Bollinger Bands

Quick Reference

Property	Value
Dimension	technical
Category	technical
Version	v0.9.0 (Beta)
Output Column	bollinger_pct_b

Bollinger Band %B: (price - lower) / (upper - lower) - normalized price position within volatility bands

Formula

(price - (rolling_mean(price, window) - (rolling_std(price, window) * k))) / ((rolling_mean(price, window) + (rolling_std(price, window) * k)) - (rolling_mean(price, window) - (rolling_std(price, window) * k)))

CDM Inputs

Column	CDM Table	Description
price	cdm_*	CDM source table

Parameters

Parameter	Type	Default	Description
window	integer [2, 500]	20	Window for moving average and standard deviation
k	float [1.0, 5.0]	2.0	Number of standard deviations for band width

Output

Column: bollinger_pct_b

%B indicator: 0=at lower band, 1=at upper band, <0 or >1=band breakout

Market Intuition & Trading Rationale

Bollinger Bands %B normalizes price position within volatility envelopes: (price - lower) / (upper - lower), where lower = SMA - k×σ and upper = SMA + k×σ. %B = 0 means price is at the lower band; %B = 1 means at the upper band; %B > 1 means price has broken above the upper band (overbought breakout); %B < 0 means broken below the lower band (oversold breakdown).

%B is more useful than raw band values because it's normalized — it works across instruments and time periods without recalibration. The default parameters (20-period window, k=2) create bands that contain approximately 95% of price observations under normality. The window controls responsiveness (shorter = faster, noisier) and k controls band width (larger k = wider bands, fewer touches).

In range-bound markets, price tends to oscillate between %B = 0 and %B = 1 — buy at the lower band, sell at the upper. In trending markets, price can "ride the band" — %B stays above 0.8 in uptrends or below 0.2 in downtrends for extended periods. The %B value distinguishes these two regimes in real time.

Usage Cases

• Mean reversion in ranges: Buy when %B < 0 (below lower band), sell when %B > 1 (above upper band). Only do this in confirmed range-bound markets — in trends, band touches are continuation signals, not reversal signals.
• Breakout confirmation: %B sustained above 1.0 for multiple bars after being range-bound → genuine bullish breakout. %B briefly crossing 1.0 and immediately reverting → false breakout (bull trap).
• The Squeeze: When band width (upper - lower) reaches a multi-period minimum, the market is "squeezing" — a period of exceptionally low volatility that typically precedes a large directional move. %B's behavior after the squeeze resolves indicates direction.

YAML Definition

name: bollinger_bands
description: 'Bollinger Band %B: (price - lower) / (upper - lower) - normalized price
  position within volatility bands'
category: technical
version: v0.9.0 (Beta)
dimension: technical
status: Pre-release
required_inputs:
  - price
output_column: bollinger_pct_b
output_description: '%B indicator: 0=at lower band, 1=at upper band, <0 or >1=band
  breakout'
parameters:
  window:
    type: integer
    description: Window for moving average and standard deviation
    required: false
    default: 20
    constraints:
      min: 2
      max: 500
  k:
    type: float
    description: Number of standard deviations for band width
    required: false
    default: 2.0
    constraints:
      min: 1.0
      max: 5.0
formula: (price - (rolling_mean(price, window) - (rolling_std(price, window) * k)))
  / ((rolling_mean(price, window) + (rolling_std(price, window) * k)) - (rolling_mean(price,
  window) - (rolling_std(price, window) * k)))