Overview
Average True Range (ATR) answers a single, practical question: how much does this market typically move in one bar? It is a pure measure of volatility — the size of the move — and says nothing about direction. That makes it one of the most useful numbers a trader owns, because the right stop distance and the right position size both depend on how much a market is moving right now.
Volatility is not constant. It expands and contracts in clusters — quiet stretches and stormy ones — and ATR is the standard way to put a number on the regime you're currently in.
Origins & history
- 1978J. Welles Wilder Jr. introduced ATR — along with the RSI, ADX and Parabolic SAR — in New Concepts in Technical Trading Systems.1
- WHYWilder traded commodities, where gaps and limit moves are common. A simple high-minus-low range misses the jump from yesterday's close — so he defined True Range to capture it.1
How it works
True Range is the greatest of three distances — it deliberately includes the overnight gap:
The first ATR is a simple average of the first 14 true ranges; after that Wilder smooths it: ATR = (prior ATR × 13 + current TR) ÷ 14. The result is a stable, slow-moving line that rises when bars get bigger and falls when they shrink — exactly what you want from a volatility gauge.1 ATR is quoted in the market's own units (dollars, points, ticks); to compare across instruments, traders often divide it by price to get ATR%.
Why volatility clusters
ATR works because volatility is persistent: large moves tend to be followed by large moves, and quiet by quiet. This "volatility clustering" is one of the most reliable empirical facts in markets — quiet regimes and turbulent regimes both tend to last — and it is the basis of an entire family of volatility models (ARCH/GARCH).3 Because today's volatility carries information about tomorrow's, a measure like ATR is genuinely useful for sizing — even though it says nothing about which way price will go. The mechanism is human and structural: fear and forced liquidation feed more movement, while calm, balanced auctions beget more calm.
Honest assessment
Strengths
ATR turns "this market is wild / quiet" into an objective number. It adapts automatically across instruments and regimes, it's the cleanest input for volatility-based stops and position sizing, and — because it captures gaps via True Range — it's honest about overnight risk in a way that high-minus-low is not.
Evidence rating: volatility clustering (the reason ATR is useful for sizing) is strongly supported empirically. ATR itself is a descriptive statistic, not a directional edge — it tells you how big the moves are, never which way.
Weaknesses & failure modes
- NO DIRECTIONIt cannot call direction. A high ATR means big moves are likely — up or down.
- LAGIt lags. As a smoothed average, ATR reacts after volatility has already changed.
- UNITSAbsolute units mislead. An ATR of 2 is huge for a \$20 stock and tiny for a \$2,000 one — normalise to ATR% to compare.
- SHOCKSOne outlier distorts it. A single limit move or gap can inflate ATR for many bars.
Professional uses vs. retail misuses
How professionals use it
- Volatility stops — place the stop a multiple of ATR beyond entry (e.g., 2–3× ATR) so normal noise doesn't stop you out.
- Position sizing — size so that a fixed dollar risk equals a chosen ATR multiple (see risk & position sizing).
- Regime read — is volatility expanding or contracting? Scale expectations and size accordingly.
Common retail misuses
- Treating a high or low ATR as a buy/sell signal — it is neither.
- Using a fixed dollar stop regardless of how much the market is moving.
- Comparing raw ATR across very different-priced instruments.
Going deeper
ATR stops: a trailing stop set a multiple of ATR below the highest close since entry (the "Chandelier Exit") keeps you in trends while respecting current volatility. ATR position sizing: shares = dollars-at-risk ÷ (ATR-multiple stop distance) — so quieter markets earn larger size and wild ones smaller, holding risk constant. Relatives: Bollinger Bands measure volatility with standard deviation; Keltner Channels and Supertrend are built directly on ATR. ATR and standard deviation usually agree on the regime but differ in detail — ATR weights the gap, standard deviation weights dispersion around the mean.
Practice
Quiz 1 — ATR jumps sharply. Is that a signal to buy or sell?
Neither. ATR measures the size of moves, not direction. A rising ATR says bigger moves are likely — but up or down is a separate question entirely.
Quiz 2 — Why did Wilder use True Range instead of just High − Low?
To capture gaps. True Range includes the distance from the previous close, so an overnight jump (common in commodities) isn't ignored. It's the greatest of: High−Low, |High−PrevClose|, |Low−PrevClose|.
Quiz 3 — Stock A has ATR 2 at price \$20; stock B has ATR 2 at price \$400. Which is more volatile?
Stock A, by far. In ATR% terms A is moving ~10% per bar while B is ~0.5%. Always normalise ATR to price before comparing instruments.
This concept in the knowledge graph
Resources
- CONCEPTRisk & position sizing (where ATR earns its keep) & Bollinger Bands.
- TRADERJ. Welles Wilder Jr. — the creator of ATR.
- GLOSSARYVolatility.
References (primary where possible)
- J. Welles Wilder Jr., New Concepts in Technical Trading Systems (1978) — origin of ATR and True Range — overview.
- Average True Range — formula & Wilder smoothing — StockCharts ChartSchool.
- Volatility clustering & ARCH/GARCH models — overview.