Home / Wiki / A / Average True Range (ATR)

Average True Range (ATR)

Trading and SpeculationUpdated 16 Nov 2026

Average True Range (ATR) is a technical indicator that measures how much a market typically moves in one period. It is a smoothed average of the true range, a version of the daily high-to-low range that also takes account of any gap from the previous close. ATR is expressed in the price units of the instrument and says nothing about direction: a high value means large moves, up or down.

The indicator was introduced by J. Welles Wilder in his 1978 book on technical trading systems, which also presented the Relative Strength Index. Wilder designed it for commodity futures, where a market can open far from the prior close or lock at its daily price limit, so that the simple range of the day understates the real movement. It is now a standard tool of technical analysis in all markets, used less to generate signals than to scale other decisions to current volatility.

01

Formula

The true range of a period is the largest of three distances:

TRt=max⁡(Ht−Lt,  ∣Ht−Ct−1∣,  ∣Lt−Ct−1∣)TR_t = \max \left( H_t - L_t, \; \left| H_t - C_{t-1} \right|, \; \left| L_t - C_{t-1} \right| \right)

The average true range is then updated each period with Wilder's smoothing:

ATRt=ATRt−1×(n−1)+TRtnATR_t = \frac{ATR_{t-1} \times (n - 1) + TR_t}{n}
  • HtH_t and LtL_t are the high and low of the current period.
  • Ct−1C_{t-1} is the close of the previous period.
  • nn is the number of periods, 14 in Wilder's original and in most software.

The first value is a simple average of the first nn true ranges. Wilder's smoothing is equivalent to an exponential moving average with a smoothing factor of 1/n1/n, so it reacts more slowly than a standard exponential average of the same length.

02

Worked example

Suppose a share closed yesterday at 50 and today opens higher, trading between a low of 51 and a high of 53. The three distances are 53−51=253 - 51 = 2, ∣53−50∣=3|53 - 50| = 3 and ∣51−50∣=1|51 - 50| = 1. The true range is 3, larger than the day's own range of 2 because of the gap.

If the 14-period ATR stood at 2.00 yesterday, today's value is (2.00×13+3)/14=29/14=2.07(2.00 \times 13 + 3) / 14 = 29 / 14 = 2.07. One unusually large day lifts the average only slightly, and its influence fades gradually over the following weeks.

03

Uses

  • Stop placement. Traders set a stop-loss order at a multiple of ATR from the entry or from a recent extreme, so that the stop lies outside normal noise. The Chande Kroll Stop and similar trailing methods are built this way.
  • Position sizing. Dividing the amount a trader is willing to lose by the stop distance gives the position size. With 1,000 at risk and a stop two ATRs away, at an ATR of 2.07 the stop distance is 4.14 and the size is about 241 shares. Volatile instruments thus receive smaller positions.
  • Channels and filters. Bands drawn at a multiple of ATR around a moving average form a volatility channel, comparable in purpose to Bollinger Bands. Systems for breakout trading may require a move of at least one ATR before acting.
  • Market state. A falling ATR indicates a quiet, narrowing market and a rising one an active market. The Choppiness Index uses it as an input.

Sizing by ATR was a central rule of well-known trend following programs of the 1980s, which equalized risk across futures markets of very different volatility.

04

Comparison with other volatility measures

ATR works on price ranges and is stated in currency or points. The standard deviation of returns, the basis of realized volatility, works on closing prices and is stated as a percentage. Because ATR uses the high and low, it captures movement within the session that close-to-close measures miss. Because it is in price units, it cannot be compared directly between instruments or over long periods in which the price level has changed; for that purpose analysts divide ATR by the closing price to obtain a percentage form. Other range-based indicators include Chaikin Volatility.

05

Limitations

ATR is a lagging description of recent conditions and does not forecast. Volatility can change abruptly, and a stop or position size set from a calm period may prove inadequate when conditions shift. The choice of period length and of the multiple applied to it is arbitrary, and results from backtests are sensitive to both. On intraday charts the gap term matters only at the session open, and in markets that trade around the clock it has little effect.

Last updated 16 Nov 2026