Thursday, April 8, 2010

The Relativity of Volatility

Apologies for the sparse posting as of late. My time has been eaten up by other endeavors leaving little room for nurturing my little corner of the blogosphere. Rest assured this next week will be more interesting as I've been tinkering with a few posts in my workshop of ideas. So how about kicking it off with a few volatility musings...

When analyzing volatility remember that there aren't really any absolutes. Take the VIX for example. While it would be nice to assert that the VIX is "low" when at 20 (cheap options) or "high" when at 30 (expensive options), the truth is its all relative. The VIX could be at a crazy low level like 10 and options could still be overpriced. On the other hand it could be in the stratosphere at 80 and options could still be underpriced. Whether implied vol was too high or too low when you entered an option trade depends on the volatility of the underlying throughout the duration of the trade. Suppose you think SPY options are cheap right now so you buy an ATM straddle. Currently the implied vol on the May 119 straddle is around 14.5%, which implies a little less than a 1% daily move should occur approximately 68% of the time in the SPY. That's a lot of numbers so read that last sentence again if you need to. Even though 14.5% vol seems cheap, what if the SPY is only able to muster up a .5% move on average each day (which coincidentally is pretty much happening right now)? Well, sorry charlie but you overpaid for that straddle. Though you looked at a low absolute number like 14% and made the assessment that options were a buy, reality is they were still a sell.

One lesson I can draw from my experience selling condors back in 2006 and 2007 is the fact that options' volatility can continue to be expensive even with the VIX sitting at barn burning low levels like 10. Between Oct 2006 and Feb 2007 the VIX found a nice home right around 10; a level which historically speaking is very low. How much volatility did the SPY actually realizes over the same time frame? About 7%, which made selling condors a lucrative proposition (click image to enlarge).

[Source: MachTrader]

This perhaps reiterates why it's beneficial to use a short term historical volatility reading like 10 or 21 days when assessing whether options seem cheap or expensive. It keeps us grounded in reality and serves as a useful benchmark for gauging what's happening in the here and now. While looking at the VIX now compared to 2008 certainly makes it seem like options are a steal of a deal, comparing it to recent realized volatility paints a different picture altogether.


Lluís said...

Hi Tyler,

As usual an excellent explanation of volatility nuances.

By the way, I've found an easy to use spreadsheet to calculate the underlying movement range for different standard deviations. It could be useful for your readers, I can send it to you.

Have a good weekend,


Tyler Craig said...

Thanks Siull. I actually have a similar spreadsheet that calculates the expected move over any time frame, but I'd be interested in taking a look at yours. My e-mail address is