Average True Range (ATR)

Average True Range (ATR) is a volatility indicator that measures how much a price typically moves per period, including jumps between one period and the next. Developed for trading in financial markets, not retail, it can still help merchants measure how unstable a product’s market price or a competitor’s pricing is.

Where ATR comes from

ATR was introduced by J. Welles Wilder Jr. in his 1978 book on technical trading systems. Traders use it to judge how far a stock, commodity or currency usually moves in a day or week, often to size positions or place stop levels. It measures the size of moves, not their direction. A rising ATR means larger swings, whether prices are going up or down.

Retail prices behave differently from traded assets. They change less often, usually in steps, and are set by people or rules rather than by an order book. Treat ATR in pricing as a borrowed tool, useful for comparing volatility, not as a forecast.

The true range formula

Each period has a high, a low and a closing price. The true range (TR) is the largest of three values:

  • High minus low for the current period
  • The absolute difference between the current high and the previous close
  • The absolute difference between the current low and the previous close

The second and third values capture gaps. If a price closes at 50 one day and opens at 57 the next, the simple high-minus-low range ignores that jump, while true range includes it. ATR is then the average of true range over a chosen number of periods, commonly 14. Wilder’s version smooths it: new ATR equals the previous ATR times (n minus 1), plus the latest TR, all divided by n.

A worked example in its original setting

For example, assume an asset with these daily prices, using a 3-period ATR to keep the arithmetic short. Day 1: high 52, low 48, close 50, so TR is 4. Day 2: high 53, low 50, close 51; the largest value is 3, so TR is 3. Day 3: high 55, low 51, close 54; TR is 4. The first ATR is the simple average, (4 + 3 + 4) / 3, about 3.67.

On day 4 the price gaps up: high 60, low 57, close 58. The high-minus-low range is only 3, but the high is 6 above the previous close, so TR is 6. The smoothed ATR becomes (3.67 times 2, plus 6) / 3, about 4.44. The gap raised measured volatility even though the day itself was calm. These numbers are invented to show the method.

Adapting ATR to product prices

To use the idea in retail, define each period as a day or week of price observations. The high and low are the highest and lowest prices seen for the product during that period, and the close is the last observed price. With a long price history, these values can be built for every competitor and product.

For example, assume a competitor’s blender is observed weekly. Week 1: high 89, low 84, last 85, giving a TR of 5. Week 2: high 85, low 79, last 79; TR is 6. Week 3: the price jumps to 92 and stays there all week. The weekly range is 0, but the gap from 79 makes TR 13. Week 4: high 92, low 88, last 88; TR is 4. A simple 4-week ATR is (5 + 6 + 13 + 4) / 4, which is 7 dollars. These figures are assumptions for illustration.

Dividing ATR by the current price gives a percentage that can be compared across products. Here, 7 divided by 88 is about 8 percent per week, which is high for a household appliance.

What the measure tells a pricing team

  • Which competitors reprice aggressively and which keep prices steady
  • Which products sit in unstable markets where frequent checks are worth the effort
  • Whether a market became more volatile after a new competitor arrived or during a sales season
  • How wide a tolerance band a repricing rule needs, so it does not react to normal noise

The last point is practical. If a product’s market price moves about 2 dollars per week on average, a rule that reacts to every 50-cent change will churn prices for no benefit.

Limits and common mistakes

  • Using ATR as a prediction of where prices are going; it only describes past movement
  • Comparing raw ATR values across products with very different prices instead of using a percentage
  • Building it on sparse observations, which can miss changes and understate volatility
  • Ignoring promotions, which create short spikes that inflate ATR for weeks with smoothing
  • Mixing mismatched or out-of-stock listings into the high and low values

Putting it to work

Start with a small group of important products and one consistent period length, such as a week. Calculate ATR and ATR as a percentage of price for each main competitor. Rank products by volatility, then decide how often to review each group and how wide the tolerance band in each pricing rule should be.

Revisit the numbers each quarter. A product whose ATR doubles may have a new aggressive competitor, and that is often worth investigating before the next price change. Keep the period length and the observation schedule fixed, because changing either one changes the result as much as a real shift in the market would.

Key takeaways

  • ATR is a trading volatility indicator created by J. Welles Wilder Jr. in 1978.
  • True range is the largest of high minus low and the two gaps from the previous close.
  • Applied to retail, it measures how much a product or competitor price typically moves.
  • Express ATR as a percentage of price to compare products fairly.
  • Use it to set review frequency and tolerance bands, not to forecast prices.

Frequently asked questions

Who created Average True Range?

J. Welles Wilder Jr. introduced ATR in 1978 as part of a set of technical trading indicators. It was designed for financial markets such as commodities and stocks, not for retail pricing.

What is the difference between range and true range?

Range is the high minus the low within one period. True range also considers the gap from the previous close, so a price jump between periods is counted even if the period itself was flat.

Can ATR predict competitor price changes?

No. ATR describes how much prices have moved, not which way they will move next. It is useful for comparing volatility and setting tolerance bands, not for forecasting.

What period should I use for product prices?

Choose a period that matches how often you observe prices, often a day or a week. Traders commonly use 14 periods for the average, but fewer periods react faster in retail markets with infrequent changes.

Related terms

In BlackAha, Price Tracking helps you monitor competitor prices and stock status automatically.

Back to the Pricing Dictionary