Average True Range (ATR) — basics, formula and the 14-period default

Marcin traded GBP/JPY with a thirty-pip stop loss — the very one that had earned a steady profit on EUR/USD for two years. In his first three weeks on the cross he was knocked out of fifteen trades in a row, every one in the red, even though the direction proved correct several times afterwards. The fault was not in his analysis but in a number he had never checked: the pound against the yen moves more than twice as far in a day as the euro against the dollar. That typical width of movement is exactly what the Average True Range measures — and it is where sensible stop-loss and position sizing begins.

What the Average True Range really is

The Average True Range (ATR) measures the average amplitude of price movement across a chosen number of recent sessions. J. Welles Wilder introduced it in 1978 in New Concepts in Technical Trading Systems, the same book that debuted the RSI, the Parabolic SAR and the ADX. He designed it for commodity markets, where gaps between sessions made a plain high-minus-low range meaningless, but the solution proved universal enough that it is now the default volatility tool across every asset class, from equities to currency pairs.

What matters most is what ATR does not do: it does not show direction, only amplitude. A rising reading tells you the market has started to move more energetically, and nothing more; whether that energy is up or down, the indicator does not know. That is why ATR is a risk-management tool rather than a signal generator: the trader decides on direction from price analysis first, and only then uses ATR to size the stop loss and the position. The same line separates it from the ADX, which gauges trend strength — one says how far, the other how decisively, neither which way. The default period is 14 sessions, set that way in MetaTrader 4 and 5, in TradingView and in cTrader, and the reading is in the units of the instrument itself: pips on currencies, points on equities and commodities.

How true range and the indicator are calculated

To understand ATR you first need to understand true range, which Wilder defined as the greatest of three values measured for each session. The first is the plain session range, the difference between the highest and the lowest price of the candle. The second is the distance between the previous close and the current high, the third between the previous close and the current low — both taken as a positive number. The indicator picks the largest of the three, and that figure is the true range of the session.

Why bother with the complication? Because a plain range cannot see gaps. A purely illustrative example on EUR/USD: on Monday the market travelled 70 pips with no gap, so the true range is 70 too. On Tuesday the session covered only 65 pips but opened 10 pips higher after a weekend gap — and it is that gap that lifts the true range to 75, even though the day's own range misses it. Average a few such consecutive readings and you get a simplified ATR; the real indicator computes it over 14 sessions with Wilder's smoothing method, which every platform has built in, but the logic is identical.

A realistic ATR(14) reading on EUR/USD sits on the daily timeframe usually between 60 and 90 pips, and on GBP/JPY between 150 and 250 — and that difference explains why a stop loss carried wholesale from one pair to the other is mathematically destined to fail.

Why the 14-period setting became the standard

The number 14 did not come from nowhere. Wilder tested lengths on commodities from 5 to 30 sessions, and 14 — roughly three trading weeks — produced the best compromise between responsiveness and stability: long enough that a single extreme candle does not dominate the reading, short enough that a regime change reaches the indicator within days. Shorter settings suit scalpers and longer ones position traders, but any change should follow a backtest, never an intuition.

ATR as a volatility measure — and its limits

ATR reads in the units of the instrument on which it is applied, which is precisely why two ATR readings from two different markets are not directly comparable. Sixty pips on EUR/USD is an average daily range, whereas 60 points on the S&P 500 signals a crisis session. The lesson: always compare an instrument's ATR with its own history, never with a reading from another market — a separate piece on ATR as a volatility measure develops that comparison in full. For the wider picture, ForexMechanics keeps a concise reference on market volatility.

The second limit is behavioural. ATR rises quickly during panic and falls slowly. If the pre-session reading is twice the average of the last fifty sessions, the market has entered an elevated-risk phase — spreads widen, slippage becomes routine, and stop losses based on a regular multiple land in the wrong places. The trader should then halve the position or refrain from trading until the reading settles. Conversely, a very low ATR describes a market sluggish enough that most momentum strategies stop working.

Stop loss based on ATR — from 1.5x to 2x

The most common and most valuable use of ATR is stop-loss placement. The rule for a long position: place the stop below the entry price by an amount equal to ATR(14) times the chosen multiplier; for a short, the direction flips. Standard multipliers are 1.5x for day trading and 2x for swing trading, and specific pairs show the value of that rule. When ATR(14) on EUR/USD reads 80 pips, the 1.5x multiplier gives a 120-pip stop and the 2x multiplier a 160-pip one; on GBP/JPY, with an ATR around 200 pips, the same multipliers give 300 and 400 pips. Each level looks different in absolute terms, but each represents the same percentage of the daily range — that is the entire idea of an ATR-based stop. The same pip-distance on every pair, the classical beginner error, instead guarantees that volatile markets are stopped out by ordinary noise. It is exactly what undid the Marcin from the opening.

Position sizing computed from ATR

The second layer of professional ATR use is position sizing. Take the amount you are willing to lose on a trade and divide it by the risk of one lot, which follows from the stop-loss width times the pip value — and since the stop width comes from ATR, the indicator feeds straight into that calculation. Purely as an illustration: a trader with a €10,000 account and the one-percent rule risks €100 per trade. On EUR/USD the stop is 120 pips (1.5x an ATR of 80) and the pip value €10 per standard lot, so €100 of risk divided by those figures gives a position of about 0.08 lots. On GBP/JPY, where the stop is 300 pips, the same calculation gives roughly 0.05 lots — a smaller lot, but identical exposure, because the maximum loss stays the agreed one percent of capital. This is what discipline around the one-percent rule requires — not the same number of lots, but the same monetary exposure across pairs.

Trailing stops based on ATR

The third critical use is the trailing stop — a stop that follows price as the trade goes the trader's way. The classical implementation tightens it by a multiple of ATR behind current price: two times for day trading, three times for swing trading. The most famous version is Chuck LeBeau's Chandelier Exit: three times ATR(22) subtracted from the highest high of the last twenty-two sessions.

The whole point of such a trail is to separate noise from a genuine retracement. A fixed pip-distance trail either gets knocked out on every market sneeze when volatility expands, or leaves the position unprotected when volatility contracts. An ATR-based trail adjusts itself: when the market accelerates it grants a larger cushion, and when it slows it tightens and locks in more of the gain. A fuller picture, with variants and pitfalls, is the subject of a separate article on advanced ATR trailing-stop techniques.

"A pip-only stop loss makes sense in exactly one case: when the trader trades one instrument, in one volatility regime, for an entire career. In any other version they need an indicator that adapts to what the market is actually doing this week." — J. Welles Wilder, New Concepts in Technical Trading Systems, Trend Research, 1978.

What to do tomorrow

The theory of ATR is simple, but it only sinks in once you see it work on your own chart and your own account. Here are four steps that close the lesson in practice.

1. Open the daily chart of the three pairs you actually trade, apply the ATR indicator with its default 14-period setting, and record the readings — you will see for yourself how widely typical volatility differs between the euro-dollar and the pound-yen.
2. Convert your existing stop loss on each of those pairs into a multiple of ATR, and check whether on the more volatile pair you are running too tight a stop that ordinary noise knocks out rather than a genuine change of direction.
3. Size every future position from the risk amount and the stop-loss width taken from ATR, so that the maximum loss on each trade equals the same one percent of capital whichever pair you are trading.
4. Once a week, compare each pair's current ATR with the average of its last fifty sessions and halve your position wherever the reading has jumped by more than half — your early signal that the market has entered an elevated-risk phase.