Average True Range, or ATR, is a smoothed average of a price range called True Range. This includes the distance between an interval’s highest and lowest prices and, when relevant, the distance from the previous closing price. It measures the size of price fluctuations, not their direction.
Petr opens a chart and sees quiet candles alongside others covering a wide price range. A candle summarizes a time interval: High is its highest price, Low its lowest, and Close its last price. TradingView describes True Range, or TR, as the largest of High minus Low and the distances of those prices from the previous Close. Distances are taken without a sign. The values are not added; the largest is selected. The result is never negative and can be zero if prices stay completely still. [1]
Imagine a new interval starting well above where the previous one ended, then barely moving within that interval. High minus Low alone would show a small range. Fidelity explains that True Range also includes the distance from the previous Close, so the jump does not disappear from the measurement. Petr is following the connection between neighboring intervals, not just the size of the current candle. A numerical example appears below. [2]
One turbulent candle does not describe the longer picture. ATR therefore smooths individual ranges into an ongoing average. Fidelity gives 14 periods as a common length. A period is not automatically a day: on an hourly chart it means hourly intervals, on a daily chart days. Before comparing values, Petr checks both the chart interval and the indicator’s length. The same name without the same settings does not ensure comparable values. [2]
Fidelity writes the next value as the previous ATR multiplied by (n − 1), plus the current TR, with the whole sum divided by n. The letter n is the chosen length. The old result carries some history, and the new range gradually changes it. TradingView uses this RMA smoothing by default; settings also offer SMA, EMA and WMA, other ways of averaging. Initialization can affect the result too. Petr therefore checks the method, starting point and data source. [2] [1]
If the chart quotes price in USD/BTC, ATR uses that same unit. It is a range in the price per bitcoin, not the total amount of Petr’s purchase. For relative comparison, TradingView offers ATR%: ATR / Close × 100, expressing the range relative to price as a percentage. Close must be nonzero. This helps compare different price levels, but does not remove differences in chart intervals, smoothing or data. [3]
Petr checks again and ATR has changed before the candle ends. TradingView explains that High, Low and the current Close can change during an open interval. Current TR and ATR can change with them. This alone is not a malfunction. What matters is distinguishing decisions made during the candle from decisions made after it closes. A historical test must not give an earlier decision final values that were unavailable at the time. [4]
Fidelity and TradingView both stress that ATR is nondirectional. It can rise during a sharp price increase or decline. Petr may use it to set the distance of a protective order, but an ATR multiple does not guarantee a maximum loss. An order being triggered does not guarantee its execution price. The outcome also depends on the amount of bitcoin traded, fees and slippage, the difference between expected and actual execution price. [2] [5]
ATR helps Petr describe how turbulent recent conditions have been. It does not choose position size or entry and exit times for him. TradingView’s strategy documentation distinguishes rules from simulated execution. Decide in advance when to read the value and how to act on it, and include costs. Overfitting means tailoring settings so closely to past results that they fail elsewhere. Further testing therefore belongs on data not used for tuning. [5]
A small candle, a larger jump between intervals
For the clearest picture, read this entry together with Volatility, Candlestick chart, Stop-loss, Position Sizing, Backtesting, Technical indicator. The reverse links also lead from Average Directional Index, On-Balance Volume.
01Does rising ATR mean bitcoin is rising?+
No. The ATR line follows range size, not price growth. It can rise during heavy selling as well as buying. Direction comes from prices, not from ATR alone.
02Does an ATR-based stop guarantee a known loss?+
No. It only sets the chosen rule for order distance. Actual execution, quantity, fees and slippage can change the outcome. An average of past ranges is not an upper bound on future movement either.