How to Use Average True Range (ATR): Beginner’s Guide

Average true range (ATR) measures how much an instrument typically moves in a period, in the instrument’s own units. It says nothing whatsoever about direction. That single fact is the most misunderstood thing about it: a rising ATR does not mean prices are rising, it means the moves are getting bigger, up or down. Everything useful about ATR follows from treating it as a ruler rather than a signal.

What follows is the formula with the arithmetic actually worked through, then the practical applications, then an honest section on where ATR misleads you. If you are coming from chart patterns, ATR pairs naturally with the reading process described in how to interpret candlestick patterns, because it puts a number on whether a candle’s range was unusual.

What ATR actually measures

ATR answers one question: over the last N periods, how far has this thing typically travelled per period? If a stock has a 14-day ATR of $0.98, an average day covers about a dollar of range. If a different stock has an ATR of $6.40, an average day covers six and a half dollars.

Three consequences fall out of that definition immediately. ATR is always positive, so it has no bullish or bearish reading. ATR is expressed in the instrument’s price units, so $0.98 of ATR means something completely different on a $50 stock than on a $600 stock. And ATR is a smoothed average of the recent past, so it describes the volatility regime you have just been in, not the one you are about to enter.

Where ATR came from

ATR was developed by J. Welles Wilder Jr. and introduced in his 1978 book New Concepts in Technical Trading Systems, the same book that introduced RSI, the Directional Movement Index and the Parabolic SAR. Wilder was working primarily with commodity futures, and the design reflects that: commodities gapped and hit limit moves, so a range measure based only on the high minus the low would systematically understate how far price had actually moved. The True Range definition exists specifically to capture gaps. StockCharts’ ChartSchool entry on ATR documents both the attribution and the calculation.

The formula, worked through

True Range is the greatest of three values

TR = the greatest of:
       (1) High - Low
       (2) abs( High - PreviousClose )
       (3) abs( PreviousClose - Low )

First ATR   = simple average of the first 14 TR values
Each ATR    = ( PreviousATR x 13 + CurrentTR ) / 14

Values (2) and (3) are the gap-catching terms. On a day with no gap, the previous close sits somewhere inside the day’s range, so the high minus the low is the largest of the three and True Range equals the ordinary range. On a day that gaps up, the previous close sits below the day’s low, so (2) becomes the largest and True Range stretches back to include the gap.

A worked 14-period example

Assume the day before this table began closed at $50.00. Every True Range below is the largest of the three columns to its left.

So this stock’s 14-day ATR is about $0.98 at a price near $51.70, and days 4 and 11 show the gap terms doing their job: on both, the ordinary high minus low was only 0.80 and 0.90, but True Range came out at 1.55 and 1.35 because price had gapped up away from the previous close.

Wilder’s smoothing versus a simple average

The first ATR value is a plain average, but every subsequent one uses Wilder’s recursive smoothing. Suppose day 15 runs 52.20 high, 51.00 low, 51.15 close, after a previous close of 51.70. Its True Range is 1.20 (the high minus the low, since the gap terms give only 0.50 and 0.70).

Wilder smoothing:
  ATR = ( 0.9786 x 13 + 1.20 ) / 14
      = ( 12.7214 + 1.20 ) / 14
      = 13.9214 / 14
      = 0.9944          ->  $0.99

Simple 14-day average of TR (days 2 to 15):
  ( 13.70 - 0.90 + 1.20 ) / 14
      = 14.00 / 14
      = 1.0000          ->  $1.00

A one-cent difference here, but the mechanism matters. Wilder’s version never fully drops an old value, it only decays it, so it reacts more slowly and produces a smoother line. A simple moving average of True Range drops day 1 entirely the moment day 15 arrives, which makes it jumpier. Most platforms label the Wilder version simply “ATR”; some offer an RMA, SMA or EMA choice, and the values will differ slightly. Pick one and stay with it.

Why the default is 14

Because Wilder used 14 for most of the indicators in the 1978 book, and every charting package inherited it. There is no deep statistical justification. Shorter periods react faster and are noisier; longer periods are stabler and lag more. If you change it, change it because you have tested the change, not because a shorter number felt more responsive.

What ATR is actually good for

1. Volatility-based stop losses

The common construction is a stop placed a multiple of ATR away from entry, for example 2 x ATR below a long entry. In the example above that is $51.15 minus $1.98, so a stop at $49.17.

The reason this beats a fixed percentage stop is that a fixed percentage means different things on different instruments. A 3% stop on our example stock is $1.53, or roughly 1.5 ATRs, which is a reasonable amount of room. The same 3% stop on a quiet utility with an ATR of 0.6% of price is about five ATRs away, which is far more risk than intended. On a volatile small cap whose ATR is 5% of price, 3% is less than one ATR, meaning an average day’s noise will take you out. One number, three completely different levels of tightness. ATR normalizes that.

2. Position sizing so dollar risk stays constant

This is the application most worth learning, because it lets you trade instruments of wildly different volatility with the same risk per idea.

Shares = ( Account equity x Risk per trade ) / ( ATR multiple x ATR )

Example A: quieter stock
  Equity $25,000, risk 1% = $250
  Price $51.15, ATR $0.99, stop 2 x ATR = $1.98
  Shares = 250 / 1.98 = 126.3  ->  126 shares
  Exposure = 126 x 51.15 = $6,444.90
  Risk     = 126 x 1.98  = $249.48

Example B: same price, much more volatile
  Price $51.15, ATR $2.50, stop 2 x ATR = $5.00
  Shares = 250 / 5.00 = 50 shares
  Exposure = 50 x 51.15 = $2,557.50
  Risk     = 50 x 5.00  = $250.00

Same account, same risk, same price per share, and yet one position is two and a half times the size of the other. That is correct and it is the entire point. Volatility, not price, determines how much you can hold. Notice also that the risk figure is fixed by design while the exposure floats, which is the opposite of how most beginners size positions.

3. Trailing stops and profit targets

ATR-based trailing stops ratchet an exit level along behind price, widening in volatile conditions and tightening in calm ones. The best-known formulation is the Chandelier Exit, developed by Charles Le Beau: for a long position, the stop sits at the highest high of the last 22 periods minus 3 x ATR(22), and it only ever moves up. There is a fuller treatment in the guide to using the Chandelier Exit for swing trading.

Targets work the same way in reverse. Setting a target at 2 or 3 ATRs of profit is a way of asking for a move that is large relative to normal daily movement, rather than picking a round dollar figure. Be aware of the arithmetic, though: a 1 x ATR target is hit often and pays little, and a 4 x ATR target pays well and is hit rarely. Neither is better; they just imply different win rates.

4. Filtering dead and frantic markets

Comparing current ATR to its own recent range tells you which regime you are in. If ATR has fallen to the bottom of its six-month range, ranges are compressed, breakout attempts tend to fail and the reward available on any trade is small relative to costs. If ATR has spiked to the top of that range, stops must be much wider, which means positions must be much smaller, and a setup that looked attractive at normal volatility may not be worth taking at all.

Used this way ATR is a permission check rather than a trigger, which pairs well with pressure-based readings such as the Elder Ray Index.

5. Comparing volatility across instruments

Here is the trap. ATR is absolute, so it cannot be compared between instruments at different price levels. Our $51.70 stock with $0.98 of ATR is about 1.9% of price per day. A $600 stock with a $6.00 ATR has six times the absolute ATR but only 1.0% of price, so it is actually the calmer of the two in relative terms.

To compare, normalize. ATR percent (sometimes ATRP or normalized ATR) is simply ATR divided by the close, times 100. Use raw ATR when you are calculating stop distances and share counts in one instrument, and ATR percent whenever you are ranking or comparing instruments.

Choosing an ATR multiple

Adding ATR to a chart

Every mainstream charting platform ships ATR, so this is generic on purpose. Open the indicator or study list, search for “Average True Range” or “ATR”, and add it. It appears in a separate pane below price because its scale is dollars of range, not price. The one input that matters is the period, which will default to 14.

Two things worth checking in whichever platform you use, because they differ and both change the numbers:

1. The smoothing method. Some platforms expose a choice between Wilder’s smoothing (often labeled RMA), a simple average and an exponential one. The classic ATR is Wilder’s.
2. Whether extended-hours data is included. On US stocks, including pre-market and after-hours prices changes the highs, lows and gaps, and therefore changes ATR. Two platforms can legitimately show different ATRs for the same symbol on the same day for this reason.

The limits of ATR

It is entirely backward-looking. ATR is an average of what already happened. It does not anticipate the earnings report tomorrow morning or the macro release on Friday. Volatility clusters, which is why the recent average has some usefulness, but a low ATR is not a forecast of calm.

Gaps inflate it, and then linger. The True Range definition deliberately captures gaps, which is right, but one large gap raises ATR and, because of the recursive smoothing, keeps it elevated for many periods after the event is over. Around earnings dates, ATR-derived stops are often much wider than current conditions justify.

It says nothing about direction. Worth repeating because people persist in reading it as a signal. A rising ATR in a downtrend and a rising ATR in an uptrend look identical. Pair it with something directional; on its own it can only tell you how much room to give a position.

It needs rescaling across timeframes. A 14-period ATR on a 5-minute chart measures a completely different quantity from a 14-period ATR on a daily chart. If you analyze on daily bars and execute on 15-minute bars, the ATR you use for the stop must come from the timeframe your holding period matches, which is usually the higher one.

It is a risk tool, not an edge. ATR will not tell you what to trade. Its honest contribution is making your risk consistent, which is the same principle behind spreading capital across uncorrelated things generally, whether that is index funds or property in different state markets. The SEC’s investing basics guidance is a reasonable reality check on where active trading sits in a financial plan.

Common mistakes

1. Reading ATR as bullish or bearish. It has no sign. Rising ATR means bigger moves, nothing more.
2. Comparing raw ATR across instruments. Use ATR as a percentage of price for any comparison or ranking.
3. Choosing a multiple to fit a desired position size. This inverts the logic. Set the multiple from the strategy, then accept the size it produces.
4. Recalculating the stop every day on an open position. If ATR contracts, a recalculated fixed stop moves closer to price and can eject you from a fine trade. Either fix the stop at entry or use a proper trailing rule that only moves in your favor.
5. Forgetting that ATR is still elevated after a gap. Check whether one event is dominating the average before sizing off it.
6. Mixing timeframes. A daily-chart thesis needs a daily-chart ATR.

Frequently asked questions

What is a good ATR value?

There is no good or bad ATR value, because it is measured in the instrument’s own currency units. The only meaningful comparisons are ATR against that same instrument’s own history, and ATR as a percentage of price when comparing different instruments. A $2 ATR is enormous on a $20 stock and trivial on a $2,000 one.

Does a high ATR mean the price will go up?

No. ATR is directionless by construction, because True Range uses absolute distances. High ATR means large moves are happening, and those moves can be in either direction. Reading it as a bullish signal is the single most common error with this indicator.

What ATR period should I use?

Start with 14, the default that comes from Wilder’s original work. Shorter periods such as 7 respond faster to a change in conditions but produce noisier stop distances; longer periods such as 20 or 22 are smoother and slower. Change it only if testing shows the change helps your specific rules.

Is ATR better than a fixed percentage stop?

For trading across multiple instruments, generally yes, because a fixed percentage is a different amount of room on every instrument depending on its volatility. ATR gives each position comparable breathing space. For a single instrument you always trade, the difference is much smaller.

How is ATR different from standard deviation or Bollinger Bands?

ATR averages the full high-to-low range including gaps, so it captures intraperiod movement. Standard deviation, which drives Bollinger Bands, measures the dispersion of closing prices around their mean. A market that swings wildly intraday but closes near the same level daily will show high ATR and low standard deviation.

Can I use ATR with candlestick patterns?

Yes, and it is one of the better pairings, because they answer different questions. The pattern suggests a direction to consider and a level that invalidates it; ATR tells you whether that candle’s range was unusual and how far away a stop needs to sit. Start with the basics of candlestick patterns if you need the vocabulary.

Wrapping up

Learn ATR as a measuring device and it will earn its place permanently. Its three real jobs are setting a stop distance that reflects how much the instrument actually moves, converting that distance into a position size so every trade risks the same dollars, and trailing an exit that adapts to conditions. Those are risk-management jobs, and they are where most of the difference between traders actually shows up.

What ATR will never do is tell you what to buy or when. If you catch yourself looking for direction in it, you have the wrong tool out. Add ATR to your charts, get comfortable reading it as “an average day is about this big,” and pair it with whatever directional method you are testing, whether that is trend structure or the reading process in interpreting candlestick patterns.