What Research Actually Says About Stop Hunts
If you’ve spent any time around ICT or Smart Money Concepts content, you’ve absorbed one idea as a near-law of the market: price sweeps a level to grab liquidity, then reverses. It’s in nearly every explainer, every YouTube breakdown, every “how to spot the reversal” thread. It’s treated as settled.
It isn’t. And the actual research on stop-loss order behavior — not trading-education content, but peer-reviewed market microstructure research using real order-book data — points in a different direction.
Table of contents
- What a “stop hunt” or liquidity sweep actually is
- What the popular narrative gets right (and skips)
- What the research actually found
- Why the reversal story feels true anyway
- What this means practically
- Where this leaves you
What a “stop hunt” or liquidity sweep actually is
Strip away the branding and the mechanics are simple. Retail and institutional traders alike cluster stop-loss and take-profit orders around obvious levels — recent highs and lows, round numbers, the edges of a range. When price reaches that level, those resting orders execute. In FX and other liquid markets, that cluster of executions can itself move price further in the direction of the sweep, because a stop-loss sell order is functionally a market sell order once triggered.
The “smart money” framing describes this as intentional: bigger participants push price into the cluster to fill their own orders against it. The mechanism is real. What’s contested is the outcome — specifically, the assumption that once the liquidity is taken, price reliably turns around.
What the popular narrative gets right (and skips)
To be fair to the narrative: the mechanism it describes — orders clustering at obvious levels, and those levels acting as magnets for price — is genuinely documented in market microstructure research. This isn’t a fabricated idea from scratch.
What gets skipped is the second half of the claim: that sweeping the level is itself the signal for a reversal. That’s the part carrying all the trading weight, and it’s the part with the thinnest evidence.
What the research actually found
The most relevant academic work here comes from Carol L. Osler, whose research is built on something most trading-education content doesn’t have access to: actual bank order books.
Her 2003 paper in the Journal of Finance, “Currency Orders and Exchange Rate Dynamics,” analyzed the complete order book of RBS/NatWest — a major FX dealer — covering August 1999 through April 2000. The dataset: 9,655 orders, representing more than $55 billion in face value, across USD/JPY, GBP/USD and EUR/USD. This is about as close as trading research gets to “we watched the actual orders,” rather than inferring intent from price action after the fact.
An earlier, related paper — a New York Federal Reserve Staff Report from 2002, “Stop-Loss Orders and Price Cascades in Currency Markets” — states its finding plainly: stop-loss orders “propagate trends and are sometimes triggered in waves, contributing to price cascades.”
Read that again. Not “and then price reverses.” Propagate trends. Price cascades. The dominant, documented effect of a cluster of stop-loss orders executing is that price continues in that direction — sometimes sharply, as one triggered order pushes price into the next cluster, which triggers more orders, and so on.
That’s the opposite of “sweep, then reversal.”
Why the reversal story feels true anyway
If the research says continuation is the more common documented pattern, why does “sweep and reverse” feel so intuitively correct to anyone who’s watched enough charts?
A reasonable explanation is selective memory, not malice. A sharp reversal off a swept level is dramatic and visually obvious — it looks exactly like the “smart money grabbed liquidity” story, and it’s the kind of chart that gets screenshotted and taught. A swept level that simply continues doesn’t get the same treatment. Nobody posts “price hit the low and just kept going, unremarkably” as a lesson. It doesn’t feel like a pattern — it feels like nothing happened, even though from a research standpoint, it’s the more commonly documented outcome.
This is a known bias in how trading concepts get taught: memorable examples get repeated, unremarkable ones get filtered out, and after enough repetition, the memorable case starts to feel like the rule rather than one of several outcomes.
What this means practically
None of this means liquidity sweeps are meaningless, or that Smart Money Concepts as a framework is wrong. It means the sweep alone is not sufficient information. A level being taken tells you liquidity was there and got used. It does not, by itself, tell you which of two very different things happens next: price cascading through in the same direction, or price reversing because the sweep exhausted the move.
If you want a number for how often standalone sweep setups actually work, the honest answer is: nobody has published a rigorous study with disclosed methodology on this specific question. What exists is a rough band from community backtesting — figures in the 35–55% range for standalone sweep entries turn up repeatedly in trader-run backtests, but these are self-reported, without disclosed sample sizes or methodology, and should be treated as a community estimate, not a study — a meaningfully different thing from Osler’s order-book data.
What the research supports is a shift in what to look for after the sweep, rather than treating the sweep itself as the trigger:
- Did price simply tag the level and continue (consistent with the cascade pattern Osler documents)?
- Or did price sweep the level and then break structure in the opposite direction — a genuine shift, not just liquidity being taken?
That second case — sweep followed by a confirmed structural break — is a materially different situation from a sweep alone, and it’s the distinction that separates a documented cascade from an actual reversal.
Where this leaves you
Doing this distinction by eye, on one chart, at one liquidity level, is manageable. Doing it consistently — across multiple concurrent liquidity levels, on a higher timeframe context you’re not staring at in the moment, across several instruments — is where manual reading starts to break down. That’s a real, practical limitation, not a hypothetical one.
One way to handle that gap is a tool that explicitly separates these two outcomes rather than labeling every swept level the same way. SSM IQ Market Structure Liquidity draws exactly this distinction — SWEEP versus BREAK versus RETEST — rather than treating “liquidity taken” as a single, undifferentiated event. That’s not a claim that it predicts which sweeps reverse; it’s a tool for seeing the distinction the research says actually matters, instead of eyeballing it under pressure.
Whatever you use — this tool, a different one, or nothing — the takeaway from the research stands on its own: a swept level is not, by itself, a reversal signal. It’s a piece of information. What happens on the next close is the rest of it.
Related SSM IQ Tools: SSM IQ Market Structure Liquidity · SSM IQ Liquidity Breakout Premium Related Articles: How to Verify a TradingView Indicator Doesn’t Repaint Sources: Osler, C.L. (2003). “Currency Orders and Exchange Rate Dynamics.” Journal of Finance, 58(5). · Osler, C.L. (2002). “Stop-Loss Orders and Price Cascades in Currency Markets.” Federal Reserve Bank of New York Staff Report No. 150.
SSM IQ tools are analytical software, not investment advice. Trading involves substantial risk. Author: SSM IQ Research · We Are Tech Sp. z o.o. · Updated: 2026-08-16