What Is Average True Range (ATR), and How Do Traders Use It?
Average True Range (ATR) is a single number that measures how much a stock actually moves in a normal session, averaged over a recent stretch of days — not the direction, just the size of the move. Traders use it to size stops and targets in units the market itself sets, instead of a round number that means something different on every ticker.
That's the whole answer. The rest of this page is the math behind it, why "true" range isn't the same as the high-minus-low you'd guess, and where ATR actually earns its keep versus where people lean on it too hard.
What does "true range" actually measure?
Most people's first guess at a day's range is high minus low. That's close, but it misses the case where the stock gapped overnight — the move that happened between yesterday's close and today's open never shows up in today's high-low spread at all.
True range fixes that by taking the largest of three numbers for a given session:
- Today's high minus today's low
- Today's high minus yesterday's close (captures a gap up)
- Yesterday's close minus today's low (captures a gap down)
Whichever of those three is biggest is the true range for that day. On a normal session with no overnight gap, it's just the high-low spread, same as your first guess. On a session that gapped, it correctly counts the gap as part of the day's real movement instead of ignoring it.
Average True Range is exactly what it sounds like: true range, averaged over a lookback window — 14 sessions is the standard, though there's nothing sacred about that number. The result is one figure, in the stock's own price units, that answers "how much does this thing normally move in a day."
Why not just use yesterday's range, or a round number?
Two reasons, and they compound.
A single day is noisy. Yesterday's range might have been unusually quiet or unusually wild for reasons that have nothing to do with what tomorrow holds — an average smooths that out and gives you a figure that reflects the stock's actual behavior over a real sample, not one session's mood.
A round number ignores the stock entirely. A fixed cents-wide stop is enormous on a low-priced stock and irrelevant on a high-priced one. ATR is denominated in the stock's own price, updated as the stock's behavior changes, so the same logic — "a stop at 1.5× ATR" — produces a completely different, and correctly scaled, number on every name you apply it to. That's the entire case for using it: not that it's smarter than a fixed number, but that it's the same rule expressed in units the stock itself defines.
The long version of this is in the book — Become a Cyclitecnical Trader: the cycle ladder, the FLD, and the eight interactions, written out end to end. It's free. Send me a copy. We email it to you. No card, and you can unsubscribe any time.
How is ATR actually used?
The three places ATR shows up most in a real process:
| Use | What it does |
|---|---|
| Stop distance | Set the stop some multiple of ATR away from entry (1×, 1.5×, 2× are common starting points) so it's wide enough to sit outside the stock's normal daily noise |
| Target distance | Frame a target as a multiple of ATR too, so the reward is measured against the same yardstick as the risk |
| Volatility filter | Compare a stock's current ATR to its own ATR from weeks or months ago to see whether the stock has gotten quieter or louder than its own recent normal |
The stop-sizing use is the one people reach for first, and it connects directly to something we've written about separately: a stop is only one of four dimensions of risk — size, stop, target, and time — and ATR is a tool for setting one of the four, not a complete risk plan on its own.
Does a bigger ATR mean a riskier stock?
Not by itself — it means a bigger-moving stock, which is a different question from risky. A high-priced stock and a much lower-priced one can carry the exact same normal-day move once you express ATR as a percentage of price instead of raw currency. Comparing raw ATR figures across different stocks tells you almost nothing; comparing ATR as a percentage of price puts them on the same footing, and comparing a stock's ATR to its own history tells you whether that specific name has gotten more or less active lately.
That last comparison — a stock's ATR against its own recent past — is the more useful read day to day. A ticker whose ATR has been climbing for two weeks is telling you something changed: bigger swings, wider stops needed, different sizing math. A flattening ATR after a stretch of volatility is often the market settling back into a range. Reading ATR as a trend, not just a snapshot, is where it earns its place next to something like relative volume — one measures how much is trading, the other measures how far price is actually moving because of it.
Does ATR tell you which way price is going?
No, and this is the mistake worth naming directly. ATR has no direction in it at all — it's an absolute value, built from range, with no sign. A stock can have a huge ATR while going nowhere on net (wide, choppy, directionless days that cancel each other out) or a small ATR while grinding steadily in one direction. ATR answers "how big are the swings," never "which way is this going." Direction is a separate question, answered by structure, trend, and — on this desk — where a market sits in its larger cyclic rhythm, not by a volatility reading.
What does ATR get used wrong for?
A few habits worth naming plainly:
- Treating it as a support/resistance level. ATR is a distance, not a price. "Price is 1 ATR below the open" is a valid observation; "ATR is at that reading, so that's a level" is not — there's no price on the chart at an ATR value, only a distance.
- Using one multiple for every setup. A breakout trade and a mean-reversion fade have different reasons to want a stop close or wide; the right ATR multiple depends on the setup, not a single number applied everywhere out of habit.
- Ignoring that ATR itself moves. A stop sized off today's ATR can look wrong two weeks later if volatility has genuinely shifted — it's a live measurement, not a constant, and stale ATR-based stops are a real failure mode.
- Skipping the time dimension entirely. ATR helps size where a stop goes; it says nothing about how long to give a trade before the setup itself has expired, which is the point of a time stop as its own separate decision.
ATR and RSI get confused for each other more than the math suggests they should: ATR measures how big the swings have been, RSI measures how lopsided the gains and losses inside those swings have been. A stock can have a huge ATR and a neutral RSI, or a small ATR and an extreme RSI — they're answering different questions and neither substitutes for the other.
Questions traders ask
What is a good ATR multiplier for a stop loss?
There's no universal number — 1.5× to 2× ATR is a common starting range for a swing setup, tighter for something meant to resolve fast, wider for something meant to survive normal chop. The right multiplier is set by what the specific setup needs to be false-signaled by noise versus stopped out by an actual thesis failure, not by a rule that applies identically everywhere.
Is a 14-period ATR the only setting people use?
It's the standard default and the one most charting platforms preset, but the period is just a lookback window — a shorter one reacts faster to recent changes in volatility, a longer one smooths more and reacts slower. Fourteen is a reasonable, well-tested default, not a law.
How is ATR different from standard deviation of returns?
Both measure volatility, but they measure different things. ATR is built from each day's actual range (including gaps) in price units; standard deviation is typically calculated on percentage returns and describes the dispersion of those returns around their average. ATR answers "how big are the daily swings"; standard deviation answers "how spread out are the day-to-day returns." They tend to move together but aren't interchangeable.
Can ATR predict a breakout?
No — it measures how big recent moves have been, not what's coming next. A falling ATR does sometimes precede an expansion (quiet ranges compressing before a move), which is why some traders watch for ATR contraction as one ingredient in a setup, but that's a pattern to weigh alongside everything else on the chart, not a signal on its own, and it says nothing about direction even when it does precede a bigger move.
ATR is a volatility ruler, nothing more and nothing less: it tells you how big the market's normal noise has been recently, in the stock's own units, so a stop or target can be sized against something real instead of a guess. Where a market sits in its larger cyclic structure — which is what actually informs direction and timing on this desk — is a separate question, covered on the methodology page.
Keep reading
Get the Cycle Pass — from the price on the page