The vocabulary is worth learning and the popular trade built on it points the wrong way. Sessions that take yesterday's low first close down about seven times in ten — the opposite of what sweep-and-reverse predicts.
Buy-side liquidity is the cluster of buy orders resting above obvious highs — mostly stop losses on short positions — and sell-side liquidity is the mirror image below the lows.
The names are backwards from what most people first assume, so it is worth being precise.
Buy-side liquidity sits above the highs. Traders who are short have their stop-loss orders up there, and a stop on a short is a buy order. So above every obvious high there is a pool of buying waiting to be triggered.
Sell-side liquidity sits below the lows. Traders who are long have their stops down there, and a stop on a long is a sell order.
The naming describes what the resting orders do, not who wants price to go there. That is the whole trick, and once it clicks the rest of the vocabulary follows.
A large fish wants to eat, but it can only eat where the small fish are shoaling. It does not swim into open water — it swims to where the shoal is, because that is the only place a big meal exists.
Obvious highs and lows are the shoals. A large participant who needs to fill a big order goes where the resting orders are, because nowhere else can absorb the size. That is a genuinely useful mental model. What it does not tell you is what happens next — and next is what this page measures.
The popular version goes: price takes the sell-side liquidity below the prior low, the sellers who were induced in get trapped, and then price reverses upward for the rest of the session.
That is a clear prediction. If a session takes yesterday's low before it takes yesterday's high, it should close up more often than not.
| Instrument | Took prior low first | …then closed UP | Took prior high first | …then closed DOWN | Took neither |
|---|---|---|---|---|---|
| XAUUSD | 558 | 31.4% | 659 | 28.5% | 15.3% |
| GBPUSD | 605 | 30.1% | 607 | 33.3% | 13.7% |
| EURUSD | 631 | 26.9% | 588 | 31.3% | 13.7% |
| US30 | 432 | 37.3% | 491 | 32.4% | 11% |
| NAS100 | 409 | 34.5% | 529 | 32.9% | 11.1% |
What was counted: For each session, the hourly candles are walked in order until one trades beyond the previous session's high or its low. Whichever side is reached first is recorded, and the session's close is compared with its open. Hours reaching both sides are discarded as undetermined.
The prediction fails, and it fails hard. On gold, sessions that took the prior low first closed up only 31.4% of the time — meaning they closed down nearly 69% of the time. Sessions that took the prior high first closed down only 28.5%, so they closed up about 72%.
Every instrument agrees. Whichever side gets taken first, the session usually keeps going that way.
Most negative results in this series say "there is nothing here". This one says something stronger: the first side taken predicts continuation, not reversal, roughly seven times in ten.
Which means a trader who systematically buys after the prior low is swept — the textbook sweep-and-reverse play — has been taking the wrong side of a fairly lopsided tendency. That is worse than trading a coin flip.
None of this means sweeps do not reverse. They plainly do, and the liquidity sweep lesson measures how often a level is taken and given back within the same session. But taking out yesterday's low is much more often the start of a move down than the end of one, and the two ideas get run together constantly.
Grace has learned that price sweeps sell-side liquidity before rallying. Yesterday's gold low was 4,392. This morning price trades to 4,388 and she buys, expecting the reversal.
What the data says about that trade. Sessions on gold that took the prior low first went on to close up only 31.4% of the time. She is taking a position that loses on roughly 69% of comparable days, and the story she learned told her it was the high-probability side.
Where the confusion comes from. She is mixing two timeframes. Intraday, a poke below a level that closes back above it within the hour genuinely does often reverse — that is the sweep. But a session that trades through yesterday's low is usually a session that is going down, and "yesterday's low was taken" is not the same event as "a wick was rejected".
What she does instead. She waits for two things rather than one: price closing back above 4,392, and then a break of structure upward on the hourly. If both happen, she has the reversal with evidence rather than with faith. If they do not, she has avoided the wrong side of a seven-in-ten tendency.
The other use she takes up. Next time she is long from lower down, she marks the buy-side pool above yesterday's high as a target — which is the part of this concept the data actually supports.
If you have been buying swept lows, the table above says you have been on the wrong side of a lopsided tendency. Replay three months, mark yesterday's high and low each session, note which went first and where the session closed. Two hundred rows will settle it for your instrument.
Check it in the free backtester →Run the drill in the free backtester → Free, no sign-in to begin.
What is buy-side liquidity?
The pool of buy orders resting above an obvious high — mostly stop losses belonging to traders who are short, since a stop on a short is a buy order. Sell-side liquidity is the mirror: sell orders resting below the lows. The naming describes what the orders do, not who wants price to go there.
Does price reverse after taking liquidity?
Usually not, at the session level. On gold, sessions that traded through the prior low before the prior high closed up only 31.4% of the time — so they closed down nearly 69% of the time. Every instrument tested shows the same. The first side taken predicts continuation.
But I have seen sweeps reverse constantly.
You have, and they do — intraday. A wick that pokes below a level and closes back above it within the hour is a different event from a session that trades through yesterday's low and keeps going. The liquidity sweep lesson measures the first; this page measures the second. Running them together is the most common error with this vocabulary.
So what is liquidity good for?
Targets and stop placement. If you are already long, the pool above the prior high is a reasonable place to expect price to be drawn. And knowing where the pools are tells you where not to hide your own stop — which is the most valuable practical output of the whole concept.
How do I trade a reversal from a sweep then?
Demand more than the sweep. Require price to close back inside the level and then produce a change of character in the new direction. The sweep on its own leans against you; the sequence is what separates the reversal cases from the continuation ones.
Why do people get the names the wrong way round?
Because "buy-side" sounds like it should be where buyers are, which would be below. It is above, because the orders sitting there are buy orders — stop losses on short positions. Once you think of it as "what the resting orders do", it stops being confusing.
Mark yesterday's high and low each session, record which went first and where the session closed, and total two hundred rows. That table will tell you whether the trade you have been taking is the one the data supports.
Open the free backtester →Beginner exploration
Open each answer for a plain-language way to read Buy-Side and Sell-Side Liquidity, test it carefully and decide what to explore next.
This page focuses on “Buy-Side and Sell-Side Liquidity”.What buy-side and sell-side liquidity mean, and the sweep-then-reverse claim tested across 6,370 sessions. Taking the prior low first led to an up close only 31.4% of the time — the tendency runs the other way.For “Buy-Side and Sell-Side Liquidity”, a beginner should identify what the learning guide measures, assumes or teaches before acting on its conclusion.Treat this page's account of “Buy-Side and Sell-Side Liquidity” as a learning reference rather than a prediction, signal or promise of future performance.
For “Buy-Side and Sell-Side Liquidity”, translate the idea into a definition you could apply the same way on two different charts.While exploring “Buy-Side and Sell-Side Liquidity”, work through one example slowly and record which inputs or observations determined the result.Keep your “Buy-Side and Sell-Side Liquidity” record honest: list the limitation or counterexample before using the concept in a trading plan.Before leaving “Buy-Side and Sell-Side Liquidity”, practise the definition on unseen history and review consistency before judging performance.
Turn one idea from “Buy-Side and Sell-Side Liquidity” into a rule with explicit inputs, dates, costs and pass-or-fail conditions.Ask AI to expose missing assumptions in that “Buy-Side and Sell-Side Liquidity” test, not to guess the next market move.Use the FXAbsolute AI Backtesting Lab to inspect calculations connected to “Buy-Side and Sell-Side Liquidity” and the assumptions behind them.Reproduce any important “Buy-Side and Sell-Side Liquidity” result and reserve unseen data before deciding that an apparent pattern is useful.
Continue your exploration of Buy-Side and Sell-Side Liquidity with the beginner AI prompt guide, or inspect public calculations in the AI Backtesting Lab.