Liquidity Sweep vs Support and Resistance: Does a Swept Level Still Hold?

Michael Whitman · Trading since 2017, full-time since 2020 · YouTube creator, 137K subscribers · Published 2026-09-08 · Updated 2026-11-02 · Reviewed 2026-11-01

Support and resistance is a price level; a liquidity sweep is a single session that trades through that level and closes back on the near side. Measured on 839 sweeps on SPY, QQQ and IWM daily charts, a swept level was usually closed through within two weeks, and surviving one sweep did not make it any sturdier.

The two ideas are often taught as old school against new school, as if the sweep had replaced the line on the chart. They are not rivals. One is the thing on the chart and the other is an event that happens to it, so the useful question is not which to believe but what a sweep tells you about the level it touched.

What each one is

Support and resistance is a price where the market has turned before, a prior swing high, a prior swing low or the edge of a range, and anyone can check it on a chart. The support and resistance page covers how levels are drawn.

A liquidity sweep is a session that pokes through such a level and closes back inside, with the reasoning that stop orders and breakout entries were resting just past the line. The liquidity sweep page explains the idea and its vocabulary (buy-side above highs, sell-side below lows).

The level can exist without a sweep; a sweep cannot exist without a level. That one-way dependence is the whole relationship.

Where they differ

Timing is the practical difference. A level can be planned against before price arrives. A sweep is known only at the close of the day that went through, which is after the moment a trader who fades the move wanted to act.

The claims differ in what can be checked. The level claims a past reaction, which the chart shows. The sweep adds a story about resting orders that a candle cannot confirm. Whether a wick or a close past a level is the better tell is the subject of the liquidity sweep vs break of structure comparison; this page asks what happens to the level itself.

How the level and the sweep were defined

The level. A five-bar swing high is a session whose high is strictly above the highs of the five sessions on each side, the same swing rule the break of structure page uses. It is usable from the session after it is confirmed, a newer confirmed swing high replaces it, and a close above it retires it. Swing lows mirror this.

The sweep and the break. A session whose high goes above the live swing high either closes at or below it (a sweep, and the level stays live) or closes above it (a break, and the level is retired). Each sweep is numbered: the first, second or third time that same level was swept.

The outcome. Whether any close in the next ten sessions went beyond the same price, and how many sessions that took. Sweeps without ten sessions of data after them are left out of the percentages.

A worked example: one level swept twice

IWM’s five-bar swing high of $302.72 was set on 1 Jul 2026. On 5 Aug the session reached $303.06, 34 cents above it, and closed at $299.77, so the level survived its first sweep. On 12 Aug IWM reached $303.41 and closed at $302.71, one cent below the line: a second sweep of the same resistance. The next session, 13 Aug, closed at $303.50 and retired the level, six sessions after the first sweep and one after the second.

SPY gave a quicker version in October 2026. Its swing high of $775.14 dated from 22 Sep. On 5 Oct SPY traded to $776.61 and closed at $774.83, 31 cents under the line, and the following session closed above $775.14. A trader who read the 5 Oct close as resistance holding had one session before the level was gone.

A support case ran the same way. SPY’s swing low of $731.53 (19 May 2026) was swept on 9 Jun, when the session traded to $722.59 and closed at $737.05, and SPY closed below $731.53 the very next session. These are three illustrations of how the rule labels a chart, not evidence about sweeps in general; the counts below are.

Which one to use: Liquidity Sweep or Support and Resistance?

Use support and resistance when you are planning before the session. It gives a price for a stop, a target or an alert while there is still time to act, and the sweep framing adds nothing until a close.

Use the sweep only when the session has closed, and read it as information about that day, not as the level proving itself. Across the three funds a first sweep was followed by a close through the same level within ten sessions 78.41% of the time, and a second sweep 78.31%. A level that survived one poke was no more likely to survive the next.

Give any trade based on a sweep a time limit and a price that cancels it. Without both, every eventual reversal can be claimed as a successful sweep and every break explained away.

Keep the sweep idea for levels a crowd can see. A daily swing high qualifies; a line from a private rule has no cluster of orders behind it for the stop-order story to rest on.

Does a swept level get stronger or weaker?

Two opposite beliefs circulate. One says a level that holds a test, sweep included, has proved itself and gets stronger. The other says each test uses up the orders resting there, so a level weakens with every touch. The sweep counts let both be checked on the same rule.

Neither showed up clearly. First, second and later sweeps were closed through at similar rates. The third-or-later group was the highest at 85.71%, which leans toward the weakening story, but it is 84 events, and the gap from first sweeps (78.41%) is within what chance produces (two-sided Fisher exact test, p = 0.15). Second sweeps sat almost exactly on first sweeps.

What this means for a chart: a sweep does not reset a level’s credit in either direction. The level was most often closed through soon after the sweep, whatever number sweep it was.

The original data

Sample one, three funds and the sweep count. Yahoo Finance daily bars to 26 Oct 2026: SPY from 29 Jan 1993 (8,493 sessions), QQQ from 10 Mar 1999 (6,951) and IWM from 26 May 2000 (6,643); flat no-trade bars dropped. Same rule throughout, highs and lows pooled.

Pooling second and later sweeps gives 220 of 273 (80.59%) against 78.41% for first sweeps, p = 0.53. Of the 839 sweeps, 470 were of swing highs and 369 of swing lows; highs were closed through somewhat more often than lows (81.21% against 74.90% for first sweeps), which fits three funds that rose over most of the period. The IWM counts match the IWM sample below event for event (152 high sweeps, 136 low).

Sample two, SPY, how far a sweep goes. SPY daily bars, 29 Jan 1993 to 2 Oct 2026, 8,477 sessions.

So a sweep is mostly a tick through and a close back. With SPY in the $700s, a 0.145% poke is about a dollar, and 70 of 191 swing-high sweeps went no more than 0.10% past the line. A stop placed a few cents past a swing high sat inside the distance half of all sweeps covered. Ten sessions after the first sweep of a swing high, SPY was lower in 43 of 121 cases (35.54%) against 39.23% of all other sessions, p = 0.46: no different from an ordinary ten-session stretch.

Sample three, IWM, how long the swept level lasted. IWM daily bars, 26 May 2000 to 26 Oct 2026.

The check. Nasdaq’s own IWM history, 26 Oct 2016 to 26 Oct 2026, run through the same code gives the identical events, date for date, on both sides. The limits: three US index funds, daily bars, one swing rule and a ten-session window; a different swing width or timeframe could give different counts. Every event is in the sweep-number file, the IWM sweeps file and the SPY swing-level sweeps file.

When it fails

Fading a sweep as proof the level held is the expensive mistake. The short goes on at the close under a swept high with a stop above the day’s extreme, and the most common next event in this data was a close through the level, a third of the time on the very next session.

Counting touches as strength fails too. A level swept twice was closed through as often as one swept once, so a second sweep is not a better entry than the first.

Drawing the level after the move fails quietly. Every turn on the chart then looks like a sweep.

A level only you can see fails the premise. The stop-order explanation needs a crowd.

Treating the stop story as observed fails on evidence. A price chart shows the poke and the close back; the orders are inferred.

Questions people ask

Is a liquidity sweep a reversal signal?

Not on these charts. Across SPY, QQQ and IWM, 78.41% of first sweeps were followed by a close through the same level within ten sessions, so the more common sequel was the break, not a reversal away from the level.

Is a liquidity sweep the same as a stop hunt or a false breakout?

They describe the same bar. A false breakout names the failed close beyond the level; a sweep or stop hunt names the run through it and adds a claim that resting stops were filled. The bar is visible on any chart, and the claim about stops is not.

Does a level get weaker every time it is swept?

Not measurably here. Second sweeps were closed through within ten sessions at 78.31% against 78.41% for first sweeps; third-or-later sweeps came out higher at 85.71%, on 84 events and inside chance.

How far past the level does a sweep usually go?

A little. The median SPY swing-high sweep went 0.145% past the level, against 0.524% for sessions that closed through it.

Related

Liquidity sweep defines the run through a level and the close back, and support and resistance covers how the level is drawn in the first place. The resting-order argument behind sweeps is set out on the liquidity page, and market structure explains the swing sequence these levels come from.

This page is educational, not financial advice. Test every idea on your own charts before risking money.