The Easy VIX: Who Wants To Ride The Next 25% Decline? Keep Your Money Instead

|
Includes: DIA, IEF, IWM, QQQ, SPY, SSO, VIXY
by: Michael Gettings
Summary

Given hyperbolic media coverage of trade skirmishes, a reticent Fed in a zero-interest world, transient inverted yields, and tweets, stocks are far more volatile than fundamentals would dictate.

What next? A buy-and-hold strategy will ride the next 10%, 25%, or 50% decline. Will you? If 500+ point swings in the Dow leave you unsettled, consider a new approach.

Big drawdowns can be avoided. Identify 11% of the time when drawdown risk is excessive and double annual returns. It’s a risk-pattern issue, not a price-pattern issue.

On any given day since 2008, next-quarter losses have exceeded 10% about 6% of the time and 5% about 14% of the time. The Easy VIX algorithm substantially avoids them.

And for false signals, when stocks don’t plummet, the algorithm produces a break-even distribution of small gains and losses.

Some will tell you that timing market turns is a fool's errand; others say they can do so by tracking fundamentals or price charts. I see no reason to be constrained by those two choices. Let me explain.

Typical market-timing advocates follow fundamentals or price-driven technical indicators. We've all seen fundamental analysts discuss why the economy is slowing and the market will turn down, only to repeat the same warning for two years while the market runs higher and higher. Then, when an inevitable correction occurs, they claim to have gotten it right again. Their fundamental data is great, but the market marches to a different beat, and timing with fundamentals can be wrong by years.

For the most part, when I read technical analysts predicting market turns, it seems to go something like this: "'A' might happen, or if 'B' happens then 'C' might happen, otherwise 'D' might happen." Then a week or a month later, there is always some wisdom to showcase in retrospect.

Let me be clear. Both fundamental and technical analyses have merit, but when it comes to avoiding big drawdowns and improving returns via timing choices, it's simply a game of probabilities. Fundamentals do drive stock prices, but the timing is extremely uncertain. Chartists get it right a reasonable percentage of the time, but anyone relying on price charts alone should probably incorporate some form of hedging or stop loss in their strategy.

A Better Way

There is another choice. Markets cycle based on risk-on/risk-off psychology. Find a way to identify the emergence of material risk-off psychology and you can avoid big drawdowns. If you can identify the emergence of sustained risk-on psychology, you can stay invested with confidence. It's still a game of probabilities, but it's a risk issue, not a price issue.

Stock prices alone do not identify emerging risk conditions until it's too late. You could measure observed volatility, but to be statistically valid, you'd need at least 30 days of trailing price action, and then, all you'd know is how much risk existed for that trailing period.

Fortunately, the implied volatility of options provides an immediate reading on the market's perception of volatility and risk - now. And while in theory, risk carries upward and downward potential, implied volatilities are far more sensitive to downside potential; that's the risk we care about here.

There is more good fortune. The VIX is designed to provide a real-time reading on the implied volatility of a broad market index, the S&P. And still more good fortune... The VIX trades on the futures market, so not only can we see the spot value of S&P's implied volatility, we can see the market's money-backed perception of how it's changing for the future.

So, what we need is a robust calibration of how changes in the VIX futures curve reflect emerging risk-on/risk-off psychology. I use the word 'robust' with purpose. Investopedia provides this definition: "Robust is a characteristic describing a model's, test's or system's ability to effectively perform while its variables or assumptions are altered." I'll have more on that later, but first, let me explain the Easy VIX model.

The Easy VIX Model

This picture might help make the concepts I'm about to discuss more tangible. If you've followed markets in the last few weeks, you'll recognize that it felt very risky two weeks ago but has settled into a calmer upward bias in recent days. Here is how the VIX futures curve changed from a risky August 27th to yesterday, September 5th. It shows a very identifiable pattern progressing from risk-off on the 27th (downward slope or backwardated) to risk-on by September 5th (upward slope or contango).

VIX Futures Curves Progressing from Risk-Off to Risk-On Psychology

Source: Michael Gettings Data Sources: VIXCentral.com, CBOE

For those who might not have had much experience with futures, the fancy word for the August 27th shape is 'backwardated', near months are trading at a premium to later months. The healthy shape of September 5th is called 'contango' when near months are discounted. The shape is a strong indication of risk-on/risk-off psychology which generally correlate with contango/backwardated conditions, respectively. More importantly, the rate of change in shape is a strong indication of the emergence of changes in risk-on/risk-off psychology, and that has impressive predictive value if one cares to avoid drawdown risk. Who doesn't?

So, the Easy VIX methodology harnesses the predictive value of changes in the VIX futures curve to forewarn of drawdown risk. When it indicates "Sell," I sell a broad-based basket of ETFs (SPY, DIA, QQQ, IWM, SSO) and buy IEF, a ten-year treasury ETF. When it says "Buy", I reverse the trade and ride the risk-on period. The system spends about 11% of the time since 2008 in treasuries, and 89% of the time in the stock ETF basket. By doing so, it substantially avoids most of the more significant drawdowns over the eleven-year period.

I won't go into detail here, but if you would like a more comprehensive understanding of the underlying model, read my last article "The Easy VIX: A Frenetic August, But No Traction! - A Small Rant and Refinements to The Risk Model" in Seeking Alpha. For now, I'll simply say that the model measures the contango or backwardated shape of the VIX futures curve, and the rate of change over two different look-back periods.

I'll also address my prior use of the word 'robust'. The model is calibrated using an artificial-intelligence algorithm to determine look-back periods for the slopes and the triggers that determine exit and reentry signals. Those calibrations are derived from periodic historical data and only applied prospectively. Since these key parameters are not being optimized in a retrospective fashion, the algorithm should perform as designed across a wide range of future conditions.

Tangible Examples

My goal today is to make the model's performance tangible to readers. To that end, I'll show graphic results of the last two corrections, plus the 2008 debacle. You will see that it's not perfect, just excellent. The benefits come from avoiding the big drawdowns, while the smaller variations between SPY and the algorithm amount to zero-sum noise. I've focused on corrections greater than 10% here, but the statistics indicate that all gains and losses which vary from a buy and hold strategy by less than 5% add up to a zero sum. Any divergence from buy-and-hold greater than 5% is advantageous. Said differently, a 5% disadvantage is as bad as it gets, and comparable gains offset those, but the advantageous outcomes in excess of 5% accumulate to approximately double benchmark returns.

If you inspect the following graphs, you'll see how the model avoids much of each period's drawdown and, to provide a full picture, how false signals do not amount to much. If you really scrutinize them, you'll note that the sell intervals are not exactly flat because the stock ETFs are replaced by IEF during those times, and treasuries tend to rally when stocks fall. Finally, for simplicity, the graphs use SPY as a benchmark, but in practice, I trade the 5-ETF basket of SPY, DIA, QQQ, IWM, and SSO with a modest overweighting of QQQ and more significant underweighting IWM and SSO.

Here is the most recent correction from December 2018:

Performance During December 2018 Correction

Source: Michael Gettings Data Sources: Fidelity, VIXCentral.com, CBOE

And the Correction of January 2016

Source: Michael Gettings Data Sources: Fidelity, VIXCentral.com, CBOE

Finally, for some real fun, here is the June to December 2008 interval. Once again, the model avoids much of the drawdown, and this time, in a severe downturn, false signals are imperceptible.

Performance, June 2008 to December 2008

Source: Michael Gettings Data Sources: Fidelity, VIXCentral.com, CBOE

Current Market Metrics

Finally, you might wonder what the algorithm says now. The bottom line is I'm becoming hopeful that with summer doldrums over, we might be setting up for a meaningful rally. Current indications are bullish and improving. For established followers, here is the Easy VIX dashboard for September 5th. For new readers, I won't explain it all here, but read the article linked earlier, particularly 'The Quant Corner' subheading to understand it well.

The Easy VIX Dashboard, September 5th, 2019

Source: Michael Gettings Data Sources: Fidelity, VIXCentral.com, CBOE

Closing

So, if you're retired or near retirement and can't afford to lose your nest egg, or if you simply want to increase returns by avoiding downturns, follow the VIX futures quotes for early warning signs.

Or better yet, follow me and get the benefit of quantitative calibrations. I post two to three articles a month and provide more frequent near-real-time updates via two to three blog postings weekly. Or just ride that next 20% downturn.

Thanks for reading and especially for following. If you're not yet a follower, consider clicking the orange "Follow" button at the top to receive more articles and market signals.

Disclosure: I am/we are long SPY. 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.

Additional disclosure: I trade all the tickers mentioned using the algorithm described. The artificial intelligence algorithm monitors daily performance and periodically recalibrates look-back horizons and triggers in a step-wise sequence. New calibrations are applied prospectively only, and never applied to the historical period from which they derived. The algorithm described and the discussions herein are intended to provide a perspective on the probability of outcomes based on historical performance. Neither modeled performance nor past performance are any guarantee of future results.