Bollinger Bands consist of two trading bands, placed at 2 standard deviations above and below a simple moving average of the underlying instrument chart. The standard deviation of the prices themselves over recent price history is used since it constitutes perhaps the simplest and most robust measurement of price volatility.
Using 2 standard deviations ensures that 95% of the price data will fall between the two trading bands thus visually distinguishing the extremes that fall outside. For the construction of the bands, John Bollinger used a default moving average period of 20.
Bollinger Bands were developed in an attempt to improve the concept of fixed-width trading bands such as “Envelopes”. Contrary to fixed-width bands, Bollinger Bands expand and contract based on the volatility of the prices. Bollinger selected the standard deviation as a measure because the calculation involves squaring from the average, thus making the system more responsive to short-term price changes.
Bollinger Bands were not intended for producing trading signals. The system’s primary purpose is to indicate whether prices are relatively too high or too low and then a trader could turn to other indicators for trade signalling.
The Bollinger Bands are comprised of three elements:
- The Upper Band (BU)
- The Middle Band (SMA20)
- The Lower Band (BL)
Middle Band = SMA20Days
Upper Band = BU = SMA20Days + (2 * S20Days)
Lower Band = BL = SMA20Days - (2 * S20Days)
Where SMA20Days is the 20-Day simple moving average of the underling instrument price and S20Days is the 20-Day standard deviation of the underling instrument price.
A full guide to the calculation and construction of the Bollinger Bands is provided in excel format for your better understanding.
Bollinger Bands (BB)
Interpretation & Trade Signals
Given that Bollinger Bands were not intended for producing trade signals, the best way to exploit their role in a trading system is in combination with other indicators or chart patterns. An excellent combination in this respect is with the RSI since it is a countertrend oscillator and therefore aims to identify possible market tops and bottoms.
When price touches the upper Bollinger Band and RSI is below 70 and rising, it is an indication that the trend will continue. If however price touches the upper Band and the RSI is above 70 or possibly even higher (in an overbought state that is), the trend may reverse and decline (example provided below).
Conversely, when price touches the lower Bollinger Band and the RSI is above 30 and retreating, it indicates the trend should continue. If however a price touches the lower Band and the RSI is below 30 and falling (in an oversold state), it indicates the trend may reverse and move upward.
There are some basic interpretation guidelines that Bollinger himself considers a trader needs to keep in mind to make the most out of Bollinger Band:
- A. A move that originates at one band tends to go all the way to the other band, setting therefore an indicative price target.
- B. When volatility lessens and the bands tighten (come together), it is usually a sign that a significant market move will result in one or the other direction.
- C. When volatility increases and the bands are unusually far apart while at the same time there is evidence of a top/bottom formation, it is often a sign that the market has moved too far and that the current trend is ending.
- D. When prices move outside the bands, it implies a continuation of the current trend.
- E. Bottoms and tops formed outside the bands followed by a failure to achieve the same extreme twice and re-entering the bands call for possible reversals in trend.
Long Entry signal: RSI14 < 50 and there is an apparent positive failure swing see RSI Analysis). At the same time the price chart forms a bottom outside the bands and moves above BL . Enter long when the price retests its low and remains within the bands, failing to form a lower low.
Change of trend warning: RSI14 > 70 and the price chart moves above BU Possible top forming.
Long Exit signal: RSI14 < 70 and there is an apparent negative failure swing forming. At the same time the price chart forms a peak outside the bands and then moves below BU Exit position when the price chart fails to rise higher than the previous peak and remains within the bands.
Tips and ideas
The standard deviation value used for constructing the Bollinger Bands can be varied to alter the sensitivity of the system. Some traders increase the value of the standard deviation from 2 to 2.1 deviations away from the moving average when using longer averages such as 50-Day. Conversely, may technicians lower the value of the standard deviation from 2 to 1.9 deviations when using short moving averages such as 10-Day.
As per John Bollinger’s recommendations, moving averages of less than 10 periods in length don’t work very well with Bollinger Bands.
Two extremely useful measures that are key to the development of trading systems via the linking of price and indicator action are %b and BandWidth. Both of these indicators were introduced by John Bollinger himself and are calculated as follows:
%b = (Last - BL ) / (BU - BL)
BandWidth= (BU - BL) / (BMiddle)
When day-trading using Bollinger Bands , pay particular attention to volatility-breakouts. Bollinger Bands’ logic requires a “squeeze” for a volatility-breakout to be traded.
When day-trading, Bollinger Bands can be used as references of overbought/oversold levels. Rallies that carry far above/below the upper/lower band should be sold/bought at the first sign of weakness.
The biggest strength of Bollinger Bands lies in their construction as it creates a volatility measuring system driven by the market itself, rather than trying to read market movements by imposing an external discipline.
Bollinger Bands adapt dynamically to price changes as volatility increases and decreases, rather than depending on any secondary source of information.
Using a single moving average of a rolling time interval introduces a lag, causing our bands to respond with a time-lag.
Also, measuring the standard deviation of this single moving average and therefore measuring the volatility of the market, means that volatility rises when prices rise/fall quickly. This will be reflected on the bands by moving far apart from each other. However even if market volatility declines after that, the volatility represented by the standard deviation will not decline as fast as expected, until it passes completely out of the calculation window. This will make it difficult for an exit signal to be delivered on time, since it is contingent on reduced volatility. Also, if a new high volatility event takes place soon after, it will not be easily recognised due to the inherent volatility in our system.
Suggested Combinations for EA development
Relative Strength Index - RSI.
Chart reversal patterns as sign of exhaustion.
Moving Average Convergence Divergence (MACD).
Price - Moving Average crossover signals can be used for confirmation help.
Basic trend analysis can help filter crossover signals.
Parabolic SAR can be used for creating Stop orders.
Use in FxPro Quant
Bollinger Bands can be found under the ‘Indicators’ group in our FxPro Quant menu. With FxPro Quant, we do not have to worry about the calculations and definitions behind the indicator. We just need to drag & drop the indicator in the main window and set the parameters we want to test.
Input: e.g. EurUsd, GbpYen | Default: Current
If left empty, the instrument symbol used in the relevant chart in MT4 will relate by default.
Input: 1m,5m,15m,30m,1h | Default: Current
If set to “current”, the time frame used in the relevant chart in MT4 will relate by default.
Input: number | Default: 20 periods
Number of periods used in the calculation formula to derive the Bollinger Bands (see excel spread sheet attached).
Input: number | Default: 2
Standard Deviations value used to derive the bands.
Input: number | Default: “0”
Shifting the bands vertically to serve the EA logic.
Please note that Bands Shift=1 means shifting the Bands one period forward (in the future).
Input: Close, Open, High, Low, Median, Typical, Weighted price. | Default: Close Price
Value used to derive the bands.
Input: Upper Line, Lower Line. | Default: Upper Line
Indicator value required for the Expert Advisor system in question.
Input: Number | Default: “0”
Number of periods the output should be shifted back or forth to serve the EA logic.
Please note that Bands Shift=1 means shifting the Bands one period backwards (in the past).
John Bollinger – Bollinger on Bollinger Bands, 2002
John J. Murphy – Technical Analysis of the Financial Markets
Steven B. Achelis – Technical Analysis from A to Z
Robert D. Edwards & John Magee – Technical Analysis of Stock Trends