814 Part VI: Measuring and Trading Volatility 
TRADING THE VOLATILITY PREDICTION 
The volatility trader must have some way of determining when implied volatility is 
sufficiently out of line that it warrants a trade. Then he must decide what trade to 
establish. Furthermore, as with any strategy- especially option strategies - follow-up 
action is important too. We will not be introducing any new strategies, per se, in this 
chapter. Those strategies have already been laid out in the previous chapters of this 
book. However, we will briefly review important points about those strategies and 
their follow-up actions where it is appropriate. 
First, one must try to find situations in which implied volatility is out of line. 
That is not the end of the analysis, though. After that, one needs to do some proba­
bility work and needs to see how the underlying has behaved in the past. Other fine­
tuning measures are often useful, too. These will all be described in this chapter. 
DETERMINING WHEN VOLATILITY IS OUT OF LINE 
There is much disagreement among volatility traders regarding the best method to use 
for determining if implied volatility is "out ofline." Most favor comparing implied with 
historical volatility. However, it was shown two chapters ago that implied volatility is 
not necessarily a good predictor of historical volatility. Yet this approach cannot be dis­
carded; however it must be used judiciously. Another approach is to compare today's 
implied volatility with where it has been in the past. This concept relies heavily on the 
concept of the percentile of implied volatility. Finally, there is the approach of trying 
to "read" the charts of implied and historical volatility. This is actually something akin 
to what GARCH tries to do, but on a short-term horizon. So the approaches are: 
1. Compare implied volatility to its own past levels (percentile approach). 
2. Compare implied volatility to historical volatility. 
3. Interpret the chart of volatility. 
In addition, we will examine two lesser-used methods: comparing current levels of 
historical volatility to past measures of historical volatility, and finally, using only a 
probability calculator and trading the situation that has the best probabilities of 
success. 
THE PERCENTILE APPROACH 
In this author's opinion, there is much merit in the percentile approach. When one says 
that volatility tends to trade in a range, which is the basic premise behind volatility trad-