Direct Answer
Average volume is the mean number of shares or contracts traded per period over a specified lookback window, most commonly the trailing 20, 30, or 90 trading days. Traders compare a security's current-day volume against this baseline to judge whether activity is unusually high or low, and screens use it as a liquidity filter to exclude thinly traded securities from consideration.
Key Takeaways
- Average volume is a simple mean of shares (or contracts) traded per period across a lookback window.
- The trailing 20-, 30-, and 90-day windows are the most common conventions, each trading off responsiveness against stability.
- It serves two distinct purposes: a baseline for spotting unusual activity, and a liquidity filter for screening.
- Relative volume - current volume divided by average volume - is the ratio traders use to quantify "unusual."
- A short lookback window reacts faster to recent changes but is noisier; a longer window is smoother but slower to adjust.
- Average volume says nothing about price direction on its own - it measures activity, not whether that activity is bullish or bearish.
- Corporate actions like stock splits and low starting sample sizes (recent IPOs) can distort an average volume figure.
How Is Average Volume Calculated?
Average volume is an arithmetic mean: sum the volume traded in each period across the lookback window, then divide by the number of periods. For a 20-day average volume. That means adding up the daily share (or contract) volume for the most recent 20 trading days and dividing by 20.
| Window | Formula | Typical use |
|---|---|---|
| 20-day average volume | Sum of last 20 days' volume ÷ 20 | Short-term baseline, common default on most charting platforms. |
| 30-day average volume | Sum of last 30 days' volume ÷ 30 | Roughly a calendar month of trading, a common screening default. |
| 90-day average volume | Sum of last 90 days' volume ÷ 90 | Smoother, longer-term liquidity baseline, less sensitive to a single active week. |
Worked example: suppose a stock trades the following volume over five consecutive days: 1.2 million, 900,000, 1.5 million, 1.1 million, and 2.8 million shares. Summing those figures gives 7.5 million shares; dividing by 5 gives an average daily volume of 1.5 million shares for that short window. If the stock then trades 4.5 million shares on the next day, its relative volume for that day is 4.5 million ÷ 1.5 million = 3x - three times its recent average, which is the kind of figure that would flag the day as unusually active. A real 20-, 30-, or 90-day average volume is calculated the same way, just summed and divided across a longer, rolling window that updates as each new trading day is added and the oldest day drops off.
Why Average Volume Matters
Average volume answers a simple question - "what's normal for this security?" - and that baseline supports two distinct uses. First, it's the denominator behind relative volume: dividing today's volume by the average volume produces a ratio that flags whether current activity is elevated or subdued relative to the security's own recent history. A reading well above 1x can accompany news, earnings, or a breakout attempt; a reading well below 1x can signal a quiet, low-conviction session. See Relative Volume for more on interpreting that ratio.
Second, average volume functions as a liquidity filter. Screens and systematic strategies commonly set a minimum average-volume threshold to exclude thinly traded securities before any other criteria are applied. This matters because low-volume names tend to have wider bid-ask spreads and greater price impact - the act of entering or exiting a position can move the price against the trader more than it would in a heavily traded name, which erodes returns independent of whether the underlying idea was sound. See Stock Screening for how liquidity filters typically fit into a broader screen.
Choosing a Lookback Window
There's no single correct window - the choice is a tradeoff between responsiveness and stability. A shorter window, such as 20 days, adjusts quickly to a genuine shift in trading activity (a new product launch, a change in float, an index addition) but is more sensitive to noise from any single unusually active or quiet day within that window. A longer window, such as 90 days, smooths out that noise and gives a more stable liquidity baseline, but it also reacts more slowly if a security's typical trading activity has genuinely changed recently.
In practice, traders often look at more than one window at once - for example, comparing a 20-day average against a 90-day average to see whether recent activity is trending up or down relative to the longer-term baseline, rather than relying on a single window in isolation.
Limitations and Common Mistakes
| Mistake | Why it's a problem | Better practice |
|---|---|---|
| Treating average volume as a directional signal | High or low volume says nothing about whether price is likely to rise or fall - it only measures how much trading activity occurred. | Pair average/relative volume with price action or another indicator before drawing a directional conclusion. |
| Ignoring corporate actions | A stock split changes share count and typically volume levels, which can create an artificial jump or drop in a rolling average that spans the split date. | Check for splits or other structural changes before trusting a rolling average that crosses that date. |
| Applying one window universally | A 20-day average calculated on a security with only 10 days of trading history (a recent IPO) is based on an incomplete, unrepresentative sample. | Confirm the security has enough trading history to fill the chosen window before relying on the figure. |
| Comparing average volume across unrelated securities | Absolute average volume isn't comparable across securities with very different share counts or price levels. | Use average volume to judge a security against its own history, or normalize with dollar volume when comparing across names. |
The Denominator Behind Every Volume Judgment
Average volume is rarely the thing you are actually looking at. It is the denominator underneath relative volume, the baseline behind a volume-confirmation check, and the screen filter that decided which names you saw at all. That makes it worth more care than a simple mean usually gets, because an error here does not announce itself. It quietly rescales every conclusion built on top of it.
Two situations break the baseline without anything looking wrong. A stock split changes the share count, so historical volume and current volume are denominated differently until the window rolls over, and the ratio you calculate spans two different units. A recently listed security has too few sessions to average meaningfully, and the figure it produces is precise-looking and close to arbitrary.
The lookback choice is the other decision worth making deliberately rather than accepting. A 20-day window adjusts quickly when trading activity genuinely changes, after an index addition or a shift in float, and it is also pulled around by any single frantic or dead session inside it. A 90-day window is steadier and slower to acknowledge that something real has changed. Neither is correct; they answer slightly different questions.
Whatever window you use, keep the measure in its lane. Average volume describes how much trading normally happens, and a day at three times normal is a statement about activity with no directional content whatsoever.
Frequently Asked Questions
What is a good average volume for a stock?
There is no single universal threshold - what counts as adequate average volume depends on position size and strategy. Screens commonly use a trailing 20-, 30-, or 90-day average volume filter to exclude thinly traded securities, but the specific cutoff a trader chooses depends on how large a position they intend to take relative to that average and how much price impact they're willing to accept when entering or exiting.
What is the difference between volume and average volume?
Volume is the number of shares or contracts traded in a single period, such as one trading day. Average volume is the mean of that daily figure calculated across a lookback window, most commonly the trailing 20, 30, or 90 trading days. Average volume smooths out single-day noise so a trader can judge whether today's volume is typical or unusual for that security.
How is relative volume different from average volume?
Average volume is the baseline figure itself - the mean shares traded per period over the lookback window. Relative volume is a ratio: current-period volume divided by that average volume, typically expressed as a multiple such as 2x or 0.5x. Average volume answers "what's normal for this security," while relative volume answers "how does today compare to normal."
Why do stock screens use average volume as a liquidity filter?
Average volume approximates how easily a security can be bought or sold without materially moving its price. Screens and strategies commonly set a minimum average volume threshold to exclude thinly traded securities, since low-volume names tend to have wider bid-ask spreads and higher price-impact costs, which can erode returns even when the underlying trade idea is sound.
Does average volume include off-exchange trading?
It depends on the source. A consolidated figure includes trades reported from off-exchange venues, which in many equity markets is a substantial share of total activity. A single-exchange figure counts only what traded there. The two can differ by a wide margin for the same security on the same day, and any threshold set against one is not valid against the other.
Should the average use a rolling window or a calendar period?
A rolling window produces a new value every session and changes smoothly as observations enter and leave. A calendar average, such as a monthly figure, updates in steps and holds the same value throughout the period. The rolling version is more responsive and more prone to the step artefact when a single large day rolls out of the window, which shows up as a sudden drop with no cause.
How much history does a stable average volume figure need?
Enough sessions that one unusual day cannot dominate. With a short window, a single event day can lift the average enough to change every subsequent relative-volume reading, which then reports normal activity as subdued. Longer windows are more stable and slower to reflect a genuine change in interest. Running two windows and comparing them is a practical way to see which situation you are in.
How does average volume behave after a large price move?
It rises sharply and then decays over subsequent sessions as interest fades, which means the average lags the underlying change in both directions. In the weeks after an event, the average is elevated by the event itself, so activity that would have looked heavy beforehand reads as ordinary. Any comparison against the average during that stretch is being made against a temporarily inflated baseline.
Does average volume mean the same thing for an ETF?
Not quite, and the difference is important for liquidity assessment. An ETF secondary market volume understates the size that can be traded, because authorised participants can create and redeem shares against the underlying basket. A fund with modest screen volume can absorb far more than that figure suggests if its holdings are liquid. Judging an ETF by its own traded volume alone systematically overstates the constraint.