Bollinger Bands (BB)

I. What is the Bollinger Bands Indicator

Bollinger Bands (BB) is a technical analysis indicator invented by American financial analyst John Bollinger in the early 1980s and officially published in 1983. In 2011, Bollinger trademarked the indicator.

Bollinger Bands belongs to the Volatility / Bands overlay indicator category and is plotted directly on the price chart. It consists of three lines:

• Middle Band: Simple Moving Average (SMA) of the closing price
• Upper Band: Middle Band + K times the standard deviation
• Lower Band: Middle Band - K times the standard deviation

The default parameters are period $n = 20$, standard deviation multiplier $K = 2$, and the data source is the closing price.

The core idea behind Bollinger Bands is that price fluctuations are cyclical, and most of the time prices move within the range of “mean +/- 2 standard deviations.” When volatility increases, the bandwidth expands; when volatility decreases, the bandwidth contracts. This “breathing” rhythm provides traders with extremely valuable market state information.

---

II. Mathematical Principles and Calculation

2.1 Core Formulas

Let the closing price series be $C_1, C_2, \ldots, C_t$, the period be $n$, and the standard deviation multiplier be $K$.

Middle Band:

$$\text{Middle}_t = \text{SMA}(C, n) = \frac{1}{n} \sum_{i=0}^{n-1} C_{t-i}$$

Standard Deviation:

$$\sigma_t = \sqrt{\frac{1}{n} \sum_{i=0}^{n-1} (C_{t-i} - \text{Middle}_t)^2}$$

Upper Band:

$$\text{Upper}_t = \text{Middle}_t + K \times \sigma_t$$

Lower Band:

$$\text{Lower}_t = \text{Middle}_t - K \times \sigma_t$$

2.2 Derived Indicators

Bandwidth:

$$\text{BandWidth}_t = \frac{\text{Upper}_t - \text{Lower}_t}{\text{Middle}_t} \times 100$$

Bandwidth quantifies the current volatility level. Extremely low bandwidth values typically foreshadow an upcoming large price movement (the “squeeze” signal).

%B Indicator:

$$\%B_t = \frac{C_t - \text{Lower}_t}{\text{Upper}_t - \text{Lower}_t}$$

$\%B$ represents the relative position of the price within the Bollinger Bands: $\%B = 1$ means at the upper band, $\%B = 0$ means at the lower band, and $\%B = 0.5$ means at the middle band.

2.3 Calculation Steps

1. Take the most recent $n$ closing prices
2. Calculate the simple moving average of these $n$ closing prices — this is the middle band
3. Calculate the population standard deviation of these $n$ closing prices
4. Upper Band = Middle Band + $K \times$ Standard Deviation
5. Lower Band = Middle Band - $K \times$ Standard Deviation

---

III. Python Implementation

import numpy as np
import pandas as pd

def bollinger_bands(close: pd.Series, period: int = 20, std_dev: float = 2.0) -> pd.DataFrame:
    """
    Calculate Bollinger Bands (BB).

    Parameters:
        close  : closing price series
        period : moving average period, default 20
        std_dev: standard deviation multiplier, default 2.0

    Returns:
        DataFrame with columns: middle, upper, lower, bandwidth, percent_b
    """
    # Middle band: simple moving average
    middle = close.rolling(window=period).mean()

    # Population standard deviation (ddof=0)
    std = close.rolling(window=period).std(ddof=0)

    # Upper and lower bands
    upper = middle + std_dev * std
    lower = middle - std_dev * std

    # Bandwidth
    bandwidth = (upper - lower) / middle * 100

    # %B
    percent_b = (close - lower) / (upper - lower)

    return pd.DataFrame({
        'middle': middle,
        'upper': upper,
        'lower': lower,
        'bandwidth': bandwidth,
        'percent_b': percent_b
    })


# ============ Usage Example ============
if __name__ == '__main__':
    np.random.seed(42)
    n_days = 100

    # Generate simulated OHLCV data
    base_price = 100 + np.cumsum(np.random.randn(n_days) * 0.5)
    df = pd.DataFrame({
        'open':   base_price + np.random.randn(n_days) * 0.3,
        'high':   base_price + np.abs(np.random.randn(n_days) * 0.8),
        'low':    base_price - np.abs(np.random.randn(n_days) * 0.8),
        'close':  base_price,
        'volume': np.random.randint(1000, 10000, n_days)
    })

    # Calculate Bollinger Bands
    bb = bollinger_bands(df['close'], period=20, std_dev=2.0)
    result = pd.concat([df, bb], axis=1)

    print("=== Bollinger Bands Results (last 10 rows) ===")
    print(result[['close', 'middle', 'upper', 'lower', 'bandwidth', 'percent_b']].tail(10).to_string())

    # Generate trading signals
    result['signal'] = np.where(result['close'] < result['lower'], 'Oversold',
                       np.where(result['close'] > result['upper'], 'Overbought', 'Neutral'))
    print("\n=== Signal Statistics ===")
    print(result['signal'].value_counts())

---

IV. Problems the Indicator Solves

4.1 Volatility Measurement

The bandwidth of Bollinger Bands directly reflects the current volatility level of the market. When bandwidth contracts to a historical low (i.e., the “Bollinger Band squeeze”), it often foreshadows an upcoming large price movement.

4.2 Overbought / Oversold Detection

• Price touches or breaks above the upper band: Possibly overbought, with short-term pullback risk
• Price touches or breaks below the lower band: Possibly oversold, with short-term rebound potential

4.3 Typical Trading Strategies

1. Bollinger Band Squeeze Strategy: Enter in the direction of the breakout when bandwidth expands after contracting to a recent minimum
2. Mean Reversion Strategy: In range-bound markets, go short when the price touches the upper band and go long when it touches the lower band, with the middle band as the profit target
3. Double Bollinger Band Strategy: Use both 1x and 2x standard deviation bands simultaneously, dividing the price space into three zones (trend zone, buffer zone, reversal zone)
4. W-Bottom and M-Top Patterns: A pattern recognition method specifically recommended by John Bollinger — a W-bottom near the lower band is a buy signal, and an M-top near the upper band is a sell signal

4.4 Combining with Other Indicators

• Combine with RSI to confirm overbought/oversold signals
• Combine with volume indicators to confirm breakout validity
• Combine with Keltner Channels to detect the “squeeze” signal

---

V. Advantages, Disadvantages, and Use Cases

Advantages

Advantage	Description
Adaptive volatility	Bandwidth automatically adjusts with market volatility — no need to manually switch parameters
Intuitive and easy to understand	The relationship between price and the upper/lower bands is immediately clear
Versatile	Can be used for both trend identification and mean reversion trading
Solid statistical foundation	Based on standard deviation theory with clear probabilistic meaning

Disadvantages

Disadvantage	Description
Lagging	Based on SMA and historical standard deviation — inherently a lagging indicator
Normal distribution assumption	Financial market return distributions typically have “fat tails,” so actual breakout frequency exceeds theoretical values
False signals in trending markets	In strong trends, price may persistently “walk the band,” causing mean reversion strategies to incur repeated stop-losses
Not reliable alone	Needs to be combined with other indicators or pattern analysis for confirmation

Use Cases

• Best suited for: Mean reversion trading in range-bound markets; identifying breakouts after volatility compression
• Moderately suited for: Trend following in combination with trend indicators
• Not suited for: Contrarian trading alone in strongly trending markets

Comparison with Similar Indicators

Indicator	Volatility Measurement	Characteristics
Bollinger Bands	Standard deviation	Sensitive to extreme values; more dramatic bandwidth changes
Keltner Channels	ATR	Smoother; less sensitive to extreme values
Envelopes	Fixed percentage	Does not adjust with volatility; simplest but least flexible