Why this indicator is underestimated
ATR — Average True Range — is one of those indicators you'll find in every charting tool but almost nobody actually uses. Wrong reflex. ATR doesn't tell you where the market is going. It tells you how far it can go. And that's the information that should drive your stop-loss distance, your position size, and your strategy selection — not your gut.
Invented in 1978 by J. Welles Wilder, in the same book that introduced RSI. But while RSI became a pop star, ATR stayed in the shadows. Unfairly. We've spent the last weeks integrating it deeply into Backtesting Arena — and in our own tests, it improves drawdowns across nearly every trend-following strategy by 20 to 40 %. That's the kind of single change that's rare in trading.
This post explains what ATR is, why it's especially indispensable for crypto, and what four ATR tools you can use in the Arena starting today.
What ATR measures
ATR measures one thing: average movement range per period. No trend, no direction, no momentum.
Concretely: ATR(14) on a daily chart tells you how far the price moved on average over the last 14 days per day — including overnight gaps. That's the key difference from a simple High-Low range. When the market opens with a gap, the "True Range" captures that gap.
An example makes it concrete. Daily ATR(14) on different assets:
| Asset | Typical value |
|---|---|
| S&P 500 ETF (SPY) | 0.8 – 1.5 % |
| Gold (GLD) | 0.7 – 1.2 % |
| Bitcoin | 3 – 7 % |
| Ethereum | 4 – 8 % |
| Altcoins | 5 – 15 % |
Bitcoin moves daily about as far as a stock ETF moves in an entire week. Anyone trading crypto with stock-sized stops gets stopped out mercilessly. That's not a question of mental toughness — it's a question of scale. ATR delivers the right scale.
The formula — short and painless
Step one: True Range per candle.
TR = max(
High − Low,
|High − Previous Close|,
|Low − Previous Close|
)
Step two: smooth over 14 periods. In the Arena we default to Wilder smoothing — that's the original method and gives the most stable values. Alternatively EMA or SMA, depending on whether you want faster reaction or less noise.
You don't need more theory than that. Things get interesting when you actually use ATR.
Four tools — all available in the Arena
1. ATR Trailing Stop (the biggest leverage)
Classic stop-loss = fixed distance below entry. Problem: in volatile phases you get stopped out, in calm phases you risk too much. Solution: a stop that follows the price and adapts to volatility.
Trailing Stop = Highest High Since Entry − (ATR × Multiplier)
Standard multiplier: 2×. Aggressive: 1.5× (tighter stops, more whipsaws). Conservative: 3× (wider stops, longer trends).
On BTC with ATR = 4 % and multiplier 2×, your stop sits 8 % below the current high. Sounds wide. It's normal for BTC. Anyone trading BTC with 2 % stops gets thrown out by noise, not by reversals.
In the Arena, the Trailing Stop is now available as a global modifier on every strategy. Tick the checkbox, choose the multiplier, done. Pro feature. In our backtests, it cuts max drawdown on trend-following strategies by 20 to 40 % at nearly identical CAGR — the most powerful single intervention we've ever added to a strategy.
2. ATR Volatility Filter (the entry gatekeeper)
Some strategies only work in specific volatility regimes. RSI mean-reversion on BTC works in calm phases — and breaks down in panic markets. Logic: in chaos phases the price jumps, "mean reversion" doesn't apply.
With the ATR Volatility Filter you set conditions for entry: only trade when ATR is below X %. Or only when ATR is significantly above its 20-day average (volatility expansion → trend beginning). Three modes:
- Low Vol: entry only when ATR is below threshold. Avoids chaos markets.
- High Vol: entry only when ATR is above threshold. Targets breakouts.
- Expansion: entry when ATR ≥ 50 % above 20-day average. Filters for trend onsets.
Analogous to the 200-WMA filter and Altcoin Season filter. Stackable with other filters.
3. Keltner Channel Breakout (standalone strategy)
An ATR-based strategy that catches trend breakouts. Keltner Channels = EMA ± ATR × Multiplier. Entry when price breaks out of the channel — exit when it falls back to the middle.
Middle line = EMA(Close, 20)
Upper line = Middle + ATR(10) × 2
Lower line = Middle − ATR(10) × 2
Long entry when close crosses above the upper line. Long exit when it drops below the middle. Clean trend-following logic with built-in volatility adaptation. Now available alongside EMA Trend Bias and WMA Trend in the trend-follower category.
4. Volatility Regime Insights (the analysis tool)
We now measure the average ATR during every backtest run and correlate it with performance. You'll find the result in Strategy Insights. Sample output:
"RSI/SMA on BTC 1D performs best with ATR between 1.5 % and 3 % (median CAGR +42 %), poorly with ATR above 5 % (median CAGR −18 %)."
This way you don't just know whether your strategy worked historically — you know in which volatility regime it worked. That's a layer we haven't seen in any other platform.
What ATR can't do
Clear line, no sales marketing: ATR has limits.
It lags. When volatility jumps suddenly (black swan, news event), it takes ATR a few periods to reflect that. It has no fixed signal levels like RSI's 30/70 — you have to develop relative measures yourself. It can be distorted for weeks by a single outlier day. And it tells you nothing about direction.
ATR isn't a signal. ATR is a tool.
How to start now
Fastest way to get a feel for ATR:
- Pick one of your saved strategies. Any of them.
- Activate the ATR Trailing Stop with multiplier 2.
- Re-run the backtest and compare drawdown and CAGR with the original run.
On most trend-following strategies you'll see the effect immediately. On mean-reversion strategies the effect is smaller, sometimes even negative — because tight exits are part of the logic there. That insight itself is valuable: ATR Trailing Stop isn't right for every strategy, but for the right strategies it's a leverage almost nothing else matches.
Anyone who has never saved a strategy in the Arena can start with our default strategies — RSI/SMA Cross on BTC 1W is a good entry point because the trailing stop effect shows up especially cleanly there.
Takeaways
- ATR measures volatility, nothing else — no trend, no direction.
- Crypto needs different multipliers than stocks. 2× ATR on BTC means 6–14 % stop distance. That's normal.
- The ATR Trailing Stop is the most powerful single intervention you can add to a strategy. 20–40 % less drawdown at nearly identical CAGR.
- ATR Volatility Filter, Keltner Breakout, and Volatility Regime Insights are now live in Backtesting Arena. Pro features.
- The best way to understand ATR isn't to keep reading — it's to re-run one of your strategies with the trailing stop now and compare the result.
[CTA Button:] Test a strategy with ATR Trailing Stop → tradingstrategies.work
Disclaimer: Backtest results don't guarantee future performance. Crypto trading carries significant risk of loss. This post is educational content, not investment advice.