Liquidity Sweeps and Stop Hunts in ICT Trading Explained

In Buy-Side and Sell-Side Liquidity you learned where stops tend to rest. This lesson looks at what happens when price reaches them. Sometimes price grabs those orders and snaps back. Sometimes it keeps going. Telling the two apart, and accepting that you will sometimes be wrong, is a core ICT skill.

What it is

Sweep

A liquidity sweep (often called a stop hunt or liquidity grab) is a brief move beyond an obvious high or low that triggers resting orders and then quickly returns inside the prior range. On the chart, it typically appears as a wick poking through the level while the candle body closes back on the original side.

Run or break

A run or break is a move through the level that holds. Price closes beyond it, often more than once, and continues in that direction. In this case the liquidity acted as fuel for continuation rather than as a turning point.

The turtle-soup idea

The name "turtle soup" comes from older trading literature that faded the breakout entries of trend-following systems. ICT borrows the concept: when price breaks an old high or low and fails, traders who bought the breakout are trapped, and their exits can add fuel to a reversal. The setup is to trade back inside the range after the false break, with a stop beyond the sweep extreme.

How to identify it

1. Mark a significant liquidity level: equal highs or lows, a previous day high or low, or a session extreme.
2. Wait for price to trade beyond it, ideally inside a kill zone such as the New York AM window (07:00–10:00 ET).
3. Check the close: a close back inside the range is the first condition.
4. Look for displacement in the opposite direction.
5. Confirm a market structure shift through the nearest swing point.
6. Identify a fair value gap or similar imbalance in that displacement for a retrace entry.
7. Place the stop beyond the sweep extreme, never inside the range noise.

Worked example: a sell-side sweep on EUR/USD

Prices are illustrative, on a 5-minute EUR/USD chart, during the New York AM kill zone, with a bullish daily bias.

Two lows at 1.0850 and 1.0851 form an obvious sell-side pool. Price drops to 1.0838, 12 pips below the lower of the two, then the 5-minute candle closes at 1.0856, back above the lows. That is the sweep.

Price then rallies in three strong candles and closes above the swing high at 1.0880: displacement plus an MSS. The move leaves a fair value gap between 1.0862 and 1.0868.

• Entry: a buy limit at 1.0865, inside the gap.
• Stop: 1.0833, 5 pips below the sweep low. Risk = 1.0865 − 1.0833 = 0.0032 = 32 pips.
• Target: the old high at 1.0929, where buy-side liquidity rests. Reward = 1.0929 − 1.0865 = 0.0064 = 64 pips.
• Risk-reward: 1:2.

Now consider the alternative. Suppose instead the candle had closed at 1.0830, below the lows, and the next two candles also closed lower with no rally. That would be a run on sell-side liquidity, not a sweep. The bullish idea would be off, and a trader who bought simply because price "hit the lows" would be fighting a breakdown.

Even the valid version can fail: price could fill at 1.0865, drop back below 1.0838 and hit the stop. In fast markets, the stop may also fill with some slippage. Size the position so that 32 pips is a small, acceptable loss using the Position Size Calculator.

Why not every sweep reverses

• The higher-timeframe draw points through the level. If the daily chart is heading for a lower target, a sell-side sweep may just be the start of the move.
• No displacement follows. A wick without strong follow-through suggests little real interest on the other side.
• The level was minor. Small intraday lows get taken constantly on the way to bigger targets.
• News. High-impact releases on the Economic Calendar can push price through levels regardless of structure.

This is why ICT traders look for a sequence (sweep, close back inside, displacement, structure shift) rather than acting on the wick alone. Related ideas include the Judas Swing, a session-based false move, and Candle Range Theory, which frames sweeps of a prior candle's range.

How traders use it

• Reversal entries in the direction of the higher-timeframe bias after the opposing liquidity is taken.
• Avoiding traps: waiting for a sweep before entering stops you from being part of the liquidity yourself.
• Stop placement: the sweep extreme offers a logical invalidation point, since a return beyond it suggests the idea has failed.
• Context for continuation: a run that holds beyond a level can confirm a break of structure instead.

Common mistakes

• Trading the wick alone. Wait for the close and some displacement.
• Fading every breakout. Many breaks are real. Use your bias to filter.
• Stops too tight. A stop one pip beyond the sweep low is easily taken by a second probe. Allow a small buffer.
• Ignoring news timing. Sweeps during data releases are especially erratic.
• Revenge trading after a failed sweep. Accept the loss, then reassess rather than re-entering immediately.

Summary

A sweep trades beyond a liquidity level and quickly returns; a run or break holds beyond it. The turtle-soup idea fades failed breakouts of old highs and lows. ICT traders confirm a sweep with a close back inside, displacement and a market structure shift, then enter on a retrace with a stop beyond the sweep extreme. Many sweeps do not reverse, so always use a defined stop and size each position for the risk you can accept.