood; much of the current vocabulary is a relabelling rather than a discovery, which is worth knowing when material is presented as hidden knowledge.

How do I know which level will be swept?

You do not, and any material implying otherwise is overselling it. What you can do is mark the levels that are obvious to the crowd before the session (previous day high and low, Asian range extremes, equal highs and lows, nearby round numbers) and then react to whichever one price actually reaches. Choosing the level after the reversal has happened is narration, not analysis.

Should I trade without a stop loss if stops keep getting hit?

No. That trades a limited, survivable loss for an unlimited one, and it is one of the most reliable routes to ending an account. If sweeps are repeatedly taking you out, the fix is to place the stop beyond the structure rather than at the obvious level, and to cut position size so the wider stop costs the same. The problem is stop placement and sizing, not the existence of the stop.

Related reading

• Liquidity Explained: The foundation this entire strategy rests on, explained from scratch.
• Break of Structure and CHoCH: The confirmation step that turns a sweep into an entry.
• Smart Money Concepts: The wider framework these terms come from, assessed honestly.
• London Open Breakout: Where sweeps happen most predictably: the session open running the overnight extremes.
• Broker Comparison: Execution model and regulation are the parts of the stop-hunt question you can actually control.