The ATR, introduced by J. Welles Wilder in 1978, measures volatility: how far price usually travels in one bar. Unlike a simple high-minus-low range, it accounts for gaps between sessions. It does not say anything about direction, which makes it a risk tool more than a signal tool.
How it’s calculated
True range (TR) is the largest of:
- current high − current low
- |current high − previous close|
- |current low − previous close|
ATR is the Wilder-smoothed average of TR, standard length 14:
ATR = (previous ATR × 13 + current TR) ÷ 14
The first ATR is usually a simple average of the first 14 true ranges.
Example
Previous close 100.00; today’s high 103.00 and low 99.00.
- H − L = 4.00; |H − prev close| = 3.00; |L − prev close| = 1.00
- TR = 4.00
If yesterday’s ATR was 2.50:
- ATR = (2.50 × 13 + 4.00) ÷ 14 ≈ 2.61
A gap case: previous close 100.00, high 108.00, low 105.00. High − low is only 3.00, but |high − previous close| is 8.00, so TR = 8.00. That is why ATR captures gap risk.
Using it for a stop: buying at 102.00 with a stop 1.5 ATR away gives 102.00 − 1.5 × 2.61 ≈ 98.09. Risking $200 on a stop distance of $3.91 gives a position size of about 51 shares.
Common mistakes
- Reading ATR as bullish or bearish. It only measures range.
- Comparing raw ATR across assets. An ATR of 2 on a $20 stock is far larger than on a $500 stock; divide by price to compare.
- Fixed stops regardless of volatility. A $1 stop-loss may be wide on a quiet day and inside the noise on a volatile one.
- Forgetting that ATR changes. Volatility expands around news; resize positions as ATR rises. A trailing stop based on ATR adapts automatically.
The position size calculator turns an ATR-based stop into a share or lot count. See also RSI, MACD, ATR and EMA explained.
In SnapPulse
SnapPulse explains the ATR reading of a scanned chart in plain language and sizes the position on the device from your balance, risk % and stop distance. Download SnapPulse.
Educational content — not financial advice.
Updated