What is a liquidity sweep?
Quick answer
A liquidity sweep is a move that pushes through an obvious high or low to trigger the stop orders resting there, then closes back inside the prior range instead of continuing. You spot one by watching for a wick beyond a clean level, a close back inside within one to three candles, and a market structure shift in the opposite direction — that sequence is the entry trigger, not the wick itself.
Liquidity, in trading, is resting orders — stop losses and pending entries clustered above old highs and below old lows. A liquidity sweep is when price intentionally runs through one of those clusters, triggering the orders, and then reverses.
The reversal is the signature. A real breakout closes beyond the level. A sweep wicks through and rejects. The market needed the liquidity to fill institutional orders in the opposite direction.
Where liquidity pools sit
- Previous day high (PDH) and previous day low (PDL).
- Previous session high and low — Asian high, London high, NY AM low.
- Equal highs and equal lows — double tops and bottoms, where stops cluster.
- Trendline liquidity — stops along a clean diagonal.
- The high or low of the opening range.
Equal highs and lows are the most reliable. The cleaner the level, the more orders sit there, the more incentive smart money has to sweep it.
How to trade a liquidity sweep
- Identify a clear liquidity pool above or below current price.
- Wait for price to spike through it on the 1m or 5m.
- Wait for a Market Structure Shift — a strong candle that closes back inside the range and breaks the most recent micro low/high.
- Enter on the retrace into the FVG or order block created by that shift.
- Stop loss just beyond the sweep wick. Target the opposite liquidity pool.
The risk-reward on a clean sweep + MSS + FVG entry is typically 3R or better. That asymmetry is why ICT traders wait for the sweep instead of fading breakouts blindly.
Sweeps inside Killzones
The highest-probability sweeps happen during the first 30 minutes of the London Open and the New York AM Killzone. These sessions deliver the day's volatility and need liquidity to operate. A sweep of the Asian range at London Open, or a sweep of the London high at NY Open, are textbook ICT setups.
Sweep vs breakout — side by side
| Liquidity sweep | Real breakout | |
|---|---|---|
| How it touches the level | Wick through, no close beyond | Candle body closes beyond |
| Next 1–3 candles | Close back inside the prior range | Continuation, level holds as support/resistance |
| Volume signature | Spike on the wick, then fade | Sustained expansion |
| What it leaves behind | An FVG or order block on the rejection | An FVG in the breakout direction |
| How to trade it | Fade it after an MSS | Enter the retest in the breakout direction |
How to spot fake sweeps
Not every wick is a sweep. Watch for these confirmations before treating it as one:
- Price closes back inside the prior range within the next 1–3 candles.
- An FVG or order block forms on the rejection.
- The sweep happens inside a Killzone, not in dead hours.
- Higher-timeframe bias agrees with the reversal direction.
Without those, you're often looking at a real breakout disguised as a sweep — the worst place to fade.
Backtest sweeps on NQ
Open NQ Replay, jump to any major economic event day (FOMC, CPI, NFP), and step through the New York AM Killzone candle by candle. Mark the previous session highs and lows before you start. Count how often the first move of NY is a sweep of one of those levels. The pattern repeats almost every day — the simulator is the cheapest way to internalise that.
Frequently asked questions
What is a liquidity sweep in trading?
A liquidity sweep is when price runs through an obvious high or low to trigger resting stop orders, then reverses. It is a hallmark of institutional order flow in the ICT framework.
How do I confirm a liquidity sweep?
Wait for price to close back inside the prior range, for a market structure shift on the lower timeframe, and ideally for the sweep to occur during a high-volume Killzone like London Open or New York AM.
What's the difference between a sweep and a breakout?
A breakout closes beyond the level and continues. A sweep wicks through, rejects, and closes back inside the prior range — typically within 1–3 candles.
How do you spot liquidity sweeps?
Mark the clean levels before the session: previous day high and low, Asian range, and any equal highs or lows. Then watch only those levels. A sweep is a wick through the level with an immediate close back inside, ideally inside the London or New York AM Killzone, followed by displacement the other way.
What time of day do liquidity sweeps happen on NQ?
Most often in the first thirty minutes of the London Open (03:00–04:00 ET) and the New York AM Killzone (09:30–11:00 ET). Sweeps during the lunch hour and in overnight dead zones far more often turn into slow drifts rather than clean reversals.
Where do I place my stop after a liquidity sweep?
Just beyond the sweep wick — that extreme is the level the move was designed to reach, so a return through it means the read was wrong. Target the opposite liquidity pool, which usually gives 3R or better on NQ.
Can I backtest liquidity sweeps?
Yes. A bar-by-bar replay tool like NQ Replay lets you step through historical NQ sessions, mark old highs and lows, and count sweeps versus real breakouts to measure the edge yourself.
Related concepts
- Fair Value Gap (FVG)
A three-candle inefficiency the market often returns to fill.
- Inverse Fair Value Gap
What an FVG becomes once it's closed through — a flipped reversal level.
- Turtle Soup
The classic equal-highs / lows liquidity sweep reversal setup.
- SMT Divergence
Correlated-asset disagreement (NQ vs ES) as a reversal filter.
Data & methodology
All NQ and MNQ examples use Databento GLBX.MDP3 1-minute OHLCV candles for continuous front-month E-mini Nasdaq-100 futures, aggregated to higher timeframes (5m, 15m, 1h, 4h) so every bar agrees across timeframes. Sessions run 01:00–16:00 ET. ICT overlays are drawn algorithmically from the same price data used in the replay trainer.