ATR Indicator: Average True Range Explained

The Average True Range (ATR) is a volatility indicator that measures how much an asset typically moves over a set period, on average. Developed by J. Welles Wilder Jr. in 1978, ATR does not predict direction. It tells you the size of recent price swings, which traders use to set stops and size positions to current market conditions.

What is the ATR indicator?

ATR is a volatility gauge introduced by J. Welles Wilder Jr. in his 1978 book New Concepts in Technical Trading Systems, the same work that gave technical analysts the RSI and the ADX. It answers one question: how much does this asset usually move in a single period? The higher the number, the wider the typical range and the more volatile the market.

Crucially, ATR says nothing about trend or direction. A stock can have a high ATR while falling, rising, or chopping sideways. That neutrality is the point. ATR is a measuring tape for movement, and traders pair it with directional tools like moving averages or support and resistance to decide what to do with that measurement.

How is ATR calculated? (the formula)

ATR is built in two steps. First you compute the true range for each period, then you average it.

The true range is the largest of three values:

True Range = max of:
  (Current High - Current Low)
  |Current High - Previous Close|
  |Current Low  - Previous Close|

Including the previous close is what makes ATR honest about gaps. If a stock closes at 100 and gaps up to open at 108, the simple high-minus-low for that day understates the real move; the true range captures the full 8-point jump.

The ATR itself is a Wilder smoothed average of the true range, with a default lookback of 14 periods:

First ATR = average of the first 14 true ranges

Subsequent ATR = [ (Previous ATR x 13) + Current True Range ] / 14

The smoothing means a single wild day nudges ATR rather than spiking it, so the line reflects a sustained shift in volatility rather than one outlier.

What ATR is actually measuring: volatility

ATR is one answer to a broader question, and the broader question is worth stating. Volatility is the dispersion of an asset's returns, how far and how fast price moves around its average. The standard statistical measure is the standard deviation of returns: a high number means returns are spread widely, a low number means they cluster tightly. Two assets can have the same average return while one whipsaws and the other drifts, and volatility is what separates them. It is a measure of uncertainty, never of direction.

There are two distinct kinds, and confusing them causes real errors. Historical (realised) volatility measures how much price actually moved over a past window. It is backward-looking and objective, computed from the price history, and it answers "how turbulent has this asset been?" Implied volatility is forward-looking, derived from option prices, and answers "how turbulent does the market expect this to be?" The two frequently diverge, with implied volatility rising ahead of known events like earnings even when recent realised volatility was calm. The VIX is the best-known example of the second kind: an index of the market's expected 30-day volatility of the S&P 500, backed out of S&P 500 option prices, which spikes when markets fall and drifts lower during calm uptrends.

ATR sits on the historical side and expresses it in price units rather than percentage terms, which is exactly why it is usable for stops and sizing while a standard deviation of returns is not.

Two behaviours make volatility partly forecastable. The first is clustering: volatile periods tend to be followed by more volatile periods, and calm by calm, so large moves arrive in bunches. The second is mean reversion: volatility returns toward its long-run average, so extreme readings, high or low, rarely persist. Neither gives a timing edge on its own, because extremes can last longer than expected. Together they explain why measuring recent volatility tells you something useful about the near future even though it says nothing about which way price will go, and why a long calm stretch is often followed by a return of bigger moves.

How do you read high vs low ATR?

Read ATR in relative terms, not absolute ones. A $5 ATR on a $500 stock is calm; a $5 ATR on a $40 stock is violent. Always compare ATR to the asset's own recent history and to its price.

• Rising ATR: ranges are expanding, volatility is increasing. This often accompanies breakouts, news events, and sharp trends. Stops need more room and positions should be smaller.
• Falling ATR: ranges are contracting, the market is quieting. Low-ATR conditions often precede a volatility expansion, which is why traders watch ATR squeezes for the next big move.
• ATR as a floor: because it represents typical movement, a stop placed inside one ATR of entry is statistically likely to be hit by normal noise, not a real reversal.

How do you use ATR for stops and position size?

This is where ATR earns its place. Because it quantifies normal movement, ATR lets you set stops and size trades to conditions instead of guessing.

For a volatility-based stop, place your stop a multiple of ATR from entry. A common range is 1.5 to 3 times ATR. In a market with an ATR of 2.00, a 2x ATR stop sits 4.00 away from entry. When volatility rises and ATR climbs to 3.00, the same 2x rule widens the stop to 6.00 automatically, so you are not stopped out by ordinary noise during turbulent sessions.

For position sizing, the stop distance feeds directly into how many shares or contracts you can hold while risking a fixed percentage of your account. If you risk 1% of a $20,000 account, that is $200 of risk; with an ATR stop 4.00 wide, you can hold 50 shares ($200 / $4.00). Tie the two together and your size shrinks automatically when volatility expands.

ATR multiples at a glance

The right multiple depends on your timeframe and tolerance for noise. The table below is a starting framework, not a rule.

Wider multiples mean fewer premature exits but larger losses when wrong, so they pair with smaller positions. The multiple and the size are two halves of the same risk decision.

Common ATR mistakes and limitations

ATR is reliable but easy to misread.

1. Treating ATR as a direction signal. It is not. A high ATR does not mean buy or sell, only that moves are large.
2. Comparing raw ATR across assets. A $3 ATR means nothing without the price. Convert to a percentage first.
3. Using one fixed stop multiple forever. The right multiple shifts with timeframe and asset; backtest before committing.
4. Ignoring gaps. ATR includes gaps by design, so it can read higher than a simple range on gappy assets, which is a feature, not a flaw.
5. Lag. Like all averages, ATR reacts to volatility changes with a delay, so it confirms rather than predicts.

Putting ATR in context

Think of ATR as the volatility lens that sits behind your other tools. A directional tool tells you which way you lean; ATR tells you how much room to give the trade and how big a position the current market can justify. Used together, they keep risk proportional to conditions instead of fixed to habit.

ATR sits in the volatility family of our technical indicators hub, alongside Bollinger Bands and Keltner Channels.