Strap in my fellow volatility junkies, let’s tackle an interesting topic today.  One of the most common and persistent questions I get asked has to do with XIV and strategy performance during the financial crisis.

How would it have done?  How would our strategy have done?  In order to attempt to answer these questions, first we’re going to have to talk about the difference between:

XIV live trading results   vs   XIV simulated data
 For those that don’t know, the XIV launched as a tradable product on November 30th, 2010 which means it really hasn’t been around very long.  For anybody who focuses on quant type strategies like we do, that is a very short data set so of course we have to be mindful of how weak our data really is and be very careful not to “overfit” or “curve fit” when testing strategies.

And for most people, they hadn’t even heard of the XIV before a couple years ago.  For me I have over 10 years of derivatives experience so when the VXX launched in 2009, as a volatility trader it just made sense and I jumped in.  I was trading VXX in 2010 before there was such thing as the XIV, but for most people this is a relatively new interest.

Have you ever heard of the Baader-Meinhof phenomenon?
I don’t mean the militant German terrorist group.  The Baader-Meinhof phenomenon is essentially when something is unfamiliar to you at first, but once an object, a subject, or a piece of information catches your attention you notice it a lot more often and it repeatedly shows up afterwards.
For years people hadn’t ever heard of the XIV. Then they read a few articles, see the recent performance, and now it’s all they see.  XIV this, volatility that, it’s all they see now and the volume on it is expanding rapidly.

But is the XIV more special now that it was 2 years ago?  Since March 1st of 2016 markets have been incredibly calm.  We’ve seen record smashing low volatility and the S&P 500 is currently in it’s longest streak ever without even a 3% correction.  Needless to say the XIV has done very well.

But so have triple leveraged equity ETFs, yet there isn’t an excess of articles and interest in them.  The TQQQ  (triple leveraged Nasdaq)  has outperformed the XIV on both an absolute and risk adjusted measure since the XIV launched, yet nobody is talking about that.  And I might point out how few people would be willing to buy TQQQ aggressively right now, yet they can’t get enough of XIV

Baader-Meinhof  πŸ™‚

Anyway, the XIV launched as a tradable product on November 30th, 2010 so our live trading data set is very small.  But remember it doesn’t derive it’s price based on supply and demand dynamics like other stocks and ETFs.  Instead it derives it’s price based on a known method of rolling the front two months of the VIX futures.  Since VIX futures did exist going back to 2004 we can actually simulate what the XIV would have done during those years even though it didn’t exist.

To each his own, but to my eye that’s pretty wild.  18 months ago it was in the teens.  It’s only because of this recent melt up in the markets that people have taken notice and of course that was not a guarantee.  Markets could easily have done the opposite and most people would have gone on their merry way without noticing XIV.

But those numbers should only be taken as a rough guide because they may not be accurate.

In the past if you wanted to hedge a portfolio or trade volatility there were the standard choices.  You could trade options on the stock market indexes like the S&P 500 for example, or you could directly trade VIX Options or VIX Futures.

Now with the introduction of these exchange traded volatility products, many investors are bypassing those and going straight to owning the much more accessible XIV or VXX for example.  And not in small amounts either.  There are tens of millions of shares being traded daily.  When you add up the trading volume on the more popular ones like XIV, VXX, SVXY, UVXY, ZIV, TVIX, VMIN just to name a few, the volume is extremely heavy.

If products like the XIV actually existed in 2004, that would have effected the buying behavior of market participants.

It would have changed the composition of VIX futures being traded.  It would have altered the volume of VIX options and S&P 500 options.  The VIX futures that the XIV hold may very well have been different, which would have changed the prices of the volatility products themselves had they existed.  The VIX index itself has likely been affected as well.  These are derivatives based on other derivatives based on other derivatives.  We can’t just retroactively throw in the XIV with 0 trading volume, it doesn’t work that way.

In the end, it’s possible that the XIV prices would not have been what they are in those simulated backtests.

Nobody knows how close or far away they would have been, but we can say they likely would not have been the same which for me is enough to discount them.  Any uncertainty in data isn’t worth using.

Our data set is already incredibly small, so it would not be wise to add in 7 years of tainted data to an already very short 7 year data set.  That could easily lead to curve fitting and strategies that fail in live trading going forward.  So getting back to the original question, and the one that fills up my email inbox.

How would our strategy have done during the financial crisis?

Using a potentially  (and likely)  tainted data set I will reluctantly share a few charts.

From 25,000 to 36 million is pretty hard to see on a 14 year scale so we can break it up into just the pre XIV launch period.  Of course for post XIV launch you can just view it on my website here.

Everything looks pretty normal and comparable to the post XIV launch and live trading results, except for that short period during the full on meltdown during the financial crisis.  Let’s zoom in on that 2 year period.  That’s the million dollar question that everybody wants to know.  How will volatility strategies do during a crisis?

Volatility markets tipped their hand over a year before Lehman collapsed so we “would have” utilized a lot of cash during that period.  Remember we only trade the high conviction signals on both the long and short volatility sides, and leave all that ambiguous middle ground to the day traders.

Again let’s be very careful when analyzing this period for two reasons.

1)  The data set may very well be tainted and inaccurate.  Like I said things get pretty messy when we try to retroactively jam a product into a chain of dependent derivatives.  Take those with a rather large grain of salt.

2)  The financial crisis, while scary, was actually very orderly with respect to volatility and the VIX Index.  It was quite a consistent, predictable, steady increase from the teens to a peak of 96.  This was an absolutely perfect volatility spike for products like the VXX, VXZ, UVXY depending on your risk tolerance.  Had they existed, all three of them “would have been” as easy profit as it comes.

But is it likely that the next recession manifests in the same way as the last one?  Maybe, but probably not.  It’ll have it’s own characteristics and quirks which is why it’s extremely important to have a strategy that isn’t curve fitted to a tiny data set.

It’s big picture only here.  This isn’t about tweaking threshold indicators down to produce the most beautiful curve fitted backtest possible on paper.  It’s about navigating inherently dangerous products in real time and turning them into safer investable assets.  But to do that it requires patience, discipline, and a long-term conservative mind set.

These products, my business, and the markets in general aren’t going anywhere.  Don’t ever feel like you’ve “missed out” on anything.  We have years, decades to do this, there’s no urgency.

As you can see in our live record as well as the longer 14 year backtest there were plenty of times when our strategies hit the volatility markets aggressively and took out a very healthy profit.  There’s also times when we will be very cautious and hold cash, sometimes for extended periods of time waiting for the best moment to strike.

Let’s not let recency bias or the Baader-Meinhof phenomenon push us into making unforced errors.

Why do I feel like my email inbox is going to be flooded with follow up questions?  Bring them on! 


Current VTS Total Portfolio Solution Allocations

25% VTS Tactical Volatility Strategy

50% VTS Tactical Balanced Strategy

25% VTS Iron Condor Strategy

VTS Conservative Vol StrategyΒ  – Β Optional replacement for lower risk tolerance investors


Want to see our current positions and daily trade signals?

Β  Consider Subscribing Β  – Β  Free 2 week Trial for all subscriptions

Β  Sign up for the Awesome VTS Newsletter Β – Β Free 1 month access

The Volatility Advisor

Making volatility investable through education and active management

Limited Offer

Start your Β FREE 2 Week Trial Β Now

Gain full access to all 4 of our successful trading strategies for a full two weeks, absolutely free.

It's time to start growing your retirement fund