Choppiness Index

A 0-100 gauge of whether a market has been trending or chopping sideways — it describes the character of price movement, not its direction.

What it is

The Choppiness Index (CHOP) is an indicator developed in the early 1990s by Australian commodities trader E.W. Dreiss (often referred to as Bill Dreiss). It is bounded between 0 and 100 and is designed to describe how a market has been moving rather than where it is heading. Higher readings are typically read as a sideways, "choppy" or consolidating market, while lower readings are read as a more directional, trending market. Importantly, the Choppiness Index is non-directional: it says nothing about whether price is going up or down, only about the character of the movement. It is a descriptive, lagging measure based on past prices and is not predictive. Nothing it shows is a signal to act on, and trading always carries a risk of loss.

How it works

CHOP compares how much ground price has actually covered over a lookback window against the net range of that window. The formula is: 100 × log10( Sum of one-bar true ranges over n bars / (Highest High over n − Lowest Low over n) ) / log10(n). The numerator inside the log — the running sum of one-bar true ranges — captures the total path length the price travelled bar by bar. The denominator — the highest high minus the lowest low — captures the net range of the window. When price travels a long, winding path but ends up inside a narrow range, the ratio is large and CHOP reads high (choppy). When price travels in a straighter, more efficient line, the path length is closer to the net range, the ratio is smaller and CHOP reads low (trending). The log10(n) term normalises the result so it stays on a roughly 0-100 scale regardless of the chosen period. The default period is 14.

How traders read it

• Traders often read high values (commonly above 61.8) as describing a market that has been moving sideways or consolidating, with little clear directional follow-through.
• Traders often read low values (commonly below 38.2) as describing a market that has been trending more directionally — though the index does not say whether that move was up or down.
• The 38.2 and 61.8 levels are Fibonacci-derived reference lines; values in between (roughly 38.2-61.8) are commonly treated as a transition or undecided zone rather than a clear state.
• Because CHOP is non-directional, traders usually combine it with a separate directional tool (such as a moving average or trend indicator) to gauge which way price has moved, while CHOP only describes how cleanly it moved.
• Some traders watch for the index falling from a high reading toward a low one as a description of a market that has been shifting from range-bound toward more directional behaviour — but this describes past bars and is not a forecast.
• All readings reflect completed price history; the index lags and can change as new bars print.

Common settings

Default period is 14. Some traders use shorter periods (e.g. 8-10) for faster, noisier readings or longer periods (e.g. 20-28) for smoother, slower ones. The 38.2 and 61.8 reference lines are the standard Fibonacci-based levels; a few traders simplify to 40 and 60. These are conventions, not settings that make the tool more "correct".

Strengths

• Puts a clear 0-100 number on something traders otherwise judge intuitively — whether a market has been trending or chopping.
• Bounded and normalised, so readings are broadly comparable across instruments and timeframes.
• Often used as a context filter: many traders use it to describe market character before applying separate, direction-specific tools.
• Built from familiar components (true range, highest high, lowest low), making its logic transparent and easy to audit.

Pitfalls to watch

• It is purely descriptive and lagging — it summarises what price has already done and offers no prediction about future direction or whether a condition will persist.
• It says nothing about direction. A low reading can accompany either an up-trend or a down-trend, so it cannot be used alone to decide anything.
• The 38.2 / 61.8 thresholds are conventions, not magic numbers; readings can hover near a threshold or whipsaw across it, especially on shorter timeframes.
• Like most range-based measures, results shift with the lookback period and the timeframe — a market can look choppy on one timeframe and trending on another.
• If the window's high and low are equal (perfectly flat bars), the denominator is zero and the calculation is undefined, so very low-activity periods can produce gaps or erratic values.
• Treating any reading as a buy or sell trigger misuses the tool. At most it is a context filter, and trading on it — or on anything — carries a real risk of loss.

Pine v6 example

//@version=6
indicator("Choppiness Index", overlay = false)

length = input.int(14, "Length", minval = 2)

// Sum of one-bar true ranges over the window (the path length)
atrSum = math.sum(ta.tr(true), length)

// Net range of the window
rangeHL = ta.highest(high, length) - ta.lowest(low, length)

// Guard against a zero range (perfectly flat window)
chop = rangeHL > 0 ? 100 * math.log(atrSum / rangeHL) / math.log(length) : na

plot(chop, "CHOP", color = color.new(color.teal, 0))
hline(61.8, "Choppy (>61.8)", color = color.gray, linestyle = hline.style_dashed)
hline(38.2, "Trending (<38.2)", color = color.gray, linestyle = hline.style_dashed)

Pro tip: Because the Choppiness Index is deliberately non-directional, traders tend to treat it as a context filter rather than a standalone tool: it is often paired with a separate directional indicator so that one component describes how cleanly price has been moving and the other describes which way. Remember the index only summarises past bars — it does not forecast, and nothing here is advice to buy or sell.

Educational only — not financial advice, not a recommendation to trade. No indicator is predictive; trading involves substantial risk of loss.