Volume profile explained: POC, value area and what you can test
A volume profile groups a period's traded volume by price instead of by time. Its point of control, value area and volume nodes can be tested, but only when the profile is frozen at the end of the previous session and the data carries volume per bar.
By Soren Huang · Published · updated · Editorial policy
Contents
- What is volume profile, and can its levels be backtested?
- Volume at price vs volume over time: what changes?
- What is the point of control (POC)?
- How is the value area calculated?
- What are HVN and LVN, and how do traders use the levels?
- What data do you need to build a volume profile?
- Can volume profile be backtested? A POC-based rule and an ES test
- Frequently asked questions
- References
What is volume profile, and can its levels be backtested?
Volume profile is a histogram of traded volume at each price level over a chosen period. Its levels, the point of control, the value area, high-volume nodes and low-volume nodes, can be backtested, but only when each level is built from bars that closed before the entry bar and your data carries volume per bar.
Think of a shop that records every sale with the shelf it came from. At the end of the day the manager can stack the receipts by shelf and see which shelf was busiest. A volume profile does the same with trades: every contract traded is stacked on the price where it traded, and the result is a sideways histogram next to the price axis. The volume profile indicator page covers the display settings; this article covers what you can test with it.
Most volume profile backtests that look good fail for one reason, and it is not that the levels are useless. It is that the profile on the chart includes the bar you are trading on, so the rule reads volume that had not traded yet. Fix that, and the levels become something you can test like any other rule.
Volume at price vs volume over time: what changes?
A normal volume bar answers "how much traded in this time slot", while a volume profile answers "how much traded at this price", so the same session gives two different pictures and only the second one gives price levels you can trade against. Volume over time tells you when the market was busy. Volume at price tells you where it was busy.
Both pictures use the same raw numbers. In the first, each trade is added to the bar of the minute it happened. In the second, each trade is added to the row of the price it happened at. A session that opens with a fast move and then spends four hours in a narrow range looks like a tall spike followed by small bars in the first picture. In the second picture the fast move leaves a thin area and the narrow range builds one thick cluster. That thick cluster is a price level; the tall spike is a time of day. A trading rule can be written against a price, which is why the profile view is the one that produces testable levels. The guide to technical analysis tools and indicators sorts other chart tools by the same question: does it give you a number a rule can read?
What is the point of control (POC)?
The point of control is the single price level with the most traded volume in the profile period, and its meaning depends on the period chosen, so a rule must state which POC it uses. In the shop comparison, the POC is the busiest shelf. On the chart it is the longest row of the histogram.
The period changes the answer. A session POC is built from one trading session and tells you where that session's trading concentrated. A weekly or composite POC is built from several sessions stacked together and often sits at a different price, because one busy day can dominate the week. Neither is "the" POC. A rule that says "buy a pullback to the POC" is not testable until it says "the POC of the previous completed session" or "the POC of the previous five sessions". Write the period into the rule before you build the profile.
One more detail matters for testing: the row height. A profile built with rows of one tick looks jagged, and the POC can jump between neighbouring ticks from day to day. A profile with rows of several ticks is smoother and its POC is more stable, but less precise. Pick one row height and keep it for the whole test, so the POC series means the same thing on every day of the history.
How is the value area calculated?
The value area is the price range around the POC that holds about 70 percent of the period's volume, built by adding rows outward from the POC until the target is reached. The 70 percent figure is a convention that traders adopted because it is close to the share of data within one standard deviation of a normal distribution, and it is now the default on most charting platforms. Its top row is called the value area high (VAH) and its bottom row the value area low (VAL).
The one formula you need:
value area target = 0.70 × total volume of the period
Here total volume of the period is the sum of all rows in the profile, and the target is the amount of volume the value area must enclose before you stop adding rows. The steps are:
- Find the POC row and start the running total with its volume.
- Look at the two rows touching the current range, one above and one below.
- Add the larger of the two to the range and to the running total.
- Repeat until the running total reaches the target. The highest row added is the VAH, the lowest is the VAL.
Hypothetical example
A profile has seven rows, from the top down: 30, 80, 200, 300, 220, 100 and 20 contracts, for a total of 950 contracts. The target is 0.70 × 950 = 665. Start at the POC row of 300. The neighbours are 200 above and 220 below; add 220, running total 520. The neighbours are now 200 above and 100 below; add 200, running total 720. That is above 665, so stop. The value area is the three rows holding 200, 300 and 220 contracts. The 200 row is the VAH and the 220 row is the VAL.
Some platforms add two rows at a time instead of one, and some compare pairs of rows. The results differ by a tick or two at the edges, which is why a value area copied from one chart rarely matches another exactly. The same 70 percent idea is used in market profile, where the rows count time rather than contracts; the market profile guide to the value area and POC explains how the two methods differ.
What are HVN and LVN, and how do traders use the levels?
High-volume nodes are prices where many contracts traded and tend to be used as support, resistance or mean-reversion targets, while low-volume nodes are thin prices that traders expect price to move through quickly, so each node type pairs with a different kind of rule. A high-volume node (HVN) is a thick cluster of rows in the profile, a price where trading paused and both sides did business. A low-volume node (LVN) is a thin gap between clusters, a price the market passed through without stopping.
The reasoning behind each use is about memory. At an HVN many positions were opened, so when price returns there, traders expect those positions to be defended or closed, which slows the move. At an LVN few positions exist, so there is little reason for price to pause there. Whether that reasoning holds on your market and bar size is exactly what a backtest is for. The two node types call for different rule shapes:
| Question | High-volume node (HVN) | Low-volume node (LVN) |
|---|---|---|
| Shape in the profile | Thick cluster of rows | Thin gap between clusters |
| What traders expect when price returns | Slow down, stall, or turn back | Move through quickly to the next HVN |
| Rule type that fits | Mean reversion: fade a move into the node, target the POC | Breakout: enter when price leaves the LVN, target the next HVN |
| Natural stop | Beyond the far side of the node | Back inside the LVN |
| What to check in the test | Does the average trade survive slippage at a crowded level? | How often does the fast move reverse before the next HVN? |
Many traders add a filter to the node rule so that it does not fire on every touch. A common pairing for the HVN fade is an oscillator reading at the same time, for example the KD stochastic indicator reading at the moment price reaches the node, so that the rule only fades a move that the oscillator also calls stretched. Levels from other sources are used the same way: gamma exposure levels built from option positioning are another map of prices where traders expect price to slow down, and the test questions for them are the same as for an HVN.
What data do you need to build a volume profile?
Tick data gives the exact profile, 1-minute bars give a close approximation when their volume is assigned to the bar's price range, and daily bars cannot build an intraday profile at all, so the bar size you own decides which profile rules you can test. Volume per bar is the first thing to check: a data file with open, high, low and close but no volume column cannot build any profile.
The difference between the three data types is where each contract is placed on the price axis:
| Question | Tick data | 1-minute bars | Daily bars |
|---|---|---|---|
| How volume is placed on price | At the exact trade price | Spread across the bar's range, or placed at one price such as the close | One number for the whole day |
| Profile accuracy | Exact | Close approximation; edges of the value area can differ by a few ticks | None for an intraday profile |
| Rules you can test | All | Session and composite levels | Only multi-day composites built from daily volume |
| Typical file size | Largest | Medium | Smallest |
With 1-minute bars you have to decide how to spread each bar's volume across its range. Dividing it equally over every tick between the low and the high is the most common choice. Placing all of it at the close is simpler but shifts volume toward the end of each bar's move. Either method is acceptable for a test as long as it is the same on every bar of the history, because the levels only need to be consistent, not perfect. When your rule needs exact node edges on a thin market, tick data backtesting for futures explains what changes when you move below bars.
Can volume profile be backtested? A POC-based rule and an ES test
Yes, if the profile is frozen at the end of the previous session and the rule reads only that frozen POC and value area. The profile drawn on a live chart grows all day: every new bar adds volume and can move the POC. A backtest that reads today's POC during today's session is reading a level that did not exist yet, which is look-ahead bias. Look-ahead bias is any use of information in a backtest that was not available at the moment the trade was taken, and the one-bar look-ahead check catches most cases of it.
A testable POC rule therefore works in two steps across two sessions:
- Session 1 builds the profile. At the session close, record the POC, VAH and VAL and lock them.
- Session 2 trades the locked levels only. For example: buy when a bar closes back inside the value area after opening below the VAL, exit at the frozen POC or at the session close. Sell the mirror image above the VAH.
- At the close of session 2, rebuild. The new profile becomes the locked set for session 3.
The run behind this article shows what such a level test looks like on a report, with one honest limit. It used ES daily bars from 2019-01-02 to 2025-06-30, and daily bars cannot build an intraday profile, as the previous section explained. So the run uses a stand-in rule: a Highest and Lowest rule built from standard indicators, which enters and exits on the highest high and lowest low of a lookback, with no stop, no target and no filters. It stands in for a level rule; it is not a volume profile rule. The box shows the one candidate the search kept out of 1,024 it tried, chosen for its return against its drawdown.
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) | $110,185 |
| Max drawdown | $64,017.5 |
| Trades | 92 |
| Win rate | 71.74% |
| Profit factor | 1.886 |
| Average trade | $1,197.66 |
| Time in market | 89.09% |
Read the numbers the way you would read your own. The net profit of $110,185 sits against a max drawdown of $64,017.5, so the account needed to hold more than half of its total gain in open losses at the worst point. The 92 trades over six and a half years are a small sample for a per-trade figure. And the candidate was picked after the fact, as the best of 1,024 for its return against its drawdown; that selection flatters it. Bailey, Borwein, López de Prado and Zhu (2014) show that the best backtest in a large search is expected to look good even when no rule in the search has an edge, and Harvey and Liu (2015) propose discounting a result for the number of tests behind it. Nothing in the box says the Highest and Lowest rule earned that money or that a reader would. AlphaPrime (Valynox) charged the $20 commission and $25 slippage inside each of the 1,024 evaluations and filled every order on the next bar, which is the backtesting setting a level rule needs most, because a fill at the level itself on the bar that touched it is the second most common way a profile test fools its author.

Park and Irwin (2007) surveyed decades of studies on technical rules and found that the results that survive are the ones tested with transaction costs and without choosing the rule after seeing the data. A volume profile rule is no different. Your next step: open your own data and confirm it has volume per bar at 1-minute or finer resolution. Then build yesterday's POC, VAH and VAL for ten recent sessions by hand, using the steps in the value area section, and check that none of the ten levels moves when you add today's bars. If they stay fixed, you have a profile you can test.
Frequently asked questions
What is the POC in trading?
The POC, or point of control, is the price level where the most volume traded during the period the profile covers. On the chart it is the longest row of the sideways histogram. A session POC and a weekly POC can sit at different prices, so any rule that uses the POC must say which period it is built from.
Is volume profile the same as market profile?
No. Volume profile counts contracts traded at each price. Market profile counts how many time periods, usually 30-minute blocks, touched each price, so it measures time at price rather than volume at price. Both produce a POC and a value area, and the two can sit at different levels on the same day.
Can a volume profile backtest use the current session's profile?
Not for entries inside that session. The current session's profile changes with every new bar, so a rule that reads it during the session is reading volume that had not yet traded at the time of the trade. Freeze the profile at the previous session's close and read only the frozen POC, VAH and VAL during the current session.
Does a volume profile repaint?
The live profile of an open period changes as new volume arrives, which is why its POC can move during the day. A profile of a completed period does not change. For testing, that means a rule should read only profiles of completed periods, and the profile period and row height should stay the same on every day of the history.
References
- 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.
- Harvey, C. R., & Liu, Y. (2015). Backtesting. The Journal of Portfolio Management, 42(1), 13–28.
- Park, C.-H., & Irwin, S. H. (2007). What Do We Know About the Profitability of Technical Analysis? Journal of Economic Surveys, 21(4), 786–826.

