Average True Range (ATR)

A volatility gauge that measures how much price moves on average over a chosen period, expressed in price units.

What it is

ATR (Average True Range) is a volatility indicator developed by J. Welles Wilder Jr. and introduced in his 1978 book "New Concepts in Technical Trading Systems." It measures the average size of price movement over a set number of bars, expressed in the instrument's price units (e.g. pips, dollars, points). ATR does not tell you the direction of a trend — only how much the market is moving. A rising ATR reflects larger bar ranges (more volatility); a falling ATR reflects smaller bar ranges (calmer conditions). It is most often plotted as a single line in a separate pane below the price chart.

How it works

ATR is built on the concept of "True Range," which captures the full size of a bar's move including any gap from the previous close. True Range for each bar is the largest of three values: (1) the current high minus the current low, (2) the absolute difference between the current high and the previous close, and (3) the absolute difference between the current low and the previous close. Using the previous close in two of these lets True Range account for overnight or session gaps that a simple high-minus-low would miss. ATR is then a moving average of True Range over a chosen lookback (Wilder's default is 14 periods). Wilder used his own smoothing method — often called Wilder's smoothing or RMA — which behaves like an exponential moving average with a smoothing factor of 1/length, so older values fade gradually rather than dropping out abruptly. Because it averages absolute ranges, ATR is always non-negative and reads in the same units as price, so its typical value depends entirely on the instrument and timeframe.

How traders read it

• Traders often read a rising ATR as expanding volatility (wider bar ranges) and a falling ATR as contracting volatility (calmer, tighter price action).
• Because ATR is reported in price units, traders commonly reference it when sizing stop-loss distances or positions relative to current volatility — for example placing a stop some multiple of ATR away from entry. This is a way of measuring volatility, not a recommendation to trade.
• ATR is direction-agnostic: a high ATR can occur in both up and down moves, so it is typically read alongside a separate trend or direction tool rather than on its own.
• Some traders treat very low ATR readings as a sign of a quiet, range-bound market, while spikes in ATR are often read as bursts of activity around news or breakouts — though ATR describes past movement and is not predictive of what comes next.
• ATR values are not comparable across different instruments or timeframes without normalising, since a $200 stock and a $5 stock will naturally show very different ATR magnitudes.

Common settings

Length 14 is Wilder's original default. Shorter lengths (e.g. 7 or 10) react faster and look noisier; longer lengths (e.g. 20–50) are smoother and slower. ATR is usually plotted in its own pane. Volatility-based stop or sizing approaches that reference ATR often use a multiplier such as 1.5x, 2x, or 3x, chosen by the trader; these are illustrative settings, not recommendations.

Strengths

• Captures gaps and overnight moves via True Range, giving a fuller picture of volatility than simple high-minus-low.
• Reported in price units, which makes it intuitive for gauging volatility-aware stop distances and position sizes.
• Works on any market and any timeframe, with the lookback length being the main parameter to set.
• Simple, well-documented, and built into virtually every charting platform (ta.atr in Pine), so its behaviour is consistent and easy to verify.

Pitfalls to watch

• ATR gives no directional information whatsoever — it cannot indicate whether price will rise or fall, only how large recent moves have been.
• It is a lagging, smoothed measure: ATR reacts to volatility that has already happened and does not forecast future volatility or price.
• Raw ATR values are scale-dependent, so comparing ATR between assets (or even the same asset at very different price levels) is misleading unless you convert it to a percentage of price.
• A single large bar or gap can lift ATR for several periods because of the smoothing, which can distort volatility-based calculations if not accounted for.
• No indicator, including ATR, is predictive; all trading carries risk of loss, and ATR-based stops or sizing do not guarantee protection against adverse moves, gaps, or slippage.

Pine v6 example

//@version=6
indicator("Average True Range", overlay = false)

length = input.int(14, "ATR Length", minval = 1)

// ta.atr applies Wilder's smoothing (RMA) to True Range
atrValue = ta.atr(length)

plot(atrValue, "ATR", color = color.orange, linewidth = 2)

Pro tip: To compare volatility across different instruments or price levels, divide ATR by the current price to get a percentage (ATR%). This normalises the reading so a quiet $300 stock and a quiet $20 stock can be assessed on the same scale.

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