ATR Bands

Volatility envelopes plotted a multiple of Average True Range above and below price, sized by how much the market has actually been moving.

What it is

ATR Bands are a volatility envelope built by adding and subtracting a multiple of the Average True Range (ATR) from a price source, typically the close. The result is two lines that track price directly, with a gap that widens when recent bars have been large and narrows when they have been small. The idea is straightforward: instead of drawing an envelope a fixed percentage away from price, size the envelope by the market's own recent movement, so 'far from price' means the same thing in calm and volatile conditions. This is also the cleanest way to answer how ATR Bands differ from Keltner Channels: ATR Bands are centered on price itself (or another raw source), while Keltner Channels place the same kind of ATR-based envelope around an exponential moving average — a smoothed centerline rather than raw price. The distinction matters because Keltner Channels lag price twice (once in the EMA, once in the ATR) while ATR Bands lag only through the ATR. ATR Bands describe recent volatility around price; they do not predict where price will go. This is educational material only, not financial advice, and trading always involves risk of loss.

How it works

The engine is the Average True Range. Each bar's true range is the largest of: high minus low, the absolute distance from the previous close to the current high, and the absolute distance from the previous close to the current low — a definition that captures gaps as well as intrabar range. ATR is then a smoothed average of true range, and Wilder's original smoothing (the RMA, which is what ta.atr uses in Pine Script) responds gradually to changes in bar size. The bands are simply: upper band = source + multiplier × ATR, lower band = source − multiplier × ATR, with 14-period ATR and a multiplier around 2.0 as common defaults. Because the source is usually the raw close, the bands follow every wiggle in price, and only the width evolves smoothly. Compare the family: Bollinger Bands use standard deviation of price around an SMA basis; Keltner Channels use ATR around an EMA basis; ATR Bands use ATR around price itself. Each choice trades centerline smoothness against responsiveness. Related tools such as the Chandelier Exit and SuperTrend take the same ATR offset but add a ratchet or flip rule so the line trails price instead of straddling it.

How traders read it

• Band width as a volatility gauge: a widening gap between the bands means recent true ranges have grown; a narrowing gap means bars have been small. This describes the recent past, not the future.
• Distance context: the bands show how far price would need to travel to make a move that is 'large' relative to its own recent behavior — a two-ATR excursion means the same thing on a calm chart and a volatile one.
• Stop and target sizing: many traders use the band offset (a multiple of ATR) as a reference for placing stops or judging whether a target is realistic given current volatility, rather than as an entry signal.
• Not touch-and-reverse lines: because the bands are anchored to price itself, price cannot 'break out' of them the way it can with a fixed centerline — a close outside a band mostly reflects a sudden single-bar range expansion.
• Versus Keltner Channels: if you want an envelope that can act as a trend reference with price moving inside and outside of it, that is Keltner behavior (ATR around an EMA). ATR Bands hugging raw price answer a different question: how wide is normal movement right now?
• Any reading is context, not instruction — combine with price structure and a defined risk plan.

Common settings

ATR length 14 (Wilder's original default) with a multiplier of 2.0 is the most common starting point; multipliers from 1.0 to 3.0 are widely used depending on how much room the envelope should give. The source is usually the close, though some versions center on hl2 or an EMA — at which point the tool effectively becomes a Keltner Channel. As always, the defaults are conventions rather than optimal values, and changing length, multiplier, or source changes every reading.

Strengths

• Self-scaling: the envelope automatically adapts to each instrument's and timeframe's volatility, unlike fixed-percentage envelopes.
• Gap-aware: because true range includes the jump from the prior close, the bands respond to gaps that a simple high-minus-low measure would miss.
• Transparent construction: two lines of arithmetic on top of a well-documented ATR — easy to verify and hard to get subtly wrong.
• Useful as a shared volatility yardstick for stop placement and position sizing discussions, since 'two ATRs' is unambiguous.
• A clean gateway to the ATR family: Keltner Channels, Chandelier Exit, SuperTrend, and volatility stops all build on the same offset.

Pitfalls to watch

• ATR lags: the smoothing means the bands widen after volatility has already expanded and stay wide after it has contracted, so the envelope always describes slightly stale conditions.
• Not breakout bands: centered on raw price, the bands move with every bar — treating a band touch as a signal misreads the construction, since price is always by definition near the middle.
• Multiplier sensitivity: whether the envelope looks 'respected' or 'meaningless' depends heavily on the multiplier, and a value that suited one chart or period may not suit another.
• Easy to confuse with Keltner Channels: many published scripts labelled 'ATR bands' actually center on an EMA; comparing values across platforms without checking the centerline invites false mismatch reports.
• ATR says nothing about direction: wide bands mean big bars, not a trend, and narrow bands mean small bars, not an imminent move.
• No predictive power: the bands summarize past ranges only; no envelope setting forecasts price or removes the risk of loss.

Pine v6 example

//@version=6
indicator("ATR Bands", overlay = true)

atrLen = input.int(14, "ATR Length", minval = 1)
mult   = input.float(2.0, "ATR Multiplier", minval = 0.1, step = 0.1)
src    = input.source(close, "Source")

atrValue  = ta.atr(atrLen)
upperBand = src + mult * atrValue
lowerBand = src - mult * atrValue

plot(upperBand, "Upper ATR Band", color = color.teal)
plot(lowerBand, "Lower ATR Band", color = color.teal)
plot(src, "Center (Source)", color = color.new(color.gray, 50))

Pro tip: Before comparing ATR Bands across platforms or scripts, confirm three things: the ATR length, the multiplier, and — most often missed — what the bands are centered on. A script centered on an EMA is a Keltner Channel wearing an ATR Bands label, and its values will legitimately differ from a price-centered version. Use the bands as a volatility yardstick for risk placement rather than as entry signals. Educational content only, not financial advice; all trading carries risk of loss.

Educational only — not financial advice, not a recommendation to trade. No indicator is predictive; trading involves substantial risk of loss.