What are Volatility Bands?
Volatility Bands (VBands) are price ranges (or "channels") where you could expect the price to remain "today",
based on information from "yesterday." VBands are calculated as Standard Deviations.
Standard Deviation
Standard Deviation is a way of measuring how close a particular value is from the average (mean average,
calculated by adding all values, and dividing by the number of values). In a normal distribution, values
are equally distributed above and below the mean. The graph of the number of data items at each value is
a normal or bell curve.
A common place to find a normal curve is in a poll or survey. Another common place to find a normal curve is
in probability calculations (like the count of heads versus tails for a coin toss).
As an example of a survey, if you asked 1000 people, "How many
total points (both teams) will be scored in the Super Bowl?," you might get answers in a range of 0 to 100 (after
eliminating "unreasonable" answers like -50 or 1000). Let's say that the mean average of the survey is 50.
Next, we graph the count of how many people voted for each value. Assuming a normal distribution,
the graph might look something like this:
This shows us that people generally thought that the score would be about 50 - with progressively less people
voting for scores further away from 50. Now, we apply a Standard Deviation to this graph to obtain information
about how people voted. This table shows the Standard Deviation, and the percentage of the normal
curve that is represented by that Standard Deviation.
| Standard Deviation | Percentage included | Normal Curve |
|---|---|---|
| 1.0 | 68% |  |
| 1.28 | 80% | |
| 1.5 | 87% | |
| 2.0 | 95% | |
| 3.0 | 99% |
Note - it is important to recognize that the bell curve can be more steep or more flat. If the consensus of values for the Super Bowl was more spread out (more "volatile"), then the normal curve would be flatter, and the Standard Deviation bands would have a wider range (for example, the 2.0 Standard Deviation may be from 10 to 90).
Applying Standard Deviation to Equities
Stock prices are statistically like "votes," indicating the prices that have been paid for the stock to be bought
and sold. By examining the range of the stock prices over one day, and graphing the prices based on how
many shares (volume) changed hands at that price, we get a curve which may, or may not, look like a normal curve.
In strongly trending markets, the curve may be almost flat - a steady increase or decrease in price with no major
volume spikes, for example. However, in a consolidating (sideways channeling) market where the prices go up and
down, the curve more closely is represented by a bell curve.
Applying Volatility Bands
Use the Implied Volatility Index for "yesterday's" option puts and calls (from
IVolatility.com) to calculate the Standard Deviation, which is applied
to "yesterday's" closing price to predict price ranges or channels for "today." As the Implied Volatility values
for puts and calls get further apart, the price ranges increase. When the Implied Volatility values are closer
to the same value (indicating that buyers and sellers of options are in closer agreement as to the future value),
the price ranges decrease. The VBand calculator determines price ranges based on this volatility.
How do you use VBand values?
As published by Kevin Haggerty and others, the Volatility Bands provide a price at which you might expect support
or resistance. For example, you could expect that 68% of the time (over MANY samples) that the price would stay
within the 1.0 Standard Deviation range (from Standard Deviation -1.0 to 1.0). You could expect that the price
would stay within the 2.0 Standard Deviation range 95% of the time. Just like Fibonacci Retrace levels, the VBand
price values provide a slightly more likely place where the price of the stock will reverse to stay within
the VBand. These values should NOT be treated as absolute brick-wall price barriers. However, the VBands will
provide a price range in which statistically the price will remain.
When trading stocks, you may find several price points at which you think that the stock will likely be supported
or provide resistance - that is, price points where the price is too far extended from where you think it should be,
and where it will likely reverse to get back to more of a "reasonable" value. You might calculate these price
points using Pivots, Fibonacci retrace values, VBands, trend lines, moving averages, previous support and resistance
levels, or any of many other techniques. As with any of these other price point calculations, you should always
look to other indicators for confirmation. Just because a price is approaching a VBand doesn't necessarily mean
that it WILL reverse. However, if it's also approaching a moving average, or a previous support or resistance
level, then your confidence level will increase.
This chart for Boeing (3 February 2003, 10 minute bars) shows how the Volatility Bands may offer areas of support and resistance. The leftmost three bars are the closing bars for the previous day. The VBands are calculated from the Put and Call Implied Volatility Index (from IVolatility.com) using data for "yesterday" - 2 February 2003. These VBands are projected over the chart for "today" in an effort to predict these areas of support and resistance.
The HamVBand Indicator displayed above in TradeStation is no longer available due to datafeed problems.
Final note
As with all other indicators, Volatility Bands are NOT the holy grail when considered alone. However, with
confirmation from other indicators that you use, VBands can provide additional price points to consider for
trade entry and exit.