On EURUSD since 2020, a stop at 1.5 times daily ATR ranged from 64 to 234 pips. If you risk a fixed 1% a trade, your position size has to move across that same range.
- ATR(14) is the average true range of the last 14 bars: each day’s high to low, stretched to the previous close when the market gaps, smoothed with Wilder’s formula.
- The wider the multiple, the rarer the hit: since 2020 a EURUSD stop at 2 times the previous day’s average range was reached before the close in 0.7% of sessions, and at 3 times once in 1,767 sessions.
- On 8 October 2026 a stop at 1.5x the range was 94 pips, and 1% of a $10,000 account bought 0.10 lot, $94 at risk. A fixed 20-pip stop buys 0.50.
- An ATR trailing stop follows the best price since entry at a set multiple of the range, often 3, and moves only in the trade’s favour.
Since January 2020, a 20-pip stop under a EURUSD long opened at the day’s open was hit before the close on 64% of days; a stop at 1.5 times daily ATR, on 2.7% (UTC daily candles, spread not counted). An ATR stop loss sets the stop distance as a multiple of the Average True Range, usually the 14-day version, and the position size follows from that distance so the loss at the stop stays at your fixed 1%. You have $10,000, a EURUSD long in mind, and an empty stop-loss field in the order ticket.
A twenty-pip cushion feels safe: half a lot, $5 a pip. But the median EURUSD session since 2020 travelled 69.6 pips from high to low, so a 20-pip stop gets touched by noise, not by the idea being wrong.
One day in April 2025 the wider stop failed the other way: the price ran far past it. So what did the 1% rule cost that day?
How to use ATR for stop loss
Read ATR(14) on your timeframe, multiply it by 1.5 to 3, and put the stop that far from entry. Then size the position from that distance, not the other way round.
We tested each width on 1,767 daily EURUSD candles from Alpha Vantage, cut at UTC midnight, holiday bars included. Each day we placed a long at the open and checked whether the low reached the stop by the close, or within five sessions. Spread isn’t counted; it would add hits to tight stops first:
- Fixed 20 pips: hit the same day 63.6% of the time, within five sessions 84.1%.
- Fixed 60 pips: 19.8% the same day, 55.6% within five sessions.
- 1.5x the previous day’s ATR: 2.7% the same day, 27.6% within five sessions.
- 2x ATR: 0.7% the same day. 3x: once in 1,767 sessions.
Shorts look the same: a 20-pip distance above the open was reached on 64.2% of days.
Did the stop on your last trade come from the chart, or from the size you wanted to trade?
The tightest of these 1.5x stops in six years was 64.25 pips, set for 31 January 2020. Half were wider than 111.
ATR position sizing: 20-pip vs 60-pip stop on one 1%
Same formula as for any stop:
On EURUSD with a USD account one lot pays $10 a pip. With a volatility stop the distance changes every day, so the size changes with it:
| Stop on EURUSD | Stop, pips | $2,000 | $10,000 | $50,000 |
|---|---|---|---|---|
| Fixed at 20 | 20 | 0.10 | 0.50 | 2.50 |
| Fixed at 60 | 60 | 0.03 | 0.16 | 0.83 |
| 1.5x ATR at the 30 Jan 2020 close | 64.3 | 0.03 | 0.15 | 0.77 |
| 1.5x ATR at the 8 Oct 2026 close | 94.4 | 0.02 | 0.10 | 0.52 |
| 1.5x ATR at the 27 Mar 2020 close | 234.4 | below 0.01 | 0.04 | 0.21 |
Position sizes at 1% risk, rounded down to 0.01. Source: Alpha Vantage daily EURUSD candles; Wilder ATR(14); calculation by VipTools.
You probably assumed a wider stop means a bigger loss; at 0.10 lot, the 94-pip stop costs $94, against $100 for 0.50 on a 20-pip stop. What really changes is the number in the volume box, and how often you pay.
Now the day from April. On 2 April 2025 EURUSD’s ATR was 83.4 pips, so a 1.5x stop was 125 and 1% of $10,000 bought 0.07 lot. On 3 April the pair opened at 1.08262 and reached 1.11466, a 320-pip climb, 2.56 times the stop. A short filled at its stop would have lost $87.53, under 1% because of rounding. The price ran another 195 pips; the loss stayed where the 0.07 size had put it.
In late January 2020 the size for 1% was 0.15 lot; by 27 March the stop had widened to 234 pips and the position had fallen to 0.04. Counted in pip terms, USDJPY’s average range was wider than EURUSD’s 74.2% of the time; on 7 October 2026 its 1.5x stop was 171.5 pips, or 0.09 lot on $10,000.
Put the day’s stop into the lot size calculator before you place the order.
Where the ATR stop loss misleads
On real data the rule leaks in four places:
- It lags. The average looks back 14 sessions. On 3 April 2025 the day’s true range was 342 pips, 4.1 times the previous day’s reading, so the stop was sized for a market that no longer existed.
- Your feed changes the number. Our ATR(14) from Alpha Vantage and Twelve Data differed by 4.0 pips on an average day; on 3 April 2020, 146 against 168.
- A close-only series halves it. The ECB’s euro rate is one number a day, with no high and no low. Used as a stand-in, its average absolute daily change over 14 sessions came to 0.48 of the true figure on EURUSD since 2020; on 8 October 2026 it was 39.6 pips against 63.0.
- Small accounts can’t follow it down. On 27 March 2020 a $2,000 account needed 0.0085 lot; the minimum 0.01 put $23.44, or 1.17%, at risk.
“They are based on the daily concertation procedure between central banks across Europe, which normally takes place around 14:10 CET.”
European Central Bank, euro foreign exchange reference rates, checked October 2026
One moment a day can’t show how far the price swung around it.
Alpha Vantage’s FX_DAILY returns open, high, low and close per UTC day, and its free EURUSD history goes back to 9 August 2007. Source: Alpha Vantage API documentation and our download, 9 October 2026.
To be fair to fixed stops: a fixed 111-pip stop, the median width of our volatility stops, was hit the same day 3.6% of the time against 2.7%, and within five sessions on 27.1% against 27.6%. The average range’s gain is that it adapts the width, not that it beats a fixed stop of the same size. Our position: on a daily-chart EURUSD trade, let ATR set the width and let the size shrink to fit. If 1% at that stop gives less than your broker’s 0.01, skip the trade rather than tighten it.
How does an ATR trailing stop work?
It keeps the stop a set multiple of the range away from the best price since entry and moves only one way. Defaults differ: one platform we checked uses a 21-day ATR times 3, another leaves both to you.
Our rule: the stop sits 3 times ATR(14) below the highest high since entry, is reset at each close, and is hit when the next day’s low touches it (filled at the stop, or at the open on a gap). We ran it on a EURUSD long from the close on 28 February 2025, at 1.03747. The first stop sat 180.4 pips away, so 1% of $10,000 bought 0.05 lot. The trail rode the April rally and was hit on 8 May 2025: 876 pips, $438, 4.9 times the risk.
One good start proves little, so we started the same trail on every session since 2020, long and short: 3,491 trades closed, and 43 were still open on 8 October 2026, so we left them out. The average closed trade made minus 0.006 of its risk before spread, roughly zero, and the median lasted 16 days. The best made 11.9 times its risk, but 62.4% closed at a loss.
What we couldn’t measure is the fill: daily candles show where a stop was touched, not the price a broker gave on days like 3 April 2025.
What is ATR in trading
ATR, the Average True Range, is J. Welles Wilder’s measure of how far a market moves in a bar, described in his 1978 book New Concepts in Technical Trading Systems. True range is the largest of three gaps: today’s high to low, today’s high to yesterday’s close, today’s low to yesterday’s close. Wilder’s 14-bar version is (yesterday’s value × 13 + today’s true range) ÷ 14, and says nothing about direction.
On EURUSD a reading of 0.0063 is 63 pips. On 9 September 2026 EURUSD’s daily ATR was 45.5 pips; by 8 October it was 63.0, and the same trade idea needed a wider stop and a smaller position. On 8 October that meant 94 in the stop-loss field and 0.10 in the volume box, for $94 at risk, where a 20-pip distance and 0.50 lot would have put $100.
Size your next trade
Account, risk and stop in, position size out. Free, in your browser, no sign-up.