Chaikin Volatility indicator: measure range expansion

Calculate the range EMA, then its rate of change

1. For each bar, compute R = high - low. Do not substitute close-to-close returns or true range.
2. Smooth R with an EMA of length N: E[t] = α × R[t] + (1 - α) × E[t-1], where α = 2 / (N + 1).
3. Compare E[t] with E[t-M], the EMA from M bars earlier: CV[t] = 100 × (E[t] - E[t-M]) / E[t-M]. The baseline must be positive and available.

The EMA period N and ROC lookback M are separate inputs. Fidelity's formula describes 10 for both. The code below uses 10 and 12, matching Pineify's current indicator configuration. These are starting values, not evidence of an optimal timeframe or trading edge.

A five-bar calculation you can check

Use N = 3, so α = 0.5, and M = 2. Assume the EMA at bar 0 is 2.00; this is a stated starting value, not a claim about every platform's initialization. All prices are synthetic.

At bar 2: 100 × (4.50 - 2.00) / 2.00 = +125%. At bar 3 the EMA falls from 4.50 to 4.25, yet CV remains positive because 4.25 still exceeds its two-bar baseline of 3.00. A positive CV therefore does not mean the range increased on the latest bar.

At bar 4: 100 × (3.125 - 4.50) / 4.50 = -30.56%, rounded. Keep full precision during the calculation and round only the displayed result.

Pine Script v6 with a denominator check

Paste this indicator into TradingView's Pine Editor, save it and add it to the chart. It plots CV in a separate pane and exposes the current and baseline range EMAs in the Data Window. The first M bars have no comparison value. Allow additional history for EMA initialization when comparing charts.

//@version=6
indicator("Chaikin Volatility: range change", shorttitle="CV", overlay=false, precision=2)

emaLength = input.int(10, "Range EMA length", minval=1)
rocLength = input.int(12, "ROC lookback bars", minval=1)
minimumBase = input.float(0.0, "Minimum baseline range (price units)", minval=0.0)

rangeEma = ta.ema(high - low, emaLength)
baseline = rangeEma[rocLength]
validBase = not na(baseline) and baseline > minimumBase
cv = validBase ? 100.0 * (rangeEma - baseline) / baseline : na

plot(cv, "Chaikin Volatility (%)", color=color.blue, linewidth=2, style=plot.style_linebr)
hline(0, "Unchanged range", color=color.gray)
plot(rangeEma, "Range EMA (price units)", display=display.data_window)
plot(baseline, "Baseline EMA (price units)", display=display.data_window)

crossedAboveZero = ta.crossover(cv, 0)
confirmedExpansion = barstate.isconfirmed and crossedAboveZero
alertcondition(confirmedExpansion, "Confirmed CV cross above zero", "Smoothed range crossed above its lookback baseline. Direction requires a separate rule.")

Set EMA length to 3 and ROC lookback to 2 to match the worked example's parameters; your chart prices will differ from the synthetic table. With the minimum baseline at 0, the script excludes zero but still displays small positive baselines. Blank sections indicate unavailable or excluded data.

The plot can move during the open bar. The alert condition requires a confirmed close and a crossover from at or below zero to above zero. Create an alert using that condition and choose Once Per Bar Close. A confirmed expansion alert reports a range comparison; it does not submit orders or identify a trade direction.

Use CV as a filter with an independent entry rule

1. Define a volatility state, such as a valid CV above zero. This state can persist for multiple bars; a zero-line crossover is a one-bar event.
2. Define direction separately, for example a close above the highest high of the prior 20 completed bars for a long breakout. Exclude the current bar from that reference.
3. Specify the exit, position size, fees and slippage before backtesting. Compare the same strategy with and without the filter on data outside the parameter-selection period.

This is a rule specification to test, not a validated profitable strategy. Changing sessions, chart types or timeframes changes the range input. Use standard candles and consistent session settings for a reproducible comparison.