In trading, liquidity means the resting orders that allow other orders to be filled. In the SMC reading, it clusters above obvious highs (buy-side liquidity: sellers' stops and breakout buys) and below obvious lows (sell-side liquidity: buyers' stops). Price often reaches for those areas. What it does next, turn or break, is something you measure on your trades.
You buy at a clean support, two lows at the same level. Your stop sits just below, like everyone else's. A candle dips, takes your stop by a few pips, and rallies without you. The level held. Your stop did not. This article explains why that scene is so common, and how to use it without falling for the narrative.
What is liquidity?
Strictly, the ability to buy or sell a given size without moving price too much. To buy, you need a seller, and vice versa. A large order needs a lot of counterparties.
Stop orders provide them. A buy stop (a seller's stop, or a breakout buyer's entry) becomes a buy order when price touches it. A sell stop becomes a sell order. Where they pile up, there is enough to fill a large order on the other side.
Where does liquidity cluster?
Wherever traders put their stops and breakout orders: around levels everyone can see.
| Zone | What sits there | SMC name |
|---|---|---|
| Above an obvious high | Sellers' stops, breakout buys | Buy-side liquidity (BSL) |
| Below an obvious low | Buyers' stops, breakout sells | Sell-side liquidity (SSL) |
| Equal highs | Two or more highs at the same level: a "perfect" resistance | Concentrated BSL |
| Equal lows | Two or more lows at the same level: a "perfect" support | Concentrated SSL |
| Previous day or week high/low | Levels everyone shares | Reference liquidity |
| Round numbers | 1.1000, 20,000 points | Clustered orders |
The cleaner a level looks, the more orders it attracts.
Sweep or break: how do you tell them apart?
| What price does | Reading | What often follows |
|---|---|---|
| Wick beyond the level, close back inside | Sweep: stops taken, level holds | A move the other way |
| Body closes beyond, then continues | Break: liquidity fuelled the move | Continuation |
| Close beyond, then quick return inside | Failed break | Often a strong move the other way |
The criterion is the same as for structure: the close, on the timeframe where you read the level. BOS or CHoCH covers that rule.
In the typical SMC setup, a sweep of liquidity against the higher trend comes before the structure shift and the return into an order block or an FVG. The Smart Money Concepts guide lays out the full sequence.
A worked example
EURUSD, H1, two lows at 1.0750 and 1.0751 (equal lows), bullish H4 structure. Example levels:
| Step | Level | Reading |
|---|---|---|
| Equal lows | 1.0750 | SSL clustered below 1.0750 |
| Wick below the level | 1.0738 | Sweep: 12 pips below the equal lows |
| H1 close | 1.0756 | Back above: no break |
| Entry after the close | 1.0756 | |
| Stop below the sweep wick | 1.0732 | Risk: 24 pips |
| Target: previous high | 1.0810 | 54 pips, about 2.25 R |
The trader with a stop at 1.0745, just below the equal lows, was taken out by the sweep. The one who enters after the close puts the stop below the wick, where the idea is actually invalidated, and adjusts size to that wider stop.
Log whether a sweep came before each trade →
Where should your stop go once you know about liquidity?
Where your idea is invalidated, not where everyone puts theirs.
- Avoid the stop "just below the level". That is the liquidity zone itself.
- Place the stop beyond the plausible sweep zone, and reduce size to keep the same risk: size = risk ÷ (stop distance × point value).
- If the logical stop is too far, the trade is not for you at that size. Reduce or pass.
How do you measure what price does after a sweep?
| Field | Example |
|---|---|
| Level type | Equal lows, previous day low, round number |
| Timeframe | H1 |
| Outcome | Sweep / break / failed break |
| Sweep depth | 12 pips, i.e. 0.5 ATR |
| Move after 10 candles | +40 pips the other way |
| Trade taken | Yes / no, result in R |
After 50 cases, compare: what share of sweeps produced a move of at least 1 R the other way? On which level types? At what times? That is the only reliable answer. Sample size explains why you need at least that many.
The limits of reading liquidity
- Intent is not observable. "Institutions hunt stops" is an interpretation. What you can observe is that price often runs past obvious levels before turning. The mechanism may be simpler: orders triggering in a cascade.
- Not every level gets swept. Many simply break. A sweep is only confirmed on the close, after the fact.
- Context matters. A sweep against a strong higher trend may just be the start of a break.
This content is educational and is not investment advice. Trading involves a risk of capital loss.
Measuring your sweeps on Trade Stack
On Trade Stack, add "Sweep" to your Confluences in Settings and tick it when your entry follows a liquidity grab. After 50 trades, the By setup view (Pro plan) compares the win rate and average R of your trades with and without a sweep.
Measure my sweeps on Trade Stack
Key takeaways
- Liquidity clusters above obvious highs (buy-side) and below obvious lows (sell-side).
- Equal highs, equal lows, previous day highs and lows, round numbers: the densest zones.
- Wick beyond then close back inside: sweep. Close beyond: break.
- The stop goes where the idea is invalidated, beyond the sweep zone, with size adjusted.
- What price does after a sweep is measured over at least 50 cases logged with the same grid.
Method: worked example built for illustration (EURUSD H1, fictional levels). Ratio: (1.0810 − 1.0756) ÷ (1.0756 − 1.0732) = 54 ÷ 24 = 2.25 R. No reaction rate is quoted: it must be measured on your own trades.
