EZV Algorithms: Consider Leveraged ETFs And Reduced Drawdown Risk

Nov. 29, 2019 6:15 PM ETSPY, TQQQ, SPXL, QQQ31 Comments


  • In my Marketplace service I’ve had numerous discussions about using leveraged ETFs while taking advantage of the drawdown protection offered by EZV Algorithms algorithm.
  • I’ve analyzed TQQQ and SPXL returns and drawdown performance since the inception of TQQQ in February 2010, using EZV Algorithms algorithm versus holding throughout the study period.
  • I’ve also analyzed returns, risk and Sharpe ratios for a more balanced, diversified portfolio with 3X leveraged ETFs representing 40% of the mix.
  • Until now, I’ve viewed my work as most applicable to those preparing for, or in retirement. But I’m finding noteworthy interest from aggressive investors.
  • Each investor's risk profile is unique, and I offer no suggestions in that regard, but I’ll report on how a very aggressive posture might be enhanced by the algorithm.
  • I do much more than just articles at The Easy VIX: Members get access to model portfolios, regular updates, a chat room, and more. Get started today »

When I started EZV Algorithms Marketplace service I expected my target audience to be investors preparing for retirement or those already there. The drawdown protection was the primary attraction. I viewed the near doubling of returns as being a byproduct of the mitigated loss profile. But lately I'm seeing increased interest in using the drawdown protection to take on more leveraged positions. So today I want to address a more aggressive use of the algorithm because the interest from Marketplace members might be indicative of a broader interest across the Seeking Alpha community.

If you're a long-term reader of my articles, you know that EZV Algorithms algorithm monitors changes in the shape of the VIX futures curve to assess the risk of holding equities. Data since 2008 shows that it successfully identifies significant market inflection points from risk-on to risk-off in real time, and then, by swapping stock positions for IEF, it materially reduces drawdown risk and improves returns. I won't go into much detail about the methodology in this article, so if you want to learn more follow this link.

For skeptics I will highlight one unique feature and advantage of EZV Algorithms. Most analysts who build algorithms do so by curve fitting the period studied. Effectively they know what would have worked because they have the history and they fit their parameters with perfect hindsight. In contrast, EZV Algorithms utilizes an artificial intelligence ("AI") routine that propagates decision parameters based only on the history available prior to each decision. So, since the AI routine has worked in the past without the advantage of knowing the answer, there is good reason to believe it will work tomorrow, next month and next year.

Leverage Anyone?

I normally apply EZV Algorithms methodology to a basket of ETFs consisting of SPY, DIA, QQQ, IWM, and a small component of SSO (2X leveraged S&P). While the algorithm is built on VIX futures which are, in turn, reflective of S&P implied volatilities, I use a multi-component basket to improve returns. Particularly, the addition of QQQ and the leveraged SSO add about 4 percentage points to average annual returns compared to what could be accomplished with SPY alone.

Today I'll compare SPY with TQQQ (3X Nasdaq leverage) and SPXL (3X S&P leverage). While the 5-part ETF basket performs better than SPY alone, the returns tend to vary with rebalancing frequency for the 5 components, so I decided to keep it simple with single-ticker comparisons.

I'll build the discussion around this summary table of results.

Summary Comparisons

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

Parenthetically, to avoid confusion among long-term followers I'll point out that the 7.8% worst loss for SPY is worse than the approximately 1% loss for the 5-part ETF basket. Diversity is truly helpful and the superior returns from the 5-part ETF basket enable better recovery from transient drawdowns over the course of a full year. (Throughout this article when I say "worst-case annual" I mean the worst rolling 252 trading days.)

Building on the diversity-helps theme, for members of EZV Algorithms Marketplace, I provided results for different portfolio compositions. Tests included a portfolio consisting of SPY@30%, SPXL@20%, QQQ@30%, and TQQQ@20%. Here are the annual returns comparing a buy-and-hold strategy to trading that composition on EZV Algorithms algorithm.

Returns for 4-Part Leveraged Portfolio, Buy and Hold versus EZV Algorithms

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

The algorithm does what it was designed to do; it avoids most serious losses. Internal rate of return for the approximate 10-year period was 30.0% for EZV Algorithms versus 18.5% for holding throughout the period. The table above shows calendar years, but continuous rolling time frames provide a more comprehensive test. The worst loss for any rolling 252 days was (5.6%) for EZV Algorithms trades versus (22.5%) if held throughout.

For those who like using Sharpe ratios, the buy-and-hold strategy's Sharpe ratio was 100.4% versus an outlandish 209% for EZV Algorithms series. EZV Algorithms returns in excess of an assumed 2.0% risk-free rate were much higher, and the standard deviation of those excess returns was lower than holding throughout the period. Here are the components of that calculation:

Sharpe Ratio Comparison

Members note: These results are updated to November 25th

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

If I can offer one tangential observation, I originally did these calculations for my Marketplace members with data ending November 8th and now, using November 25th data, the returns and Sharpe ratios have improved noticeably even though they represent 10-plus years and the update only covers about two weeks. Such is the strength of the current rally.

Speaking of the current rally, my last buy signal went out on August 29th and the standard ETF basket is up 8.42% since then. Yet the leveraged portfolio described above was up 14.6%. So, if you're an aggressive investor who likes to use leverage in your investing strategies, consider adding EZV Algorithms drawdown protection and accelerating returns further.

One more thing... my obligatory warning about black swans. They have no respect for statistics whatsoever. The analysis I've done here begins in 2010 because that was the inception of TQQQ. There have been no severe downturns over the nearly 10-year period, so even statistically, I would not expect these results to be fully reproducible when more dramatic downturns are averaged into a longer experience. So, be careful out there.

Thanks for reading and more so for following, and a special thanks to the members who support EZV Algorithms service.  If you find these ideas interesting, please click the orange follow button at the top.  Even better, if you've read a few of my articles and would like to get real-time signals and more customized analyses, pull the trigger and do a free trial of EZV Algorithms.  I think you'l find the profit potential is far greater than the cost of the service.

This article was written by

Michael Gettings profile picture
Author of EZV Algorithms
Drawdown protection and outsized returns by decoding the VIX futures curve

Mr. Gettings is CEO of RiskCentrix, a firm that specializes in the establishment of disciplined programs for commodity risk mitigation, the integration of enterprise risk programs with financial management, and support in the area of risk-cognizant strategy formulation. He has 35+ years experience, originally in regulatory affairs with a major utility, then as founder and president of a natural gas marketing and trading firm. As a consultant, for the last 20 years he has developed and implemented innovative approaches to risk assessment and mitigation for utilities, power generators, and energy-intensive industrial firms. More recently he is retired, but for occasional consulting. He manages his own portfolio using many of the quantitative methods deployed throughout his career.

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 a basket of ETFs and any 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 were 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.

Recommended For You

Comments (31)

To ensure this doesn’t happen in the future, please enable Javascript and cookies in your browser.
Is this happening to you frequently? Please report it on our feedback forum.
If you have an ad-blocker enabled you may be blocked from proceeding. Please disable your ad-blocker and refresh.