Twelve million shares traded is either enormous or unremarkable depending entirely on the stock. Relative volume answers the only version of the question that means anything: how does today compare to this stock’s own normal?
This article is for informational and educational purposes only. It is not financial advice, investment advice, or a recommendation to buy, sell, or hold any security, cryptocurrency, or financial product. Always verify data with official sources before making financial decisions.
Short answer: today’s volume against its own baseline
Short answer: Relative volume, usually written RVOL, divides current volume by the average volume over the same period on comparable prior days. An RVOL of 3.0 means three times normal activity. The critical detail — and the one most implementations get wrong — is that the comparison must be time-of-day adjusted, because volume is not distributed evenly across a session.
Calculating RVOL properly
The naive version divides today’s cumulative volume by the average full-day volume over some lookback. At 10:00 a.m. that produces a small number for every stock in the market, because barely any of the day has elapsed.
The correct construction compares cumulative volume at a given point in the session to the average cumulative volume at that same point across the lookback period. If a stock typically trades 18% of its daily volume in the first thirty minutes, then 10:00 a.m. volume gets measured against that 18% baseline rather than against the full day.
The lookback length is a genuine trade-off. Twenty days is common and adapts quickly to changed conditions; it also means a single prior spike inflates the baseline for a month and suppresses subsequent readings. A median rather than a mean is more robust for exactly this reason and is used less often than it should be.
Why raw share volume misleads
Two comparisons make this concrete. A stock with 50 million shares outstanding trading 5 million shares has turned over 10% of its float. A stock with 5 billion shares outstanding trading the same 5 million has turned over 0.1%. The raw number is identical; the events are not comparable.
Dollar volume — shares multiplied by price — is a better cross-sectional measure for comparing different stocks, because it reflects capital committed rather than certificates moved. RVOL and dollar volume answer different questions: RVOL says whether this stock is unusually active versus itself, dollar volume says whether it is meaningfully active in absolute terms. A microcap at 20x RVOL may still be trading trivial dollar amounts.
The thresholds traders use, and why they are arbitrary
Common conventions treat 2.0 as elevated and 5.0 as extreme. These are conventions rather than findings — nothing in the data distinguishes 1.9 from 2.1, and the appropriate threshold varies enormously by market capitalisation, sector and prevailing volatility.
A large, liquid stock rarely reaches 3x without a genuine catalyst. A thinly traded small cap can reach 3x because two institutional orders happened to arrive in the same hour. Applying one threshold across a screen guarantees the results are dominated by the least liquid names.
Volume without price movement: absorption or apathy
High RVOL with a large price move is straightforward — something happened. High RVOL with little net price change is the more interesting case and admits two opposite readings.
Absorption means a large buyer is meeting a large seller near a price level, with size transacting and neither side prevailing. This frequently precedes a decisive move once one side is exhausted, and the direction of that move is not knowable in advance.
Churn means high-frequency and algorithmic activity generating volume without net accumulation or distribution. It looks identical on a volume bar.
Distinguishing the two from public data is difficult. Intraday VWAP behaviour and whether volume clusters at particular price levels offer weak evidence. Anyone claiming to identify absorption confidently from a daily chart is inferring more than the data supports.
Catalysts that reliably produce high RVOL
- Earnings. The most reliable, and scheduled — which means an RVOL screen run the morning after earnings season returns mostly noise.
- Index changes. Additions and deletions force mechanical buying and selling by passive funds, frequently concentrated in the closing auction.
- Analyst actions and guidance changes.
- Sector-wide news. Produces elevated RVOL across many correlated names simultaneously, which is a useful filter — if the whole sector is at 3x, the individual reading carries no stock-specific information.
- Options expiry and quarterly rebalancing. Calendar effects that recur predictably and should be filtered out rather than interpreted.
Where it shows up on EskiSignal
RVOL is the underlying calculation behind most unusual volume screens, including the ones that generate the movers coverage on this site. Understanding the construction is what separates a screen result worth investigating from one that reflects a low-float stock and a slow lookback period.
What this article does not conclude
RVOL is a descriptive statistic. It identifies that participation is unusual, not why, and not what happens next. It has no directional content whatsoever — the same reading accompanies a stock up 20% and a stock down 20%.
Implementations differ substantially in lookback length, time-of-day adjustment, whether pre-market volume is included and whether the average is a mean or median. Two platforms will report different RVOL for the same stock at the same moment.