"Liquidity" is two different ideas wearing the same word, and most confusion about it comes from mixing them up.
The first meaning is boring, measurable, and true: liquidity is how easily you can trade without moving the price against yourself. The second meaning is the Smart Money Concepts (SMC) story: pools of resting stop orders that "smart money" supposedly hunts. The first you can verify on your own account. The second is mostly a narrative you tell after the candle has already printed.
The plain meaning: how easily a trade fills
In market microstructure, liquidity is the ability to buy or sell quickly, in size, at close to the current price. The European Central Bank frames it along three dimensions that you can actually observe: tightness (the bid-ask spread), depth (how much volume sits in the order book near the price), and resiliency (how fast the price recovers after a large order hits it) (ECB: understanding financial market liquidity).
Those dimensions are not abstract. They show up as costs you pay on every fill: a wider spread, and slippage when your order eats through more than one price level of the book (Bookmap: evaluating stock liquidity, InvestingCube: liquidity in trading).
A market is "liquid" when the spread is tight, the book is deep, and a normal order barely nudges the price. It is "illiquid" when the spread is wide, the book is thin, and your own order becomes the news. This meaning is objective: you can measure it before and after you trade.
The SMC meaning: liquidity as pools of resting orders
The retail SMC vocabulary uses the same word for something else: clusters of pending orders sitting at obvious price levels.
- Buy-side liquidity is the stop-loss and buy orders resting above the market — above swing highs and resistance (XS: buy-side and sell-side liquidity).
- Sell-side liquidity is the stop-loss and sell orders resting below the market — below swing lows and support (LiteFinance: buy-side and sell-side liquidity).
- Equal highs (EQH) and equal lows (EQL) are two or more highs or lows at almost the same price. The idea is that stops pile up just beyond them, forming an obvious pool.
- A liquidity sweep or stop run is a sharp poke past that level that triggers the stops, followed by a reversal — read as "smart money grabbed the liquidity it needed to fill, then went the other way" (LiteFinance).
Here is the honest catch: buy-side liquidity sits above price, but the name comes from the buy stops resting there. The labels invert what most beginners expect, which is one reason the framework feels slippery.
| Microstructure liquidity | SMC "liquidity" | |
|---|---|---|
| What it is | Spread, book depth, resiliency | Clusters of resting stop orders |
| How you observe it | Quotes, order book, your own slippage | Inferred from chart levels after the fact |
| Can you measure it live? | Yes, before you trade | No; you see the raw book only if the venue shows it |
| Is it falsifiable? | Yes | Usually only in hindsight |
Where stops really cluster: the part that is true
The SMC story is not pulled from nowhere. There is a real, published microstructure kernel underneath it, and it is more specific than "smart money hunts you."
Using a large FX dealer's actual order book — 9,655 orders worth over 55 billion dollars across dollar-yen, sterling-dollar and euro-dollar, from August 1999 to April 2000 — Carol Osler documented that stop and profit orders cluster in predictable ways: **stop-loss sell orders cluster just below round numbers, stop-loss buy orders cluster just above them, and take-profit orders cluster on the round numbers themselves** (Osler: Stop-Loss Orders and Price Cascades in Currency Markets). She also found that exchange rates tend to move fast after reaching those stop clusters — self-reinforcing "price cascades" — which is a microstructure explanation for why support and resistance levels sometimes break violently (Osler: Currency Orders and Exchange-Rate Dynamics, FRBNY Staff Report 125).
So the observable facts are real: stops do cluster at obvious levels, and triggering a cluster can accelerate a move. That is the grain of truth the SMC narrative is built on.
The narrative problem: mostly post-hoc
The jump from "stops cluster and clearing them moves price" to "an institution deliberately hunted your stop" is where testability falls apart.
Notice how the story is usually applied. Price pokes above a high and keeps going: that was a breakout. Price pokes above the same high and reverses: that was a liquidity sweep. Both outcomes confirm the framework, which means the framework predicted nothing. A claim that is compatible with every result is not wrong — it is unfalsifiable, which is a different and worse problem.
Even the researcher who documented the clustering is careful here. Osler notes plainly that statistical analysis of prices near round numbers cannot prove the specific order-flow interpretation is the cause (Osler, price-cascades paper). And among traders themselves, whether brokers or "smart money" actively hunt stops is a long-running, unresolved argument, with plenty of participants pointing out that a big player's real problem is getting large orders filled at all, not persecuting a retail account (BabyPips forum: is stop hunting real?).
What an objective, testable definition would need
If you want to trade a liquidity idea instead of narrate it, you have to convert the story into rules a computer can check without knowing the outcome.
That means committing, in advance, to answers like:
- What counts as a pool? Exact rule for equal highs/lows — how many touches, within how many pips, over what lookback.
- What counts as a sweep? A precise trigger: how far beyond the level, closing back inside within how many bars.
- What is the edge, measured? Entry, stop, target, and expected value across hundreds of instances — including the sweeps that kept running and never reversed.
- What does it cost? Real spread, slippage and commission, because sweep levels are exactly where spreads widen and fills get worse.
Do that and "liquidity sweep" stops being a vibe and becomes a hypothesis with a hit rate. Most of the time, forcing the definition is what kills the edge — which is the point of forcing it. If you want the general version of that discipline, see why your backtest lies and out-of-sample testing in trading.
Frequently asked
Is buy-side or sell-side liquidity above the price?
Buy-side liquidity sits above the current price (the buy stops and stop-losses that would fire on a move up), and sell-side liquidity sits below it. The names refer to the order type that gets triggered, not the direction of price, which is why they feel backwards at first.
Are liquidity sweeps and stop hunts real?
The measurable part is real: stops genuinely cluster near round numbers and obvious highs and lows, and clearing them can accelerate price. The disputed part is intent — whether a specific reversal was a deliberate "hunt." On a normal chart you cannot see who traded or why, so that layer stays an assumption.
Is SMC liquidity the same as market liquidity?
No. Market liquidity is spread, depth and resiliency — how cheaply a trade fills. SMC "liquidity" means pools of resting orders inferred from chart levels. Same word, two different concepts; a chart can look "liquid" in one sense and thin in the other.
Can I backtest a liquidity-sweep strategy?
Only if you define the pool and the sweep as exact, mechanical rules before you look at the result, then test them over many instances with real costs. Vague post-hoc labels cannot be backtested because they are assigned after the outcome is known.
The stubborn takeaway
The measurable half of liquidity — spread, depth, the slippage on your own fills — you can check on any account, any day. The "smart money grabbed it" half you usually cannot, and a level you can only recognise after the reversal is not a signal, it is a story. If a liquidity idea cannot survive an exact rule and an honest cost model, it was never liquidity you could trade. It was liquidity you could talk about.