As you can see in the following chart, the ProShares Ultra VIX Short-Term Futures ETF (UVXY) is trading higher in one of the largest rallies seen since the crisis earlier this year.
I am currently bearish the VIX over the next 4 weeks but bullish the VIX through year-end. However, due to UVXY’s specific instrument concerns, I am bearish the long-term prospects of the ETF and believe that this rally makes for a strong selling and/or shorting opportunity.
The reason why this is important to note is that it can help frame up the current movements of the VIX. Since we started this movement from around the long-run average level of the VIX (a little over 19), we can safely say that the mean-reversion trade at work from April through August had run its course. That is, the market had reverted to the mean and therefore traders should start to be more open to movements against the mean (as we are seeing now).
What this means for the VIX traders of today is that we can now start to look at mean-reversion trades once again and with greater confidence. For example, here is a chart that shows the probability that the VIX is higher 1 month into the future grouped by the level of the VIX at any given time.
This chart uses over 27 years of data which makes the underlying dataset fairly robust. What this data shows is that given that we’re seeing the VIX in the territory of 28-30, there’s about a 70% chance that the VIX will fall somewhat over the next month. For an idea as per the magnitude, historical analysis would suggest that we will see about 3 points (or 10% at today’s level) reduced from the index as an average expectation.
In other words, at this moment, I am bearish the VIX over the next month. However, I believe that seasonality factors suggest that while we may see a correction in the VIX over the next 20 market days by around 10%, the numbers also suggest that between now and the end of the year, we can expect a general increase in the level of the VIX.
Put simply, the data indicates that volatility tends to increase from September through October and the last quarter of the year tends to be the most volatile time period of the year.
And another angle to consider is that the market is still fairly technically extended and the current setup appears to suggest further downside over the next few weeks.
From a classic technical standpoint, the above chart tells me that we’re seeing a trend failing at the time of writing. Price is breaking through the ascending trendline on growing bearish momentum and a conservative price target would be the prior resistance levels of June and July. Markets rarely move in straight lines, so there’s a good chance that over the next few days, we’ll see some strength in the S&P 500; however, it appears that the technical tide has shifted, and at minimum, we will likely see the market trade down 3-5% over the next few weeks. The reason why this matters for VIX traders is that a movement down of this caliber will likely result in the VIX rallying by around 10%.
Wait, didn’t we just say that mean reversion suggests that we’ll see the VIX correct and yet we have both seasonality and technical analysis suggesting that we’ll see it rise? How do we rectify these two opposing views?
As I see it, our first study which looks at market probabilities as it relates to the outright VIX is a general compass of sorts, whereas the other studies are contextual. What this means to me is that yes, we should be mindful of mean reversion from these levels (with historical data suggesting a 10% drop over the next month as the average expectation from here). However, in context of today’s market action as well as the time of the year we are in, the data suggests that we’re going to continue to see the VIX push higher which will sweeten the odds for the volatility bears.
Also, it is important to note the timing: it’s entirely possible for the VIX to correct over the next month (as odds strongly suggest it will) and then continue higher through October and November in line with market tendencies. We don’t know the exact course of things, but this kind of scenario is a very real possibility and more in line with the nature of markets which is to fluctuate around general statistical tendencies.
What this means for VIX traders is that we should be looking for shorting opportunities on each further upwards movement in the VIX. We know mean reversion is strongly suggestive of a reversal of the VIX pop and we also know that there are seasonal factors which suggest the possibility of more volatility to the upside. This allows us the opportunity to sell into rallies if we have a long enough trading and holding time frame. This is how I’m trading UVXY and what I’ll cover in the next section.
UVXY is an interesting ETF in that it allows investors the opportunity to make leveraged bets on an index which is generally declining through time. UVXY follows the S&P 500 Short-Term VIX Futures Index (see more about it here) and it does so on a 1.5x leveraged basis. If you follow the link in the previous sentence and play around with the data, you’ll likely notice something shocking: this index declines at a rate of about 50% per year consistently. In fact, there’s basically never been a time period in available history in which one could buy and hold the index for more than 1-2 years and make money. In other words, it is a consistent loser over long periods of time and the only profits made by long traders of the ETF are likely made over very short time periods.
The reason for this loss is roll yield: VIX futures are in contango (priced above the spot level of the VIX) most of the time (~85% of all days). And through time, this difference narrows to be zero for the front contract.
Here’s a chart that shows the average level of the VIX as well as a few different futures contracts, using the last 10 years of data. This chart shows a very clear relationship: futures tend to be priced above the spot level of the VIX and this difference narrows during a normal month.
The huge problem for UVXY traders is that they are holding and rolling the first and second month VIX futures contract. UVXY starts a month 100% in the front month contract and ends the month 100% in the second month contract and it keeps this process up in perpetuity.
The problem for UVXY traders is that since you are holding and rolling this exposure at the front of the curve, you are greatly exposed to the downwards slope seen in each of the front two contracts in the chart above. To get an idea as per the magnitude here, recall the unleveraged return of the index: it drops at an annualized pace of about 50% per year. In other words, simple back of the envelope math would say that a base expectation for a UVXY trader is a loss in the territory of 70-75% per year.
Not only is this the case, but pops in the methodology tend to lead to stronger drops in performance.
For example, as you can see in the above chart, there’s a clear relationship in that the stronger the short-term return in the index, the greater the losses that can be expected in the longer term. Put simply, this means that pops (like we are seeing now) should be faded.
How I’m trading this data is this: in UVXY, I have long-dated puts that are fairly far out of the money to capture this tendency. Through time, UVXY has a baseline expectation of a 75% drop per year due to futures convergence. There absolutely are large variances around this mean expectation; however, the numbers pretty clearly suggest that shorting pops in the VIX as well as pops in this index make for solid long-term plays.
I cannot suggest that traders outright short this index (because volatility can move up hundreds of percentage points in very short order), so I suggest that if you find this piece convincing, you only trade it through puts or put spreads.
Historical data suggests that we’ll see a short-term drop in the VIX while medium-term data suggests we may continue to see upside in the index. UVXY is following a methodology which drops consistently through time which means that pops make for good shorting opportunities. Long-dated puts remain a strong option for trading UVXY to the downside.
Disclosure: I am/we are short UVXY. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.