Liquidity Sweep (Stop Hunt): How to Spot One and Whether They Reverse

A plain-language guide with live examples and backtest data across BTC, ETH, and SOL.

Jay Sharma · Founder, Botsfolio

Published August 18, 2026

What it is

A liquidity sweep is a two-part price action pattern. First, price pushes past an obvious support or resistance level where stop-loss orders are clustered. This triggers those stops, generating a burst of orders in one direction. Second, price fails to hold beyond the level and quickly reverses back into the prior range, leaving a wick past the level rather than a sustained break.

The mechanic behind the pattern is straightforward. Retail traders place stops in predictable locations (just beyond a recent swing high, below equal lows, past a round number). Those stops represent orders in the opposite direction of the initial position. Large participants who want to buy in size need someone to sell to them; the concentrated stops beyond a swing low provide that supply. A brief push below the level triggers the stops, provides the liquidity, and the position is filled. Price then reverses because the actual intent was the opposite of the wick direction.

The distinction between "liquidity sweep" and "stop hunt" is largely semantic. Stop hunt implies intent by identifiable market participants. Liquidity sweep is the neutral, structural term that describes the same price action without assigning motive. This guide uses liquidity sweep as the primary term because it is more accurate about what can actually be observed on a chart.

Two things people commonly get wrong about sweeps, which this guide will not:

• Not every wick past a level is a sweep. The reclaim (price closing back inside the range) is what confirms the pattern. A wick past that closes below produces a sustained break, not a sweep.
• Not every sweep reverses meaningfully. Around 30 to 35 percent of sweep-and-reclaim patterns in our data grind sideways rather than producing a clean reversal move. The setup fires; the follow-through does not always come.

How to spot one on a chart

Liquidity sweeps form in three phases. All three should be visible before treating the pattern as valid.

1. The liquidity pool

Start by identifying where stops are likely to cluster. Common locations are:

• Just above a recent swing high, where breakout traders' stops sit
• Just below a recent swing low, where trend followers' stops sit
• Beyond equal highs or equal lows (double tops and bottoms), where the pattern is obvious
• Just past a round number ($65,000, $70,000, etc.), where round-number stops cluster

The more obvious the level, the more stops cluster there, the higher the liquidity pool.

2. The sweep candle

Once liquidity exists, the sweep is a single candle (or occasionally two) that wicks past the level with a strong move but does not sustain the break. For a bullish sweep at prior lows, the candle wicks below the low and closes back above it. For a bearish sweep at prior highs, the candle wicks above the high and closes back below.

3. The reclaim

The reclaim is what separates a sweep from a genuine breakdown or breakout. After the wick past the level, the same candle or the next candle must close back inside the prior range. Without the reclaim, the pattern is not a sweep; it is a directional continuation.

How Botsfolio detects and measures it

Every candle close, our analysis engine scans recent price structure for stop-cluster locations (equal highs and lows, prior swing points, round numbers). When a candle wicks past one of these zones and closes back inside, we flag it as a potential sweep. Three pieces of data are recorded:

• The level swept (price and structural type: prior high, equal low, round number, etc.)
• The wick depth beyond the level (in dollars and percent of price)
• The reclaim distance and confirmation strength

The reaction is then tracked continuously: reclaimed and reversed, reclaimed but chopping, reclaim failed. Every outcome becomes a row in our backtest.

Two things worth naming, because they shape the numbers:

• We only surface sweeps with a minimum wick depth relative to trailing 14-period ATR. Very shallow wicks technically qualify but historically underperform.
• Our cost model assumes 0.12 percent round-trip fees, already deducted from every expectancy R and annual gain. Real fees vary by exchange and tier.

Full methodology at our methodology page.

A live example on BTC right now

Here is a live liquidity sweep on BTC, drawn as it looked when it formed, alongside how the pattern has performed across timeframes. If nothing is currently active, the widget shows the most recent formed example and its outcome.

Historical performance across coins and timeframes

The table below is aggregate performance of liquidity sweep setups across the coins we backfill. Book-early management means partial off at first target with the remainder trailed. Both directions combined.

Reading the table honestly, three observations:

Sweeps have among the highest win rates of any SMC pattern in our data, particularly on higher timeframes. On BTC 4H the win rate sits at 62 percent; on BTC 6H it reaches 65 percent; on BTC 1D it approaches 68 percent. This aligns with the intuition that stop-hunt patterns are cleaner when they occur on higher timeframes because the liquidity pools are larger.

Sample sizes are smaller than for order blocks or FVGs, especially on higher timeframes. Sweeps require specific structural conditions (identifiable liquidity pools) that do not form every day. Around 20 to 30 percent of the frequency of order blocks on the same timeframe.

Popular sources cite sweep win rates in the 70 to 80 percent range. Our data does not support that. Even on the daily timeframe, the honest measured win rate under fee-inclusive backtest is around 68 percent. The higher numbers in circulation typically ignore fees, count only clean-reversal outcomes, or use small hand-picked samples.

What tends to invalidate a sweep

Every backtested setup carries a reversal rate: the percentage of instances that reached +1R at some point, then finished at or below breakeven.

Three failure patterns account for most sweep invalidations in our data:

No reclaim. The sweep candle wicks past the level but closes below (for a bullish sweep) or above (for a bearish sweep). This is a sustained break, not a sweep. Traders who read the wick alone and enter get run over. Our detector requires a genuine close back inside the range before flagging the pattern.

Weak reclaim followed by re-test. The reclaim closes back inside but a subsequent candle taps the swept level again. Re-tests of swept levels have low base rate follow-through in our data. Traders who add to the initial sweep entry on a re-test often see the re-test become a sustained break.

Sweep during trending momentum. Sweeps that occur mid-trend, especially in strong directional moves with elevated volume, often fail to reverse. The sweep triggers, briefly reclaims, then continues in the impulse direction. Trend context matters enormously for this pattern.

How sweeps interact with other patterns

Liquidity sweeps combine powerfully with other Smart Money Concepts patterns. The strongest sweep setups have confluence.

• Sweep plus order block. When the sweep candle originates from or reacts inside an order block, the confluence is strong. The sweep provides the trigger; the order block provides the structural context.
• Sweep plus fair value gap. When the sweep-and-reclaim impulse creates a fair value gap, price often returns to fill the gap after the initial reversal. Two setups in one move.
• Sweep plus break-of-structure. The strongest sweep pattern is a sweep of a swing low followed by a break of the prior swing high (change of character). This confirms the reversal thesis structurally, not just from the wick.

Related: sweeps that fail to reclaim often lead to strong directional continuation, marking the strength of the underlying trend. See Break and Retest for the follow-through continuation pattern.

How reads of this concept commonly go wrong

Five patterns show up repeatedly when we look at how liquidity sweeps get misread. Each is framed as an observation about the data.

1. Entering on the wick without waiting for the close. A wick past a level without a close-back-inside is a directional break, not a sweep. Entering on the wick alone flips the base rate against the trader.
2. Ignoring the trend context. Sweeps in the direction opposite of the higher-timeframe trend have lower base rates. A bullish sweep against a strong bearish 1-day trend underperforms one aligned with the trend.
3. Trading sweeps of shallow or thin liquidity pools. Not every wick past a small local low is a meaningful sweep. The level must have been genuinely obvious, with clustered stops likely present.
4. Assuming every sweep leads to a full reversal. Around 30 percent of sweep-and-reclaim setups in our data grind sideways or produce only shallow reversals. Position sizing off "full reversal expected" produces oversized risk.
5. Reading equal highs or lows as sweep candidates without volume context. Equal highs or lows on low-volume conditions often do not have real stops behind them. The pattern looks the same but the mechanic that drives sweeps is absent.

Frequently asked questions

Related concepts

Educational analysis, not financial advice. Past performance does not predict future results.