Where resting orders cluster

Here, liquidity means resting orders, mostly stop losses and breakout entries. They gather where many traders draw the same line, because that is where most of them put the stop. Traders call the orders above highs buy-side liquidity and the orders below lows sell-side liquidity.

There is evidence for the clustering. A study of a large currency dealing bank’s order book by Carol Osler of the Federal Reserve Bank of New York (2001) found that stop-loss buy orders clustered just above round numbers and stop-loss sell orders just below them. That was currency data from one bank, but the logic carries over to any market where traders watch the same levels. The usual places:

  • Equal highs and equal lows: two or more swings at almost the same price look like strong resistance or support, so stops pile up just beyond them.
  • The prior day’s high and low, and the high and low of the pre-market session.
  • Obvious swing points on the chart you trade, especially the last swing before a strong move.
  • Round numbers, such as $50.00 or $100.00.
  • The opening range: the high and low of the first 15 or 30 minutes.

Turn your phone sideways to read the image. Tap to close.

An intraday chart with five levels. Above price: the prior day high, equal highs and the round number 50.00, each with small x marks just above it, and the note buy stops above. Below price: a swing low and the prior day low, each with x marks just below it, and the note sell stops below.

Why sweeps happen, without the myths

A stop order becomes a market order once its price trades. When price reaches a cluster of sell stops under a low, those orders hit the market together and push price lower for a moment. On the other side sit resting buy orders: traders who wanted to buy at that price, market makers, and larger buyers who need sellers to fill a big order. When the burst of stop orders is used up and no fresh sellers arrive, the selling stops and price can snap back above the level.

That is all a sweep needs. It does not require an institution aiming at your stop, and a chart cannot show intent. Large orders do need someone on the other side, and stops near obvious lows do provide it, but the same wick can come from a headline, a fast market or plain noise. Treat “stop hunt” as a description of where orders sat, not a story about who moved price. The same mechanics sit behind the spring below a trading range in the Wyckoff method.

Sweep vs real breakout: what happens next

At the moment price trades through the level, a sweep and a breakdown look identical. The difference shows in the next few candles. Decide your reclaim rule in advance, for example a close back above the level within two candles; without it, every wick becomes a sweep after the fact. A sweep is a false breakout seen from the order book, and the same filters apply.

  • Sweep: price closes back above the level on the same bar or one of the next two, the reclaim. Breakdown: it closes below the level and stays there.
  • Sweep: displacement follows, fast, large candles in the other direction. Breakdown: the bounce is weak and overlapping.
  • Sweep: the move closes above the last lower high, a change of character in market structure terms. Breakdown: price retests the broken level from below and turns down again.
  • Sweep: the extreme is not revisited. Breakdown: lower lows keep coming.

Turn your phone sideways to read the image. Tap to close.

Two panels with the same dashed level under a few candles. Left, sweep: a candle wicks below the level and closes back above it, circled, then large green candles close above a dashed red line at the last lower high, labeled CHoCH, with the note closes back above, displacement, CHoCH. Right, real breakdown: candles close below the level, a bounce back up to it from below fails, circled, and price keeps falling, with the note closes below, the retest from below fails.

How to trade a liquidity sweep: the reclaim and the entry

The sweep tells you where to look; the entry comes after. There are two common entries, both shown in the worked example below. Entry A buys the reclaim close: earliest and cheapest, with the least proof. Entry B waits for the change of character and buys the first pullback: later and more expensive, with more proof.

  1. Mark the pools before the session: equal highs and lows, the prior day’s high and low, round numbers. Check the higher-timeframe direction.
  2. Wait for price to trade through a pool and close back inside: the reclaim. Entry A buys here.
  3. For entry B, wait for displacement to close above the last lower high, then buy the first pullback that holds above the swept level, on a candle that closes back up. An order block or a fair value gap left by the displacement can mark where that pullback may stop.
  4. Stop: beyond the sweep extreme with a buffer, never exactly at the wick, where the next round of stops will sit.
  5. Target: the next pool on the other side, such as equal highs or the prior day’s high. Check the risk-reward ratio before you enter, and size from the stop.

Worked example: a long after a sweep of the prior day’s low (hypothetical)

Hypothetical, on the 5-minute chart of XYZ, whose 1-hour trend is up. The prior day’s low is $71.40, the morning drop has left a last lower high at $72.10, and equal highs at $74.10 sit above.

At 10:05 a.m. ET a candle trades to $71.10, under the prior day’s low, and closes at $71.52, back above it: the sweep and the reclaim. Two green candles follow, and the second closes at $72.30, above the $72.10 lower high: displacement and a change of character. Price pulls back to $71.80, holds above $71.40, and a candle closes back up at $72.00: entry B.

Entry A would have bought the reclaim close at $71.52 with the same $71.00 stop: $0.52 of risk for $2.48 to the target, about 4.8R, but before any change of character. With $100 of risk that is 192 shares. Entry B pays more for the extra proof.

The failure, same setup on another day: after your entry, price rolls over, slices through $71.40 and the $71.10 sweep low, and the stop fills at $71.00 or lower in a fast market: −1R, about $100 plus slippage. The sweep was the start of a real breakdown. Do not rename it; log it.

  1. Entry B: $72.00, on the close back up after the change of character.
  2. Stop: $71.00, $0.10 under the $71.10 sweep low. Risk: $1.00 a share.
  3. Target: $74.00, just under the $74.10 equal highs. Reward: $2.00, or 2R.
  4. Size: a $20,000 account risking 0.5% accepts a $100 loss. $100 ÷ $1.00 = 100 shares, about $7,200 of stock.
  5. Result: at the target, 100 × $2.00 = $200, or +2R; at the stop, −$100, or −1R, plus costs.

Turn your phone sideways to read the image. Tap to close.

The worked example on a 5-minute chart: the prior day low (PDL) at 71.40, a sweep candle with a wick to 71.10 that closes back above at 71.52, marked A, a close above the 72.10 lower high labeled CHoCH, and a pullback with a close at 72.00, marked B. Dashed lines mark the stop at 71.00, the target at 74.00 and the equal highs at 74.10. A card compares the entries: A, 71.52 − 71.00 = 0.52 of risk and 74.00 − 71.52 = 2.48 of reward, 4.77R; B, 72.00 − 71.00 = 1.00 and 74.00 − 72.00 = 2.00, 2R.

Common liquidity sweep mistakes

Most losses on sweeps come from a handful of habits:

  • Buying the wick: catching price while it is still below the level, before any reclaim.
  • A stop exactly at the sweep low, where the next sweep will look for orders.
  • Trading every sweep in the first minutes after the open, when both the early high and the early low often get taken out.
  • Fighting the higher timeframe: a sweep of a low in a daily downtrend is more often a pause than a turn.
  • Relabeling in hindsight: if every failed sweep becomes “the real one is lower”, the method predicts nothing; a trading journal keeps score.