The implied volatility skew between each strike is an important measurement not only for options trading skew based strategies like vertical spreads, but for movement in the underlying stock. ORATS presents various ways to get a sense of the richness or cheapness of the skew.
Here's a picture of a typical strike skew for a month.
We have talked about modeling the skew in other posts, but today we want to get a sense of how to understand the valuation levels. We will focus on the most important measurement, the slope of the curve, the aqua colored line above.
The slope of the line with implied volatility on the y-axis and call delta on the x-axis is communicated as the percent the 10 delta lower strike is higher than the higher delta strike. Since the low strikes tend to have higher volatility than the high strikes, this makes the communication of the skew slope easier.
For example, for the ticker NVDA, the slope of the expirations around 30 days is 1.6% and the slope. The average of the long term slopes is about 2.2%. So, how do we know if this is low or high?
One way to assess the slope is to compare it to historical levels. Another way is to compare it to a related ETF. From the table below, we can see the important slope measurements from the ORATS General Core data. NVDA's slope is in the 8.7 percentile meaning that 91.3% of the time this year the slope was higher. The average slope for the year was 2.6% and the standard deviation of the slope was 0.8% so we can say that the current slope of 1.6% is 1.25 stdevs less than the mean.
We can also look at the slope of NVDA in comparison to its best related ETF, the XLK. The slope of NVDA/XLK = 30%, and that has been its ratio for the month, but the year ratio was 50% with a stdev of 20%. So we see that NVDA is 1 stdev below the mean relationship to the ETF.
These are some pretty killer measurements when looking at the skew slope. These are meticulous and intensive calculations that can give you an edge especially when trading options strategies with sensitivities to slope like verticals.
Here are the definitions for the above from our Data API documentation.
Here are some more skew articles: