Most new traders set stop losses the same way: pick a round number of points, ticks, or pips that "feels" reasonable, and use it on every trade regardless of what the market is doing. The problem is that a 10-point stop on a quiet, low-volatility day and a 10-point stop during a volatile news-driven session are not the same risk at all — one gets hit by ordinary noise, the other leaves room for a real move against you. Average True Range (ATR) solves this by measuring volatility directly, so stops and position sizes can adapt to actual market conditions instead of a fixed number picked once and never revisited.
What Is Average True Range?
ATR is a volatility indicator developed by J. Welles Wilder that measures the average range a market has moved over a given period, typically 14 days or bars. Unlike a simple high-minus-low range, ATR accounts for gaps between sessions by using "true range" — the largest of three values.
True Range = Max[(High − Low), |High − Previous Close|, |Low − Previous Close|]
ATR is then a moving average of true range over the chosen period — most commonly a 14-period Wilder's smoothing, though simple or exponential moving averages work too. The result is a single number, in the instrument's native units (points, dollars, pips), representing how far that market typically moves per bar.
Why ATR Beats a Fixed-Point Stop
- A fixed stop distance doesn't know the difference between a calm range-bound market and a breakout day — ATR does, because it's recalculated from recent price action every bar
- Stops sized in ATR multiples widen automatically when volatility expands and tighten when it contracts, without the trader having to manually adjust
- It works across instruments with wildly different price scales — a 2×ATR stop makes the same relative sense on a $5 stock and a $500 stock, where a flat "$1 stop" would not
- It removes a common source of emotional stop placement: picking a stop distance based on what "feels safe" rather than what the market is actually doing
Setting an ATR-Based Stop Loss
The standard approach is to place the stop a multiple of ATR away from entry — commonly 1.5× to 3× ATR, depending on strategy and timeframe. A tighter multiple suits scalping and mean-reversion setups where a quick invalidation matters; a wider multiple suits swing and trend-following setups that need room to breathe through normal pullbacks.
| Instrument | 14-Period ATR | Stop Multiple | Stop Distance |
|---|---|---|---|
| Stock (e.g. $85 name) | $1.20 | 2× | $2.40 |
| E-mini futures (ES) | 18 points | 1.5× | 27 points |
| EUR/USD | 55 pips | 2× | 110 pips |
As the instrument's volatility changes — a stock starts trending harder, a currency pair goes quiet — the ATR value updates and the next trade's stop distance updates with it, without any manual recalibration.
Using ATR for Position Sizing
ATR plugs directly into the same position sizing formula used for any fixed stop — the only change is that the stop distance itself comes from a volatility measurement instead of a static number.
- 1Calculate the ATR-based stop distance (e.g., 2× the 14-period ATR)
- 2Decide the dollar risk for the trade (account balance × risk %)
- 3Divide dollar risk by the ATR stop distance, converted to dollars per unit, to get position size
- 4Size scales down automatically on high-ATR (volatile) instruments and up on low-ATR (calm) ones, keeping dollar risk consistent trade to trade
This is the real payoff of ATR-based sizing: two setups with identical dollar risk can end up with very different share or contract counts, purely because one instrument is currently twice as volatile as the other. Without ATR, a trader sizing by feel tends to under-risk calm markets and over-risk volatile ones — exactly backwards from what a consistent risk plan requires.
TradeEdge lets you log the ATR, stop distance, and position size for every trade alongside your entry and exit — so you can see in your own performance analytics whether ATR-based stops actually improved your results, instead of guessing.
Common ATR Mistakes
- Using a single ATR multiple across every instrument and timeframe without back-testing whether it fits the strategy
- Treating ATR as a take-profit target rather than a stop-distance and sizing input — it measures typical range, not directional bias
- Ignoring that ATR lags — it's built from past bars, so it reacts to a volatility spike or collapse a few bars after it starts, not the instant it begins
- Recalculating position size once at the start of a trading session instead of per trade, letting stale ATR values drive sizing on a symbol whose volatility has since shifted
- Confusing a wider ATR-based stop with "more risk" — the position size should shrink to compensate, keeping dollar risk constant regardless of the stop distance
ATR Stops Don't Replace a Risk Plan — They Refine It
ATR answers "how far should my stop be, given current volatility?" — it doesn't answer "how much of my account should I risk?" or "is this setup worth taking at all?" Those are still separate decisions. A trader who plugs a well-reasoned ATR stop into an otherwise undisciplined process — no max drawdown limits, no consistent position sizing, no review of what's actually working — just gets a more precisely calculated version of the same undisciplined outcome.
The value shows up when ATR-based risk becomes part of a logged, reviewable process: every trade's ATR, stop multiple, and resulting size recorded next to its outcome, so a trader can eventually answer a concrete question — does a 1.5× or 2.5× ATR stop actually produce a better expectancy on this setup — with real data instead of a guess.