Is It Safe to Fade the Latest VIX Rip?

Using the VXO as a VIX proxy, there are some similarities between 1987 and 2015

Aug 24, 2015 at 9:59 AM
facebook twitter linkedin

As I mentioned in Part 1 of this series, there is a way to go back further in time and look at volatility blasts, even in the absence of a volatility index. Long story short, before there was the CBOE Volatility Index (VIX), there was the CBOE OEX Implied Volatility Index (VXO). VXO was -- and is, as it still exists -- a volatility index that proxies 30-day implied volatility on S&P 100 Index (OEX) options. Prior to 2004, the index now known as VXO was known as ... VIX. The index now known as VIX didn't exist yet.

Think of VXO as "Original VIX." As I mentioned, it proxies volatility on OEX options, because back in the '80s and early '90s, OEX was the busiest option around. It also only uses near-the-money strikes in its calculation, whereas current VIX uses a much larger and varying number of strikes. Thus, current VIX factors in many high vol out-of-the-money puts, whereas original VIX did not.

So, how does VXO help us at all? The data stream goes back further, all the way to 1986 -- even though it didn't start as a published index until the early '90s. Thus, traders during the '87 crash didn't actually "see" original VIX hit 150 because it didn't exist yet. I'd note that from my experience as a floor trader from the late '80s into 2001, no one paid much attention to VIX anyway; you just knew volatility.

Anyway, here's Vol Pops: VXO Edition. It's similar, but not identical to, using VIX. That is to say, inferior methodology, but more data to work with. Here are all the days where VXO closed 40% above its 10-day simple moving average (SMA), as well as the one-month and three-month returns in OEX, if you bought the close those days.


Last Friday was actually the fourth biggest stretch ever -- though, as you can see, the first two were rather insane. Also, your performance timing a fade into the crash depended entirely on what day you faded -- which is a good argument for using an averaging strategy, if you are of a mind to fade now.

How about our "40%" strategy? Go long when VXO gets 40% above the 10-day, hold for one month or three months.


Well, now we have 15 separate incidents vs. eight when we just used VIX and only went back to 1993. And it includes a disaster trade from the '87 crash, as well as a mediocre one from 2008. Even with that, the one-month results are pretty much nothing. You won nicely eight times and won slightly twice, while losing four times. Out three months, we actually beat the market on the average results, 3.67% to 2.17%.

The scary similarity to now is that 1987 started out with a couple ugly days into expiration, followed by a disaster the day after expiration. Back in 1987, there were no weekly or dollar strikes, so part of the issue was a market cascading between strikes much further apart in both absolute and percentage terms.

Remember, on expiration, deltas go "binary" -- everything's either zero or 100. Short option squeezes forced selling into the next strike, and the next short option squeeze. Today, with weekly and tight strikes, tons more derivatives and like products, there's way more moving parts to play with. So it's not clear the similarity is that big a deal. The encouraging sign, though, is even including 1987 numbers, fading the biggest VIX rips didn't fare badly net-net.

Disclaimer: Mr. Warner's opinions expressed above do not necessarily represent the views of Schaeffer's Investment Research.

Now is the time to join our thriving community of Event Traders who consistently profit from every earnings season. With this discounted subscription opportunity, you'll stay ahead of the curve and seize opportunities others miss. Do not let Q3 earnings season pass you by – subscribe now and supercharge your portfolio with expert insights that turn market reactions into profit-generating opportunities!! Don't waste another second... join us right now before the next trade targeting +200% is released!