Liquidity zones indicator: what it shows and how to test it
A liquidity zones indicator marks price levels where stop and limit orders are likely to rest, inferred from past swing highs, swing lows and volume clusters, not from real orders. Its value is measured by the reaction rate at those zones on data the settings never saw, after costs.
By Soren Huang · Published · updated · Editorial policy
Contents
- What does a liquidity zones indicator show?
- How do you draw liquidity zones?
- Can a liquidity indicator see the resting orders?
- How do you turn the zones into explicit rules?
- Can you backtest liquidity zones, and how accurate are they?
- Which indicators pair with liquidity zones in a test?
- Frequently asked questions
- References
What does a liquidity zones indicator show?
A liquidity zones indicator marks price levels where resting stop and limit orders are likely to sit. It infers them from past swing highs, swing lows and high-volume areas, not from any list of real orders. To test it, write the zones as exact rules and count how often price reacts at them on data the settings never saw.
A liquidity zone is a price band where an indicator expects a cluster of orders to be waiting. Think of a fisherman who marks the spots where he caught fish last season. The marks are a reasonable guess about where fish gather, but they are made from memory, not from a sonar reading of the water today. A liquidity zones indicator works the same way: it reads past bars and draws where orders probably sit now.
Most scripts of this type, including the liquidity indicators traders share on TradingView, draw three kinds of zone:
- Zones above swing highs. Traders who sold short near a prior high often place a buy stop just above it. The indicator shades a band above the high and treats it as a pool of buy orders.
- Zones below swing lows. Traders who bought near a prior low often place a sell stop just below it. The band below the low is treated as a pool of sell orders.
- Volume-based zones. A price where an unusually large number of contracts traded is treated as a level where traders hold positions and will defend or exit them. This is the same idea a volume profile uses.
Where these zone tools sit among the wider set of technical analysis tools and indicators you can actually test is simple: they are a level-drawing tool. They give a location. They say nothing about when price will arrive there or what it will do on arrival. The rest of this article is about measuring that.
How do you draw liquidity zones?
A liquidity zone is drawn by a fixed recipe, and the recipe must be written down before you can test it. Two traders using the same chart and the same settings must get the same zones. If a zone depends on where someone chose to look, it cannot be backtested.
A swing high is a bar whose high is higher than the highs of the N bars on each side of it; a swing low is the mirror image. The number N is the swing lookback. The recipe below uses round example settings:
- Choose the swing lookback. Example: 10 bars on each side. A small lookback such as 3 finds many small swings and many zones; a large lookback such as 20 finds few, larger swings. A swing high with lookback 10 is confirmed only after the 10 bars to its right have closed, so the zone appears 10 bars after the high was made. A test that draws the zone at the high itself is reading the future.
- Mark every swing high and swing low that price has not yet traded through. These are the untested levels. A level that price already crossed has, by this logic, had its orders filled.
- Add the bands where volume clustered. For a session or a lookback window, find the price where the most contracts traded and shade a band around it. The volume profile indicator is the usual way to find this price.
- Give every zone a width in ticks. Example: 8 ticks. Without a width, a touch is undefined, because price almost never returns to one exact tick. Width also sets how many touches you will count: wider zones mean more touches and weaker meaning per touch.
- Mark the zone swept when a bar trades through it. A sweep is a move through the zone that, under the indicator's logic, has triggered the stops resting there. Decide once whether a swept zone is deleted or kept as a reclaimed level, and keep that choice fixed for the whole test.
Steps 1 and 4 are the two settings that change the result most. Record both before the first run, and when you later compare results, compare only runs that share them.
Can a liquidity indicator see the resting orders?
No. The indicator has no view of the order book, so it guesses where stops might be from the shape of past prices. The real resting quantity at each price is visible only in the exchange's depth of market, and it changes every second.
The order book, also called the depth of market, is the exchange's live list of limit orders waiting to buy or sell at each price. A liquidity zones indicator never reads it, for two reasons. First, bar data stores only open, high, low, close and volume; the book is not in the file. Second, even a live book would not show the orders the indicator cares about most. A stop order is not a resting limit order. It waits on a server and only becomes a market or limit order when price reaches its trigger, as the SEC's investor bulletin on stop orders explains. Until that moment it is not displayed anywhere a trader can see.
| Question | Liquidity zones indicator | Live depth of market |
|---|---|---|
| What it reads | Past highs, lows and volume from bar data | Current resting limit orders at each price |
| Shows stop orders? | No, it infers where they might be | No, stops are not displayed until triggered |
| Available in a backtest? | Yes, from any bar file | Only if you recorded the book yourself, tick by tick |
| How fast it changes | Once per bar | Many times per second |
| What it is good for | A repeatable location to test | Reading current supply and demand while trading |
This is not a weakness to fix; it is the nature of the tool. A zone is a statement about price shape, and price shape is testable from bars. If you want to work with actual order flow, the data requirements are different and are covered in order flow indicators: delta, footprint and what a backtest needs.
How do you turn the zones into explicit rules?
A zone becomes testable only when sweep depth, close-back condition, wait time, entry bar, stop distance and exit are each written as a number, so the same chart always produces the same trades. Until then, "price swept the zone and reclaimed it" is a story told after the fact, and stories cannot be counted.
Lo, Mamaysky and Wang (2000) showed that visual chart patterns can be turned into algorithms and tested like any other rule, and that the step from picture to algorithm is where the real decisions are made. The same applies here. The most common zone trade is the sweep and reclaim: price pushes through a zone below a swing low, triggers the stops there, then closes back above it, and the trader buys. Written as numbers, it has six parts:
| Part of the rule | What it fixes | Example value |
|---|---|---|
| Sweep depth | How far below the zone a bar must trade for the sweep to count | At least 4 ticks |
| Close-back condition | Which bar must close back inside or above the zone | Within 2 bars of the sweep bar |
| Wait time | How many bars after the sweep the setup stays valid | 2 bars, then cancelled |
| Entry bar | When the order is placed | Market order at the open of the next bar |
| Stop distance | Where the protective stop sits | 6 ticks below the sweep low |
| Exit | When a trade that is neither stopped nor reversed is closed | After 12 bars |
Two of these parts hide the most common mistakes. The entry bar must be the bar after the signal closes, because the close-back condition is only known at the close. And the swing lookback from the drawing recipe must be applied with its delay, so the zone exists in the test only from the bar on which a live trader could have seen it. A rule that enters on the signal bar's close, or that uses a zone before it was confirmed, will test well and trade badly. In AlphaPrime (Valynox) this is the point where a zone idea enters strategy generation: the entry and exit conditions and the parameter ranges are written as rules before any bar is read, so the trade list never depends on who is looking at the chart.
Can you backtest liquidity zones, and how accurate are they?
Yes, you can backtest liquidity zones once they are drawn by the recipe above, and their accuracy is the reaction rate measured on data the settings never saw, set against a chance rate of about half. Judge the rule by its out-of-sample trades after commission and slippage, not by how the zones look on a chart.
The reaction rate is the share of zone touches after which price moved away from the zone by a set distance before moving through it by the same distance. The one formula you need:
reaction rate = reactions ÷ touches
Here a touch is any bar whose range enters the zone, and a reaction is a touch followed by a move away of, say, 8 ticks before a move through of 8 ticks. Because the two distances are equal, a market with no memory would react about half the time, so a rate near 50 percent means the zone adds nothing. The rate on the data you used to choose the lookback and width will be higher than the rate on fresh data, because you picked the settings that looked best. Bailey, Borwein, López de Prado and Zhu (2014) call this gap the selection bonus, and it grows with the number of settings you tried. The split that exposes it is explained in in-sample vs out-of-sample: how to split and how much is enough.
Hypothetical example
You try six lookbacks and four widths on the first four years of ES 15-minute bars and keep the pair with the highest reaction rate: 58 percent from 120 touches. You then run that pair unchanged on the next two years and get 51 percent from 60 touches. The 7-point drop is the selection bonus leaving. At 51 percent on 60 touches the zone is indistinguishable from chance, and the rule should be set aside or redesigned, not re-tuned on the second period.
Level-based rules have been tested this way since Brock, Lakonishok and LeBaron (1992) put trading-range breakouts through a formal test, and the lesson has not changed: a level that is touched often enough to measure, and that reacts clearly more than half the time on unseen data, is rare. Reaction rate is only the first check. The second is the trade list after costs, because a zone that reacts 55 percent of the time can still lose money if the average winner is smaller than the average loser plus commission and slippage.
To show what the second check looks like on a real report, the figure below comes from an AlphaPrime run. It did not test a liquidity zone rule. It tested a stand-in built from a standard indicator: a Rate Of Change rule as the only entry and exit, no stop, no target, no filters, on ES daily bars from 2019-01-02 to 2025-06-30, long and short, one contract, with $20 commission and $25 slippage charged on every trade. The search tried 1,024 candidates (count from the run dated 2026-10-31) and kept one for its return against its drawdown. Picking the kept candidate after the fact flatters it; its numbers describe that one survivor of the search, not what the rule or the indicator earned or what a reader can expect.
Source: AlphaPrime product engine (CPU), run 2026-10-31. ES daily bars, 2019-01-02 to 2025-06-30, long and short, 1 contract, $20 commission and $25 slippage per trade.
| Number | Result |
|---|---|
| Net profit (closed trades) | $68,585 |
| Max drawdown | $64,677.5 |
| Trades | 47 |
| Win rate | 57.45% |
| Profit factor | 1.638 |
| Average trade | $1,459.26 |
| Time in market | 94.01% |
Three things on this report are worth reading the same way for a zone rule. The max drawdown of $64,677.5 is almost as large as the net profit of $68,585, so the path to the final number was rough, and the out-of-sample stretch is where you check whether the slope held. The 47 trades are a small sample; a zone rule on intraday bars will usually produce far more, which is the one advantage it has here. And the 94.01 percent time in market is the opposite of a zone rule, which is flat most of the time and trades only at its levels, so its drawdown shape will look different and its costs per trade matter more.
Which indicators pair with liquidity zones in a test?
A zone gives a location but no timing, so test one timing filter at a time and keep it only if the out-of-sample average trade improves after costs. Adding two filters at once makes it impossible to know which one helped, and every added filter is one more setting that can be tuned to the past.
The test design is the same for any filter. Run the zone-only rule first and record its out-of-sample average trade and trade count. Then add one filter, rerun on the same out-of-sample period, and compare. A filter that cuts the trade count by half and raises the average trade by a fraction has probably removed good trades along with bad ones; the two numbers must be read together. The comparison of indicators in what testing actually shows about the best indicator for intraday trading uses the same method.
| No filter (zone only) | DEMA slope filter | WaveTrend cross filter | |
|---|---|---|---|
| What it adds | Nothing: every sweep and reclaim is traded | Trade only in the direction of the DEMA slope | Enter only after a WaveTrend cross in the trade direction |
| Extra parameter | None | DEMA length | Channel length and average length |
| What to measure | Out-of-sample average trade after costs, as the baseline | Out-of-sample average trade after costs | Out-of-sample average trade after costs and the trade count |
The two filters in the table are common pairings because each answers a different question. A trend filter such as the DEMA, which cuts EMA lag and shows what a crossover test gives, asks whether the larger move agrees with the trade direction, so a sweep below a swing low is bought only when the slope is already up. A momentum filter such as the WaveTrend indicator with its settings and cross signals asks whether the turn has begun, so the entry waits for the oscillator to cross rather than for the close-back bar alone. The first filter tends to remove trades against the trend; the second tends to delay entries by a few bars. Which one, if either, improves the out-of-sample average trade is an empirical question for your market and bar size.
What to do next is specific. Take the six rule parts from the fourth section with their example values, fix a lookback of 10 and a width of 8 ticks, and run the zone-only rule on the first 70 percent of your bar data. Write down the out-of-sample reaction rate and the out-of-sample average trade after commission and slippage from the last 30 percent. Only then add the DEMA slope filter, rerun on the same 30 percent, and compare the two average trades. That comparison, not the chart, tells you whether the zones are worth keeping.
Frequently asked questions
Are liquidity indicators on TradingView accurate?
Accuracy has to be measured, and the honest measure is the reaction rate at the zones on data the settings never saw, compared with a chance rate of about 50 percent. A zone that reacts about half the time on fresh data adds nothing, however clean it looks on the chart. Most scripts of this type publish no such number, so you have to run the test yourself with the drawing recipe and the six rule parts written down first.
Does a liquidity zones indicator repaint?
Many do, because a swing high is only confirmed after the lookback bars to its right have closed. A script that draws the zone at the high itself is placing it earlier than a live trader could have seen it. For a fair test, the zone must appear only from the bar on which it was confirmed, and the entry must be on the bar after the signal closes.
What is a liquidity sweep?
A liquidity sweep is a move through a zone that, under the indicator's logic, triggers the stop orders resting just beyond a swing high or low. In a rule it is defined by a sweep depth in ticks and, for a reclaim trade, by a close back inside the zone within a set number of bars. Without those numbers the word describes a story told after the fact, not a condition a test can count.
Is a liquidity zone the same as support and resistance?
They overlap but are drawn from different reasoning. Support and resistance are levels where price has turned before. A liquidity zone is a band just beyond such a level, where the indicator expects stop orders to rest, so the expected behaviour is a brief push through the level followed by a return. In a test the difference shows up in the rule: a support rule buys at the level, a liquidity rule buys after the level is swept and reclaimed.
How many zone touches do I need before trusting a reaction rate?
Enough that a few touches cannot move the rate much. Sixty out-of-sample touches at 51 percent could easily be 45 or 57 percent with a different stretch of data, so a rate like that is a hint rather than a result. Several hundred out-of-sample touches is a more useful target, which on intraday bars usually means two or more years of unseen data or a walk-forward design that produces more out-of-sample trades from the same history.
References
- Lo, A. W., Mamaysky, H., & Wang, J. (2000). Foundations of Technical Analysis: Computational Algorithms, Statistical Inference, and Empirical Implementation. The Journal of Finance, 55(4), 1705–1765.
- Brock, W., Lakonishok, J., & LeBaron, B. (1992). Simple Technical Trading Rules and the Stochastic Properties of Stock Returns. The Journal of Finance, 47(5), 1731–1764.
- Bailey, D. H., Borwein, J., López de Prado, M., & Zhu, Q. J. (2014). Pseudo-Mathematics and Financial Charlatanism: The Effects of Backtest Overfitting on Out-of-Sample Performance. Notices of the American Mathematical Society, 61(5), 458–471.
- U.S. Securities and Exchange Commission. Stop, Stop-Limit, and Trailing Stop Orders (Investor Bulletin). https://www.investor.gov/introduction-investing/general-resources/news-alerts/alerts-bulletins/investor-bulletins-15

