Definition
The default ATR setting is a length of 14 with Wilder smoothing. Average True Range (ATR) averages the size of recent candles in price units, counting any gap from the last close, and says nothing about direction. J. Welles Wilder Jr. introduced it in 1978.
A rising ATR means bigger candles, whether price is climbing or sliding. Use it to decide how much room a stop needs on this coin and this timeframe.
How it works
Start with the true range of a single candle, the largest of three numbers:
- the candle's high minus its low;
- the high minus the previous close, taken as a positive number;
- the low minus the previous close, taken as a positive number.
The last two exist for gaps, where a candle opens away from the last close. Crypto trades around the clock, so a candle usually opens near the last close and high minus low wins, though a thin coin can still jump.
ATR then averages true range over N candles. The standard method is Wilder's smoothing, which platforms often label RMA:
ATR now = ((N minus 1) x previous ATR + current true range) / N
Each new candle gets a weight of 1 in N, so the line moves in small steps. Wilder's method starts the series with a simple average of the first N true ranges. A platform that seeds it another way shows slightly different values over the early history, and the difference fades as candles are added. The live candle is still changing, so read the last completed one. For arithmetic only, assume a previous close of 100 and a current candle with a high of 106 and a low of 101. The three values are 5, 6 and 1, so true range is 6. If ATR(14) was 4, the new value is (13 x 4 + 6) / 14 = 58 / 14, about 4.14.
That smoothing makes ATR slow. A candle three times the normal size, a true range of 12 against an ATR of 4, lifts ATR(14) only to about 4.57, a rise of 14%. After a shock candle the number can understate what the coin just did, so read the candles beside it.
Comparing coins and timeframes
ATR is in price units, so a raw value means something only for one coin on one timeframe. Divide ATR by price to get ATR percent. Assume a coin at 50 with an ATR of 2: that is 4% of price. Assume a coin at 60,000 with an ATR of 300: that is 0.5%. The second coin has the bigger number and the smaller average candle. A 4-hour ATR and a daily ATR for the same coin are different numbers too, so read ATR on the chart you trade.
From ATR to a stop buffer
Multiply ATR by a number you choose and add the result beyond a chart level. Assume a short that is wrong above 100, with ATR(14) at 2 and a planned loss of 100. A buffer of half an ATR puts the stop at 101. If the entry is 98, the stop is 3 away and the position can hold about 33 units. A buffer of 1.5 ATR puts the stop at 103, 5 away, and the position shrinks to 20 units. The wider buffer tolerates larger swings and costs 40% of the size.
ATR sets the room around a level. It does not find the level, which comes from the chart.
Default settings
| Setting | Length | Smoothing | When traders use it |
|---|---|---|---|
| Published default | 14 | Wilder (RMA) | Wilder's definition and the default length on the major charting platforms. |
| Variation: shorter length | 7 to 10 | Wilder (RMA) | Reacts sooner to a change in volatility and draws a noisier line. |
| Variation: longer length | 20 to 21 | Wilder (RMA) | Smoother and slower, so stop distances change less from candle to candle. |
| Variation: other smoothing | 14 | Simple, exponential or weighted | A platform preference. Values can differ from Wilder's, so keep one choice when you compare. |
Plain ATR has no multiplier. The number you multiply it by belongs to the tool or rule that uses it, and the Chandelier Exit brings its own published pair, 22 and 3. For an ATR-only stop, 1.5 to 3 times ATR(14) from entry is a commonly cited range, and this page names no single source for it. The examples here assume 0.5 and 1.5 for arithmetic, and neither is a tested value. A multiplier near 1 puts the line where ordinary candles reach it. At 5 or more, a stop hit costs a large move and forces a small position.
How it is used in this manual's setups
ATR sizes the room around a stop. It does not choose entries or predict direction.
- Trailing exits. The Chandelier Exit line is a recent high minus a multiple of ATR for a long, or a recent low plus it for a short, with a published default of ATR(22) and a multiplier of 3. The Chandelier Exit trend ride uses that line as its exit.
- Stop buffers. A setup that adds a buffer beyond its structure level sizes it with ATR so a normal swing is less likely to reach the stop. Stop-loss placement covers where the level itself goes.
- Sizing. A wider ATR stop means a smaller position, and position sizing shows the arithmetic.
ATR also tests a target. If price must travel many times its usual candle range in a few candles to get there, the target is a stretch. The volatility scanner lists the coins moving most right now. A recent surge may not show in the average yet, so check ATR(14) on the chart you trade, which is also the manual route.
Common mistakes
- Reading ATR as a trend signal. It measures candle size only. Take direction from structure or a trend tool.
- Comparing raw ATR across coins. Convert to ATR percent before deciding which coin is more volatile.
- Using ATR from another timeframe. Read it on the chart you trade, so the distance matches that chart's swings.
- An ATR stop with no level behind it. A distance with no chart reason is a guess. Find the level first, then use ATR for the buffer.
- Trusting the number after a shock candle. The average catches up over several candles. Check the live candles too.
- Widening the stop without shrinking the size. Recalculate the position whenever the stop distance changes.
What this cannot tell you
ATR(14) is an average of past candles, so it cannot say how big the next one will be, and after a shock candle it can understate the new volatility. It cannot tell you which way price will move or which side of your stop price reaches first. A stop at 1.5 x ATR with no chart level behind it is only a distance, and ATR cannot say whether that distance marks where the idea is wrong.
Questions
What is the best ATR setting for crypto?
No setting is best. The published default is 14 with Wilder smoothing. A shorter length such as 7 to 10 reacts faster and draws a noisier line, and a length near 20 gives a steadier one. Pick one and keep it across the coins you compare.
What does a high ATR mean?
Candles are larger than usual, so price is covering more ground per period. It says nothing about direction. A stop needs more room on that coin, and the position that fits a given planned loss gets smaller.
How do I use ATR for a stop loss?
Find the level the trade depends on, then add a buffer measured in ATR beyond it. Size the position from the full distance to the stop, buffer included.