ATR indicator explained: Average True Range, ATR stop losses and volatility-adjusted sizing
ATR measures how far a market has typically moved per bar, nothing more. Here is how true range and Wilder smoothing work, how traders use ATR multiples for stops, and how ATR feeds into position size.
- ATR (Average True Range) measures how much a market has moved per bar on average; it says nothing about direction.
- True range is the largest of high minus low, high minus previous close, and low minus previous close, so it counts gaps.
- An ATR stop places the stop a multiple of ATR from entry; 1.5 to 3 times ATR are commonly used ranges, not settings to copy.
- A wider ATR stop needs a smaller position to keep the same money at risk; ATR changes your size, not your risk.
- There is no good ATR in absolute terms: compare a market only with its own history and with your stop and costs.
The ATR indicator (Average True Range) measures how much a market has moved per bar, on average, over a recent window, usually 14 bars. If EUR/USD has a daily ATR of 60 pips, it has moved about 60 pips a day recently, in either direction. Traders mainly use ATR to set stop-loss distances that fit current volatility and to size positions so that a wider stop does not mean a bigger loss.
ATR is one of the least glamorous indicators on the chart and one of the most practical. It does not claim to tell you where price is going. It tells you how much room the market has been taking, which is exactly the information you need to decide where a stop belongs and how big a position you can afford.
What ATR measures, and what it does not
ATR is a volatility measure. It looks at the size of each bar, including any gap from the previous close, and averages those sizes. A rising ATR means bars have been getting bigger; a falling ATR means they have been getting smaller.
What ATR does not do:
- It has no direction. A market falling 80 pips a day and one rising 80 pips a day have the same ATR. ATR cannot tell you whether a trend is up or down, or whether one exists at all.
- It does not forecast volatility. ATR describes the recent past. Volatility does tend to cluster, so quiet periods often follow quiet periods and busy ones follow busy ones, but a scheduled central bank decision or a surprise headline can make the next bar several times larger than the average.
- It is not a target. A daily ATR of 60 pips does not mean price will move 60 pips today, or that a move of 60 pips means the day's range is "used up". Some days are a third of ATR, others are three times it.
If you are comfortable with the idea that indicators repackage past price rather than add new information, ATR fits that description well. The difference from momentum tools such as those in RSI and MACD explained is that ATR does not pretend to say anything about direction, so it is harder to misuse as a signal.
Where ATR comes from
ATR was introduced by J. Welles Wilder in his 1978 book New Concepts in Technical Trading Systems, the same book that introduced RSI. Wilder developed it for commodity futures, where limit moves and overnight gaps were common and a simple high-minus-low measure missed a lot of the real movement. His default look-back was 14 periods, and that is still the default on MetaTrader, cTrader and TradingView.
How ATR is calculated
Step 1: true range
For each bar, true range is the largest of three distances:
- The current high minus the current low.
- The current high minus the previous close (as an absolute value).
- The current low minus the previous close (as an absolute value).
On most bars the first distance wins: the true range is just the bar's own range. The other two matter when the market gaps. If EUR/USD closes Friday at 1.0895 and opens Monday at 1.0860, with Monday's whole range between 1.0860 and 1.0810, the high-to-low range is 50 pips, but the market actually travelled 85 pips from Friday's close to Monday's low. True range records 85. That is the point of the measure: it captures the movement a position held overnight or over a weekend would actually have experienced. For more on why gaps matter for stops, see gap risk and weekend risk.
Step 2: averaging
ATR is an average of the true ranges. Wilder used his own smoothing method rather than a simple average. The first ATR value is the simple average of the first 14 true ranges. After that, each new value is:
- new ATR = ((previous ATR × 13) + current true range) / 14
This works like an exponential moving average: every past bar still has some weight, but the weight fades over time, and a single large bar moves the ATR noticeably without dominating it. Some platforms let you choose a simple, exponential or Wilder-style average; the numbers differ slightly but the idea is the same. When you compare ATR readings between platforms, check which averaging method each one uses.
ATR is in price units
ATR is expressed in the same units as price. On EUR/USD a reading of 0.0058 is 58 pips. On USD/JPY a reading of 0.85 is 85 pips. On gold a reading of 45.00 means 45 dollars per ounce. That is useful for setting stops, because the number translates directly into a distance on the chart, but it means you cannot compare raw ATR values across instruments. More on that below.
Worked example: calculating ATR from five candles
Illustrative numbers only, using EUR/USD daily candles and a short five-bar window so the arithmetic stays readable. The close before the first bar is 1.0850.
- Bar 1: high 1.0890, low 1.0830, close 1.0870. High minus low = 60 pips. High minus previous close (1.0850) = 40 pips. Low minus previous close = 20 pips. True range = 60 pips.
- Bar 2: high 1.0905, low 1.0855, close 1.0895. High minus low = 50. High minus previous close (1.0870) = 35. Low minus previous close = 15. True range = 50 pips.
- Bar 3 (gap lower): high 1.0860, low 1.0810, close 1.0820. High minus low = 50. High minus previous close (1.0895) = 35. Low minus previous close = 85. True range = 85 pips, because of the gap.
- Bar 4: high 1.0850, low 1.0800, close 1.0845. High minus low = 50. High minus previous close (1.0820) = 30. Low minus previous close = 20. True range = 50 pips.
- Bar 5: high 1.0880, low 1.0835, close 1.0875. High minus low = 45. High minus previous close (1.0845) = 35. Low minus previous close = 10. True range = 45 pips.
Sum of true ranges: 60 + 50 + 85 + 50 + 45 = 290 pips. Simple average over five bars: 290 / 5 = 58 pips.
Notice that a plain average of high minus low would have been 51 pips (60 + 50 + 50 + 50 + 45 = 255, divided by 5). The gap on bar 3 added 7 pips to the average. On a market that gaps often, the difference between range and true range can be much larger.
To see Wilder smoothing in action, suppose instead that a 14-period ATR currently reads 58 pips and the next bar has a true range of 40 pips. The new ATR is ((58 × 13) + 40) / 14 = 794 / 14 = 56.7 pips. One quiet day nudges the average down by about 1.3 pips; it takes a run of quiet days to pull it down meaningfully.
The ATR stop loss
The most common use of ATR is the ATR stop: placing the stop-loss a multiple of ATR away from the entry price, or a multiple of ATR beyond a chart level. The logic is simple. If a market routinely moves 58 pips a day, a 20-pip stop on a trade you plan to hold for several days sits well inside normal movement and is likely to be hit by noise. A stop of two or three times the daily ATR gives the trade room for ordinary fluctuations.
ATR multipliers in common use
These are ranges you will commonly see in trading books and on forums, not recommendations:
- Around 1 × ATR or less: tight. The stop is within a single bar's typical movement and will often be hit before the idea has had a chance to play out.
- 1.5 to 2 × ATR: a frequently quoted middle ground, often mentioned for swing trades.
- 2.5 to 3 × ATR or more: wide, often associated with longer-term trend following, where the aim is to sit through normal pullbacks.
No multiple is correct in general. The right one depends on the timeframe, the holding period, the instrument and the method, and the only honest way to choose one is to test it on historical data and then on a small live sample. Our guide to how to backtest a strategy explains how to do that without fooling yourself.
Combining ATR with chart structure
ATR does not know where the obvious levels on the chart are. A stop placed exactly 2 × ATR below entry might sit just above a well-watched swing low, which is a poor place for it. Many traders therefore use structure first and ATR second: find the level that would show the trade idea is wrong, then use ATR to decide how much buffer to add beyond it, and to check that the total distance is sensible. Our guide on where to put a stop-loss works through that combination with examples, and support and resistance covers how to identify the levels themselves.
ATR trailing stops
ATR is also used to trail a stop behind a winning trade. A common form, often called a chandelier-style exit, keeps the stop a multiple of ATR below the highest high since entry for a long trade (or above the lowest low for a short). As price makes new highs, the stop rises; if volatility increases, the distance widens. Trailing stop losses explained compares ATR trailing with fixed-pip and swing-point methods.
Remember that any stop becomes a market order once triggered. In a fast market or over a gap, it can be filled beyond its level, sometimes far beyond. ATR helps you set a sensible distance; it does not guarantee the exit price.
ATR position sizing: volatility-adjusted size
This is where ATR earns its keep. A wider stop on the same position size means a bigger loss if the stop is hit. To keep the money at risk constant, the position has to shrink as the stop widens. The standard position-size formula is:
- position size (lots) = money at risk / (stop distance in pips × pip value per lot)
With an ATR stop, the stop distance is ATR × your chosen multiple, so the size adjusts automatically to volatility. Quiet market, tighter stop, larger size; volatile market, wider stop, smaller size; the same money at risk either way. Our position sizing guide explains the formula in detail, and the position size calculator does the arithmetic for any pair and account currency.
Worked example: 2 × ATR stop and position size
Illustrative numbers. A trader with a USD 10,000 account limits risk to 1% per trade, which is USD 100. They are looking at EUR/USD, where one standard lot is worth about USD 10 per pip for a US dollar account.
Using the ATR of 58 pips from the five-candle example:
- Stop distance = 2 × 58 = 116 pips.
- Money at risk per standard lot if the stop is hit = 116 × USD 10 = USD 1,160.
- Position size = USD 100 / USD 1,160 = 0.0862 lots.
- Round down to the broker's lot step: 0.08 lots (8 micro lots).
- Actual risk = 0.08 × USD 10 × 116 = USD 92.80, just under the USD 100 limit.
Now compare two other volatility conditions with the same rule and the same USD 100 limit:
- Quieter market, ATR 40 pips: stop 80 pips, size 100 / (80 × 10) = 0.125, rounded down to 0.12 lots, risk USD 96.
- Busier market, ATR 90 pips: stop 180 pips, size 100 / (180 × 10) = 0.0556, rounded down to 0.05 lots, risk USD 90.
The stop distance more than doubled between the quiet and the busy market, but the loss if the stop is hit stayed between USD 90 and USD 96. That is volatility-adjusted sizing. It does not make the trade more likely to work; it makes the cost of being wrong roughly constant, which is what keeps a losing streak survivable. If you want to see how a sequence of fixed-percentage losses affects an account, the drawdown and recovery guide shows the maths.
Two cautions. First, rounding down matters: rounding up even once in a while quietly raises your risk. Second, if a stop is wide enough that even the minimum lot size breaches your risk limit, the honest answer is to skip the trade or use a smaller instrument size, not to tighten the stop until the numbers fit.
ATR on gold, indices and other markets
ATR works the same way on any instrument with a high, low and close, but the units change, and some markets need a little extra care.
Gold (XAU/USD)
Gold's ATR is quoted in dollars per ounce. Because brokers disagree on what a gold pip is, it is cleaner to think in dollars than in pips. Illustrative example: if gold's daily ATR is 45 dollars and you use a 2 × ATR stop, the stop is 90 dollars from entry. On a contract of 100 ounces per lot, that is USD 9,000 per standard lot if the stop is hit. With a USD 100 risk limit, the size works out at 100 / 9,000 = 0.011 lots, so even the common minimum of 0.01 lots (one ounce per dollar, so USD 90 for this stop) uses almost the whole budget. Daily ATR stops on gold can be hard to fit into a small account at 1% risk. Our gold CFD trading guide covers contract sizes and pip conventions, and the gold lot size calculator handles the arithmetic.
Stock indices
Index CFDs are quoted in index points, so ATR is in points too. Indices often gap at the cash open and after weekends, so true range matters more than on major currency pairs during the main session. Some brokers' index CFDs trade nearly around the clock while the underlying cash market does not, which changes how bars and gaps appear on the chart. The index CFD position sizing guide shows how points convert into money per lot.
Crypto
Crypto markets trade every day, so a daily ATR on a crypto CFD includes weekends, while a forex daily chart usually does not. Daily ATR as a percentage of price is often much higher on major cryptocurrencies than on currency pairs, which pushes ATR-based position sizes down accordingly.
What is a good ATR?
There is no good ATR in absolute terms, because ATR is in price units and depends on the instrument, its price level and the timeframe. A 58-pip daily ATR on EUR/USD and a 45-dollar daily ATR on gold cannot be compared directly. Three more useful questions:
- Is ATR high or low for this market? Compare today's reading with the same market's ATR over the past few months or years. A reading well above its own history means the market is moving more than usual, which usually means wider stops and smaller positions.
- What is ATR as a percentage of price? Dividing ATR by price lets you compare markets roughly. Illustratively, 58 pips on EUR/USD at 1.0850 is about 0.53% of price; 45 dollars on gold at 4,000 is about 1.1%.
- How does ATR compare with your costs and stop? On a 5-minute chart, ATR might be 3 or 4 pips on a major pair. If the spread and commission add up to 1 pip, costs are a large share of every bar's movement. On a daily chart the same costs are a small share. This is one reason very short timeframes are harder than they look; see how to compare trading costs.
Timeframe changes ATR in a predictable direction. A daily ATR is always larger than an hourly ATR on the same market, but not 24 times larger, because intraday moves partly cancel out. Pick the ATR on the timeframe you actually trade and hold.
Other uses of ATR
- Filters. Some systematic traders avoid new entries when ATR is unusually low (little movement to work with relative to costs) or unusually high (stops would need to be very wide).
- Breakout context. A move of a quarter of ATR beyond a level is ordinary noise; a move of two ATRs in one bar is not. ATR gives a yardstick for how unusual a bar is.
- Comparing periods. A rising ATR around scheduled events, such as those on the economic calendar, shows how much more room the market takes in those weeks.
- Exits and targets. Some traders set profit targets in ATR multiples. That keeps targets proportional to volatility, but it is still a rule to test rather than a reason to expect price to reach it.
Limits of ATR
- It lags. Like any average, ATR reacts after volatility has already changed. On the day of a large surprise, ATR barely moves; it catches up over the following bars. A stop based on yesterday's ATR can be too tight for today's news.
- It treats up and down moves the same. A sharp rally and a crash look identical to ATR.
- It depends on the data. Different brokers' feeds, session times and weekend handling produce slightly different bars and therefore slightly different ATR values. A broker whose daily candle closes at New York 5pm will show different daily ranges from one that closes at midnight UK time.
- It ignores scheduled events. ATR knows nothing about the central bank meeting tomorrow. Spreads can also widen sharply around releases; see why spreads widen.
- It does not improve a bad idea. A well-sized trade on a poor setup is still a poor trade. ATR controls the size of each loss, not the number of them.
Common mistakes
- Treating ATR as directional. Rising ATR is not bullish or bearish. It means bigger bars.
- Copying a multiple without testing it. "2 × ATR" is a convention, not a law. On some methods and timeframes it is too tight; on others, too wide.
- Widening the stop without shrinking the position. This is the single most expensive error. If the ATR stop doubles, the size must roughly halve, or the risk doubles.
- Mixing timeframes. Using a daily ATR to set a stop on a trade you manage on the 5-minute chart, or the reverse, puts the stop in the wrong place for how you hold the trade.
- Ignoring gaps and slippage. ATR describes typical movement. Weekend gaps and fast news markets can carry price straight through a stop.
- Forgetting leverage. Volatility-adjusted sizing helps, but leverage still magnifies losses, and most retail CFD accounts lose money. See leverage and margin and why most retail CFD traders lose.
For a short definition to bookmark, see the glossary entry for ATR and the related entry for volatility.
Frequently asked questions
What is the ATR indicator?
ATR, or Average True Range, is a volatility indicator that averages the true range of recent bars, usually over 14 periods. It shows how much a market has typically moved per bar, in price units, without saying anything about direction.
How do you use ATR for a stop loss?
Multiply the current ATR by a chosen factor and place the stop that distance from entry, or that far beyond a chart level that would show your idea is wrong. Multiples between 1.5 and 3 are commonly used, but the right figure depends on your timeframe and method and should be tested before you rely on it.
What is the best ATR setting?
Wilder's original 14 periods remains the default on most platforms and is a reasonable starting point. A shorter period reacts faster to changes in volatility but is noisier; a longer one is smoother but slower. No setting is best in general.
How does ATR help with position sizing?
Once ATR sets the stop distance, you divide the money you are willing to risk by the stop distance multiplied by the pip value per lot. A wider ATR stop gives a smaller position, so the loss if the stop is hit stays roughly the same in money terms.
What is a good ATR value?
There is no good value in absolute terms. Compare a market's ATR with its own history to see whether it is unusually quiet or busy, divide by price to compare markets roughly, and check how ATR compares with your trading costs on your timeframe.
Does a high ATR mean price will reverse?
No. A high ATR means bars have been large recently. Markets can stay volatile for long periods, and volatility can be high in a strong trend as well as at a turning point. ATR describes the size of movement, not its direction.
Not investment advice. This is general information and education, written without knowledge of your circumstances. It is not a recommendation to buy or sell any instrument. Past performance is not a reliable indicator of future results. Trading leveraged products carries a high risk of losing money. Risk warning