ATR stop loss
Bitcoin and a small altcoin have different "noise": a 2% stop is normal for one and gets hit by a random candle on the other. ATR measures that noise, and an ATR stop adapts to each coin automatically. Here's the formula, the multipliers and what the backtest showed.
What ATR is
ATR (Average True Range) is the average true range of a candle. The indicator was introduced by J. Welles Wilder: for each candle the "true range" is the largest of three values:
Put simply, ATR shows how many dollars a coin usually moves in one candle, including gaps between candles. For bitcoin on the 4-hour chart that can be hundreds of dollars; for a $0.50 altcoin — a couple of cents. As a percentage of price, altcoins almost always have a bigger ATR: they're "noisier".
ATR stop formula
Short: stop = Entry + k × ATRk is the multiplier: usually from 1 to 3
Example: long BTC at $75,000, ATR(14) on 4H = $900, multiplier 1.5.
Stop = 75,000 − 1.5 × 900 = $73,650 — 1.8% from entry.
Take 1R = 75,000 + 1,350 = $76,350, take 2R = $77,700.
Then the position size is derived from this stop so the loss at the stop equals exactly the chosen share of the deposit — that's what the position size calculator does. A wide stop doesn't mean more risk: the position will simply be smaller. How to pick the percentage itself is covered in «How much to risk per trade».
Which multiplier to choose
- 1 ATR — a tight stop. Bigger position, but ordinary price swings often knock it out.
- 1.5 ATR — a compromise: beyond a candle's usual noise but not too far. AI Trading Room uses it in coin breakdowns.
- 2–3 ATR — a wide stop for trades that need time and for strategies with partial profit-taking and a distant target.
The key point: the stop multiplier is chosen together with the take profit. On its own it says nothing about profitability.
What the backtest showed
We tested ATR-based stops and takes on two years of Bybit coin history, fees included. A few honest takeaways:
- A high win rate guarantees nothing. A 0.5 ATR take with a 3 ATR stop gives about 88% winning trades — and almost zero in total: rare stops eat all the small wins.
- Working combinations are more modest. "Bounce in a downtrend" with a 1.5 ATR stop and 1 ATR take: 612 trades, 73% winners, +0.18R per trade on average after fees. "Inside-day breakout" with a 1.5 ATR stop and 0.75 ATR take: 633 trades, 74%, +0.07R.
- A wide stop works together with partial profit-taking. A "Bounce" variant with a 3 ATR stop, closing half at 1R and moving the stop to break-even, and running the rest to 5R showed 63–69% winners and +0.29R to +0.51R per trade on three different periods of history. It's now being tested live.
The bottom line: an ATR stop isn't a magic level, it's a way to make risk comparable across coins. Profit comes from the entry + stop + take combination, and that has to be tested on history.
Common mistakes
- ATR from the wrong timeframe. You enter on the hourly chart but compute the stop from the daily ATR — the stop ends up many times wider than intended.
- Same position size with different stops. A 1.5 ATR stop can be 6% on an altcoin and 1.8% on bitcoin. Recalculate the size, or your dollar risk jumps around.
- Stop beyond liquidation. Wide stop + high leverage = liquidation before the stop. The position size calculator shows the safe leverage.
- Moving the stop further as price approaches it. A volatility stop already accounts for noise — beyond it is just hope.
FAQ
How do I calculate an ATR stop loss?
For a long, stop = entry − multiplier × ATR; for a short, stop = entry + multiplier × ATR. Usually ATR with a period of 14 on the timeframe you trade, and a multiplier from 1 to 3.
What ATR multiplier is best for a stop?
There's no universal one: the multiplier is chosen together with the take profit and tested on history. 1.5 ATR is a common compromise between false stop-outs and position size; 2–3 ATR suits strategies with partial profit-taking and a distant target.
Which timeframe should I use for ATR?
The one you make entry decisions on. In AI Trading Room breakdowns the stop is based on the 4-hour ATR.
Is a wide stop more risk?
No, if the position size is derived from the stop. With a stop twice as wide the position is half as big, and the loss at the stop stays the same share of the deposit.
ATR stop — in every breakdown
Send a ticker to AI Trading Room and get an entry, a volatility stop, two take profits and size for your deposit.