Get The Ticker Tape delivered right to your inbox.


Putting My Foot Down. What's With the VIX Stiletto Spike?

December 19, 2014
Putting My Foot Down. What's With the VIX Stiletto Spike?

My golden rule of trading is: “If something looks too good to be true, you should do whatever you possibly can do to try to take advantage of it.”

Because that always works out. Like, never. But that’s also why when I see something that doesn’t look quite right, I literally can’t sleep until I can figure out what’s going on. 

And it wasn’t just me. I talked to our most tenured guys in the office. Former floor traders, data scientists. No one really knew what was happening. Take a look at what made me so curious (see figure 1).



An intraday chart of the CBOE VIX Index on December 17 shows a sharp “Stiletto Spike” higher. Source: TD Ameritrade’s thinkorswim® platform. For illustrative purposes only.

There’s not really a name for this phenomenon. We reached out to the CBOE and they told us that all the data was correct. But it just didn’t look right. The VIX was Stiletto Spiking™ all over the place. (And yes, that is a technical term I have given this event. Soon male traders all over Wall Street will be saying things like “the VIX is Stiletto Spiking™ again.”) Since then, several articles have been published, but no one seems to know exactly why these events were occurring. 

The first thing you have to understand is how the VIX is calculated. The CBOE published a whitepaper designed to help people understand the calculation.

It's some real "Goodwill Hunting"-style math, if you know what I mean. But I can simplify it. The VIX calculation measures 30-day expected trade volatility of the S&P 500 Index. The primary components of the VIX calculation are near- and next-term out-of-the-money put and call options. And the VIX is calculated based on the midpoint of the spread of those options.

In September, the VIX calculation was modified to include weekly options. When the VIX started Stiletto SpikingTM, a lot of traders were asking me whether it had anything to do with the change in the calculation to include weekly options. My opinion is that the inclusion of weeklys in the VIX was a good thing. Weeklys have tighter spreads and by including weeklys, the VIX is now calculated using more data. That’s a good thing.

In order to figure out why the VIX was spiking, we needed to look at the options in the seconds leading up to a VIX calculation. The VIX recalculates every 15 seconds. But not everyone has access to that kind of data. Nanex, LLC was able to provide us with the full SPX option chain raw data including bid, ask, series, and strike for all strikes and all option series for a full minute on either side of the highest Stiletto Spike of 25.20 – the high of the day on Tuesday at 9:14 AM. With this data, we could look more closely at what was going on.

We started by looking at the VIX prices before, during, and after that 25.20 print. At 9:14:04 a.m. the VIX was in line with what we would expect to see based on historical VIX price behavior. The very next print­—15 seconds later—Stiletto Spiked to our high of the week.

Stay with me here.

Each of the lines below represent a second in time (see figure 2). The X axis is the out-of-the-money strike price for the front term of the SPX options. The Y axis represents the price spread of the bid and ask. You can see that at 9:14:03, 9:14:04, and 9:14:05 everything looked fairly normal. But at 9:14:06, the otherwise tight bid/offers in the 1955 put strike disappeared and quote widened significantly. The VIX uses the midpoint between the bid and the ask.



Spreads on the out-of-the-money, near-term SPX puts from 9:14:03-9:14:06 a.m. on December 17. Source: Nanex, LLC. For illustrative purposes only.

Let’s look at the same data on the call side (see figure 3).



Spreads on the out-of-the-money near-term SPX calls from 9:14:03-9:14:06 a.m. on December 17. Source: Nanex, LLC. For illustrative purposes only.

You can also see a tremendous widening of the spread on the 1980 strike call.

A few articles have been published which suggest this was likely attributed to an algorithm adjustment and that the issue has since been corrected. But really, that’s speculation. We really have no way of knowing what caused it.

TD Ameritrade clients have been voraciously trading volatility lately. So we knew we needed to get them some answers about this event. I could tell from Twitter interactions on Tuesday that there was a lot of curiosity and thought we would share our findings with the broader trading community.

More than ever, retail has been trading volatility like an asset class. I felt like everyone—myself included—had an opportunity to better understand what was going on Monday and Tuesday.

Hope this helps.

Talk soon,



Scroll to Top