Connors RSI (CRSI) - Built-in Indicator Guide

A composite oscillator that averages RSI of price, RSI of directional streaks, and PercentRank of rate of change.

Overview

Connors RSI (CRSI) is a normalized oscillator that blends three distinct measures of price momentum and trend strength. It combines a standard RSI of closing prices, an RSI of a custom up/down streak counter, and a PercentRank of the one-period rate of change. The result is a single value between 0 and 100 that aims to smooth out noise while remaining sensitive to directional persistence.

The indicator is drawn as a blue line on a fixed scale, with a gray background band between 30 and 70 (light aqua fill) and a gray line at 50. It is typically used to identify overbought (above 70) and oversold (below 30) conditions in trending or mean‑reverting markets, where the streak component adds extra weight to sustained moves.

How it works

1. Compare current close to previous close to determine direction (up, down, or equal).
2. Update a mutable series ud that counts consecutive moves in the same direction: reset to ±1 on a direction change, otherwise increment the streak.
3. Compute a standard RSI of closing prices over len_rsi bars.
4. Compute a second RSI of the ud streak series over len_up_down bars.
5. Compute the PercentRank of the 1‑period rate of change (ROC) over len_roc bars.
6. Average the three components to produce the final CRSI value for the current bar.

Mathematical model

Where RSI is the standard Relative Strength Index and PercentRank is the rank‑based percentile normalisation provided by the platform’s library.

Logic flow

flowchart TD
    A["Get close[0], close[1]"] --> B{"close[0] > close[1]?"}
    B --Yes--> C["ud[1] <= 0?"]
    C --Yes--> D["ud[0] = 1"]
    C --No--> E["ud[0] = ud[1] + 1"]
    B --No--> F{"close[0] == close[1]?"}
    F --Yes--> G["ud unchanged"]
    F --No--> H{"ud[1] >= 0?"}
    H --Yes--> I["ud[0] = -1"]
    H --No--> J["ud[0] = ud[1] - 1"]
    D & E & G & I & J --> K["Compute rsi, ud_rsi, pr"]
    K --> L["Return (rsi + ud_rsi + pr) / 3"]

Parameters

Code walkthrough

Up/Down Streak Logic

Lines 19-31 of Connors RSI.indie5:

    is_equal = self.close[0] == self.close[1]
    is_growing = self.close[0] > self.close[1]
    ud = MutSeriesF.new(0)  # why not mut_series(init=0)
    if is_growing:
        if nan_to_zero(ud[1]) <= 0:  # why not equal nan_to_zero(ud)[1]
            ud[0] = 1
        else:
            ud[0] = nan_to_zero(ud[1]) + 1
    elif not is_equal:
        if nan_to_zero(ud[1]) >= 0:
            ud[0] = -1
        else:
            ud[0] = nan_to_zero(ud[1]) - 1

This block determines the direction of price change and updates the ud mutable series. If price is growing and the previous streak was non‑positive, the streak resets to 1; otherwise it increments. For falling prices the logic mirrors with negative values. Equal prices set ud to 0. The nan_to_zero helper ensures that a NaN previous value is treated as zero.

Component Calculations

Lines 33-35 of Connors RSI.indie5:

    rsi = Rsi.new(self.close, len_rsi)[0]
    ud_rsi = Rsi.new(ud, len_up_down)[0]
    pr = PercentRank.new(Roc.new(self.close, 1), len_roc)[0]

Three normalised components are computed: a standard RSI of the close price, an RSI of the ud streak series, and a PercentRank of the one‑period rate of change. Each uses a configurable lookback length. The [0] index extracts the current value from the series returned by the .new() factory.

Return Value

Lines 36-36 of Connors RSI.indie5:

    return (rsi + ud_rsi + pr) / 3

The final CRSI value is the simple average of the three components. Because each component is already scaled between 0 and 100, the result also falls in that range and can be plotted directly with the defined bands and levels.

Reading the chart

• The blue line oscillates between 0 and 100.
• Values above 70 (top of the gray band) suggest overbought conditions; values below 30 (bottom of the gray band) suggest oversold conditions.
• The gray line at 50 acts as a neutral midline.
• Because CRSI averages three different normalised signals, it may react more slowly to isolated price jumps but strengthens during sustained trends due to the streak component.
• The PercentRank of ROC adds a rank‑based perspective, making the indicator less sensitive to extreme price spikes than a pure RSI.

Implementation notes

• The ud mutable series (MutSeriesF) retains state between bars; its initial value is 0, and it is updated in place using ud[0] and ud[1].
• nan_to_zero is applied to ud[1] before comparisons to avoid NaN propagation when the series has not yet been fully initialised.
• All three .new() calls return series objects; the [0] index extracts the value for the current bar, while [1] would give the previous bar’s value.
• The @band and @level decorators only affect the visual overlay; they do not influence the computation.

FAQ

How do I adjust the sensitivity of CRSI?

Increase len_rsi and len_roc to smooth the indicator, or decrease them to react faster. The len_up_down parameter controls how many bars the streak RSI looks back.

What does the ud series represent?

It is a counter that tracks consecutive bars moving in the same direction. It resets to ±1 when the direction changes and increments (or decrements) during a streak, giving extra weight to sustained moves.

Can I use CRSI on timeframes other than the chart’s base timeframe?

The code does not include a sec_context or calc_on decorator, so it runs on the chart’s primary timeframe. To apply it to a different timeframe you would need to add the appropriate decorator.

Full source code

Indie Script v5, the built-in indicator as shipped with TakeProfit. Add it from the indicators menu or copy the code into the platform’s script editor.

# indie:lang_version = 5
from math import isnan
from indie import indicator, format, param, band, color, level, plot, MutSeriesF
from indie.algorithms import Rsi, PercentRank, Roc


def nan_to_zero(val: float) -> float:
    return 0 if isnan(val) else val


@indicator('CRSI', format=format.PRICE)  # Connors RSI
@param.int('len_rsi', default=3, min=1, title='RSI Length')
@param.int('len_up_down', default=2, min=1, title='UpDown Length')
@param.int('len_roc', default=100, min=1, title='ROC Length')
@band(30, 70, line_color=color.GRAY, fill_color=color.AQUA(0.1), title='Background')
@level(50, line_color=color.GRAY(0.5), title='Middle Band')
@plot.line(color=color.BLUE, title='CRSI')
def Main(self, len_rsi, len_up_down, len_roc):
    is_equal = self.close[0] == self.close[1]
    is_growing = self.close[0] > self.close[1]
    ud = MutSeriesF.new(0)  # why not mut_series(init=0)
    if is_growing:
        if nan_to_zero(ud[1]) <= 0:  # why not equal nan_to_zero(ud)[1]
            ud[0] = 1
        else:
            ud[0] = nan_to_zero(ud[1]) + 1
    elif not is_equal:
        if nan_to_zero(ud[1]) >= 0:
            ud[0] = -1
        else:
            ud[0] = nan_to_zero(ud[1]) - 1

    rsi = Rsi.new(self.close, len_rsi)[0]
    ud_rsi = Rsi.new(ud, len_up_down)[0]
    pr = PercentRank.new(Roc.new(self.close, 1), len_roc)[0]
    return (rsi + ud_rsi + pr) / 3