forex-scalping.com

ATR, The Tool That Fixes Your Stop Loss Problem

Many traders choose stop distances arbitrarily, such as "I always use 20 pips." But markets change. A 20 pip stop might be fine in quiet conditions and far too tight during volatile ones. The Average True Range (ATR) measures recent volatility, showing the average range of price movement over a number of periods, commonly 14.

How to use ATR for stops:

1. Add the ATR indicator to your chart and read its value for your entry timeframe.
2. Decide a multiple, such as 1.5 times ATR, as the minimum distance for noise.
3. Combine this with structure. Place the stop beyond the logical invalidation point, and check whether that distance is at least reasonable relative to the ATR.
4. If the structural stop is much smaller than ATR, price noise may hit it. If it's much larger, adjust your position size.
5. Calculate lot size based on the final stop distance, risking 1% or less.

How to use ATR for targets:

- Compare your target distance with ATR to see whether it's realistic
- Consider scaling targets according to current volatility

Other benefits:

- Helps adjust risk in volatile periods
- Provides context for whether a move is large or small

Remember: ATR is information, not a command. Your stop should still make sense on the chart.

Let volatility inform your decisions, and avoid one-size-fits-all numbers.

A trap with ATR on scalping timeframes: at the London open, ATR(14) on M5 is built from the last seventy minutes of Asian bars. At 07:55 it might read 3 pips on EURUSD while the first London candles run 8 to 10 pips each. A 1.5x ATR stop taken from that reading is about 4.5 pips and gets eaten by ordinary opening noise. I compare against the ATR from the same time yesterday or use a longer lookback during session transitions.

For point 5, the sizing step is simple enough to do on paper. With a $10,000 account and 1% risk you have $100 to lose. A 6-pip stop on EURUSD at $10 per pip per standard lot gives 100 / (6 x 10) = 1.67 lots. Widen the stop to 9 pips on a volatile morning and it becomes 1.11 lots. Same dollar risk, which is the whole point of letting ATR move the stop and the lot size move with it.

ATR combined with structure is a sound method. One detail to state explicitly: which timeframe's ATR. A 1.5x ATR stop on M15 is much tighter than 1.5x ATR on the daily chart, and mixing the two is a common source of stops that are either too tight or far too wide.

Writing in the plan "1.5x ATR(14) on the entry timeframe, beyond the structure point" leaves no room for interpretation. It also makes the backtest repeatable.

The 07:55 trap PropScalpDesk describes gets worse once you look at how slowly ATR(14) catches up. On M5 it needs 14 bars, so it only fully reflects London at about 09:10, long after the opening move has been traded.
The chart is a typical EURUSD morning in round numbers. ATR read 3.0 pips at 07:55 and still only 5.4 after the 08:25 bar, while the six bars from 08:00 averaged 8.5 pips each.

My fix is close to PropScalpDesk's same time yesterday, only averaged. I take the mean range of the 08:00 to 08:25 bars over the last ten sessions, which came to 7.8 pips that week, and use it as the noise floor for the open.

On FXS Prop Desk's wording, I would add which number wins when structure and ATR disagree. For me structure decides where the stop goes and ATR only sets a minimum, so if the structural stop is under one times the floor, I skip the trade rather than push the stop out to a level that means nothing.

That skip rule removed most of my 4 and 5 pip stops at the open. The sizing step from point 5 then handles the rest.