ATR (Average True Range)

J. Welles Wilder’s indicator that averages the true range of candles, usually over 14 periods. Used to gauge volatility, place stops and size positions.

ATR, or Average True Range, was introduced by J. Welles Wilder in 1978. It measures volatility as the average range the price travels per candle, taking into account gaps between the previous close and the next open. It does not indicate direction, only how much the instrument moves. The usual period is 14, applied to candles of any timeframe.

First you compute the True Range of each candle as the largest of three values: high minus low (H − L), the absolute value of the high minus the previous close, and the absolute value of the low minus the previous close. Example on daily EUR/USD: high 1.0850, low 1.0810, previous close 1.0860. H − L = 0.0040; |H − Cprev| = 0.0010; |L − Cprev| = 0.0050. The True Range is 0.0050, that is 50 pips, larger than the visible range of the candle because of the gap.

Wilder then smooths those ranges with his own average: current ATR = (previous ATR × 13 + current TR) / 14. If yesterday’s ATR was 0.0080 (80 pips) and today’s TR is 0.0050, the new ATR is (0.0080 × 13 + 0.0050) / 14 = 0.1090 / 14 ≈ 0.00779, about 77.9 pips. To start, the first ATR is the simple average of the first 14 true ranges. Some platforms use a simple moving average instead, so values can differ slightly.

Its most widespread use is the volatility-adjusted stop loss. A stop at 1.5 times ATR gives the price room proportional to its normal movement. With an ATR of 77.9 pips, 1.5 × 77.9 ≈ 116.8 pips. Many traders use between 1.5 and 3 times ATR depending on style; there is no universally correct multiple, and it should be tested with data.

Position size follows from it. With a 10,000 USD account and 1% risk, you are risking 100 USD. With a 116.8-pip stop and a pip value of about 10 USD per lot, the lot size is 100 / (116.8 × 10) ≈ 0.0856, rounded down to 0.08 lots. If tomorrow ATR shrinks, the stop tightens and the lot size grows; if it rises, the lot size falls. The money at risk stays constant.

Caveats: ATR is expressed in price units, so it is not comparable across instruments unless you normalise it by dividing by price. It is a lagging indicator: it is slow to react to a sudden change of regime. And it depends on the timeframe: a 14-period ATR on 5-minute candles and one on daily candles measure very different things.

Common mistakes: placing the stop within 0.5 ATR, where normal noise reaches it often; forgetting to recalculate the lot size when ATR changes; and using it to predict direction. It is a tool for sizing risk, not for deciding when to enter.

More terms