Volatility & bandsRelative Volatility Index · RVI
RSI's formula applied to standard deviation — a confirmation gauge for the direction of volatility.
Works in most conditionsEngine-computed on a fixed sample series
What the Relative Volatility Index measures
The Relative Volatility Index (RVI), created by Donald Dorsey in 1993 and refined in 1995, takes the familiar RSI formula and points it at volatility instead of price. Where RSI asks whether recent price changes have been mostly up or mostly down, the RVI asks whether recent volatility — measured as the standard deviation of price — has expanded more on up-days or on down-days. The output is a 0-to-100 line, but its meaning is entirely different from a momentum oscillator: it tells you the direction in which volatility is building, not the direction of price itself. Dorsey never intended it as a standalone signal generator; he designed it as a confirming filter to strengthen or veto signals from another system. For a beginner, think of it as a second opinion that measures whether the market's energy is being spent pushing prices up or driving them down. It is important not to confuse this Dorsey indicator with the unrelated Relative Vigor Index, which shares the same RVI abbreviation.
How the RVI is calculated
First compute the standard deviation of price over a short window, typically 10 days, giving a running measure of how dispersed prices have been. Then, for each bar, look at whether the close rose or fell versus the prior close; if it rose, that bar's standard deviation is added to the up-volatility bucket and zero to the down bucket, and if it fell, the reverse. These up- and down-volatility figures are smoothed with Wilder's 14-period averaging, exactly as RSI smooths its gains and losses. The RVI is then 100 times the average up-volatility divided by the sum of the average up- and down-volatility, producing a 0-to-100 reading. Dorsey's 1995 refinement runs the whole calculation twice — once on highs and once on lows — and averages the two to steady the output and reduce single-series noise.
Reading the RVI, step by step
The 50 line is the fulcrum: above 50, volatility is expanding predominantly on up-moves, supporting a bullish case, and below 50 it is expanding on down-moves, supporting a bearish one. Because Dorsey built it as a filter, the intended use is confirmation — a buy signal from your primary system is stronger when the RVI is above 50, and a sell signal is stronger when it is below 50. Some of Dorsey's own rule sets use wider bands, requiring the RVI above 60 to confirm and initiate longs and below 40 to confirm shorts, to demand clearer volatility conviction. Rising RVI shows volatility increasingly favoring the upside; falling RVI, the downside. On its own the line is deliberately unremarkable — its value comes from agreeing or disagreeing with another signal, not from crossovers of its own.
Best timeframes and settings
The RVI is a daily-chart tool by design, matching the swing- and position-trading horizon Dorsey worked on, and the standard settings are a 10-period standard deviation smoothed over 14 periods. It can be applied to intraday charts, but volatility measured over very short windows becomes erratic, so it is more dependable on daily and weekly data. Lengthening the standard-deviation window makes the volatility read smoother and slower; shortening it makes the RVI jumpier and more prone to flipping across 50. As a confirmation filter its exact tuning matters less than its consistency, so most traders leave it near the defaults and focus on how it aligns with their primary system. The responsiveness-versus-noise trade-off applies to the standard-deviation window most of all.
When and where to use it
Because it is a filter rather than a signal, the RVI is useful across regimes — trending or ranging — as a check on whichever primary tool you are using. It adds the most value in markets where distinguishing a real, volatility-backed move from a hollow one matters, helping you avoid breakouts that occur on shrinking volatility. It applies to any liquid asset with a clean price series, since standard deviation is asset-agnostic. Where you should avoid relying on it is as a solo decision tool, since used alone it produces weak, ambiguous readings that were never meant to be traded in isolation. It is also less useful on instruments with erratic or gappy pricing that distorts the standard-deviation input.
Strategies that use the RVI
The canonical strategy layers the RVI onto a primary system and takes only the signals it confirms: accept long entries from your moving-average or breakout system when the RVI is above 50, and short entries when it is below 50, discarding the rest. A stricter volatility-confirmation strategy uses the 60 and 40 bands — go long only when a buy setup coincides with RVI above 60 and short only when a sell setup coincides with RVI below 40 — to filter for stronger volatility conviction. A breakout-validation strategy uses the RVI to judge whether a price breakout has volatility expanding in its direction; a breakout above resistance with RVI climbing through 50 is far more trustworthy than one on a falling RVI. In every case the RVI is the gatekeeper, never the entry trigger itself.
Combining the RVI with other indicators
The RVI is meant to sit beside a primary directional tool — a moving-average crossover, a Donchian or price breakout, or a momentum oscillator — supplying volatility confirmation those tools lack. It complements price-direction indicators especially well because it measures an orthogonal dimension: RSI or MACD tell you where price momentum points, and the RVI tells you whether volatility agrees. It also works alongside Bollinger Bands, since both are rooted in standard deviation; a band breakout with a supportive RVI is a cleaner signal. Pairing it with ATR or historical volatility can help distinguish whether rising volatility is directional, with the RVI leaning one way, or merely broad and undirected.
Where the RVI fails
The most common mistake is trying to trade the RVI on its own, which produces vague, low-edge signals precisely because Dorsey engineered it to be a supporting filter. Around the 50 line in quiet markets it can flip back and forth, offering no clear read, so demanding a buffer such as 60 and 40 helps. Confusing it with the Relative Vigor Index leads traders to expect signal-line crossovers it does not have. Because it depends on standard deviation, gappy or thin instruments can throw distorted volatility readings that mislead the filter. The way to avoid these failures is to use it exactly as intended — as a yes or no confirmation on another system's signals — and never to expect it to generate entries by itself.
A worked example
Suppose a swing trader runs a 50-day moving-average breakout system on a stock and adds Dorsey's RVI as a filter with the default 10 and 14 settings. Price breaks above the 50-day average, generating a raw long signal at 84.00. The trader checks the RVI: it reads 63, above the 60 confirmation band, meaning the recent volatility has been expanding on up-days — the move has genuine energy behind it. The long is taken with a stop below the breakout base at 80.50. In a second case a week later, price nudges above the average again at 86.00, but this time the RVI reads 44, below 50 — volatility is actually building on the downside, so the breakout is hollow. The trader passes on that signal, and the false breakout duly fails, demonstrating the RVI's role as a veto rather than a trigger.