Normalized ATR (NATR)

TL;DR. The Average True Range divided by the close, times 100. A reading of 1.5 means the average bar has recently spanned about 1.5% of price, whether the market trades at 20 or at 5,000, so one threshold works across instruments and across years.

#What it measures

• Volatility in percent, not points. The ATR says how many points a bar usually travels. NATR says how big that is next to price, which is the number that compares across markets.
• The same smoothing as ATR. Each bar's true range (the largest of high minus low and the gaps from the previous close) is averaged with Wilder's smoothing, then divided by the close.
• A regime, not a direction. NATR rises in a fast market in either direction and falls in a quiet one. It never says which way price will go.

#Formula

#Inputs in AlgoLift

#Recommended periods

• 5 to 10 (fast): Reacts within a session to a burst of volatility.
• 14 (default): The standard reading, comparable to most published ATR work.
• 50 to 100 (slow): A background regime filter that moves over weeks, not hours.

#Outputs in AlgoLift

#How to read it

• High against its own history: bars are travelling further than usual next to price. Breakouts follow through more often, and fixed-point stops are hit more often.
• Low against its own history: a quiet, compressed market. Mean-reversion rules usually have the easier time here.
• Across instruments: a NATR of 0.8 on one future and 0.8 on another describe the same relative movement, which a point-based ATR cannot tell you.
• Rising from a low base: volatility is expanding, often the first sign that a range is ending.

#Best in, worst in

Best in filters and comparisons: turning a strategy on only when the market moves enough, or comparing volatility across instruments and across years of one instrument.

Worst in anything that needs a price distance. NATR has no unit of price, so it cannot place a level by itself.

#Three setups

#1. A volatility floor

Compare NATR(14) with a constant in a Comparison node and AND it into your entry: trade only when NATR is above, say, 0.5. A breakout rule then sits out the dead sessions where a break rarely runs.

#2. A volatility ceiling for mean reversion

Mirror the first setup: a fade rule that only trades while NATR is below its 100-bar average (a Moving Average Modifier on the NATR line) avoids fading into a market that is breaking out.

#3. Expansion from a squeeze

Enter on a Crossover of NATR above its own 20-bar average while price breaks a range, and size the stop from the ATR through a Bracket Order. NATR times the turn, ATR measures the distance.

#Common mistakes

• Wiring NATR into a stop distance. It is a percent; the stop and bracket paths read the ATR in points. Keep the two nodes apart.
• Fixing one threshold for every timeframe. A 1-minute bar spans far less of price than a daily bar. Read NATR against its own history on the timeframe you trade.
• Tuning the threshold to the decimal. A filter that works at 0.62 and fails at 0.60 has found noise (overfitting in trading).