Another Look At The NUGT-DUST Double Short

| About: Direxion Daily (DUST)


Shorting both sides of a leveraged ETF pair is a potential strategy to exploit leverage decay.

A previous article showed that rebalancing can mitigate, but not completely prevent, the main risk of this strategy.

An alternative strategy is proposed: shorting both sides of a leveraged ETF pair and holding for one year.


Daily-reset leveraged ETFs suffer from a phenomenon known as "leverage decay" or "beta slippage", in which they rapidly lose value when the underlying index is volatile but with no net change in either direction.

For a very simple real-life illustration of this concept, consider the one-year price action of the Market Vectors Gold Miners ETF (NYSEARCA:GDX), and its corresponding 3x-leveraged funds, the Direxion Daily Gold Miners Bull 3X ETF (NYSEARCA:NUGT) and the Direxion Daily Gold Miners Bear 3X ETF (NYSEARCA:DUST). Note that while GDX has been essential flat (+1.38%) over the past one year, NUGT and DUST have declined by -51.8% and -77.6%, respectively, thanks to leverage decay.

NUGT Total Return Price Chart

NUGT Total Return Price data by YCharts

The longer-term trends are even more dramatic, with NUGT dwindling to only 0.3% of its original value since inception, while DUST has performed relatively "better," with 21.9% of its original capital remaining.

NUGT Total Return Price Chart

NUGT Total Return Price data by YCharts

Given that NUGT and DUST are essentially hedged versions of each other, does this make a double-short strategy an essentially market-neutral, massively profitable, trade? Not so fast, as shown in my previous article, "Shorting Leveraged ETF Pairs: Easier Said Than Done," the main risk of this strategy is that sustained moves in one direction will crush the trade. This is because the winning security will become continually larger in value (leading to greater losses on the short side), while the losing security will become continually smaller in value (leading to reduced gains on the short side). One potential solution to ameliorate this problem is via regular rebalancing, but as shown in the article above, the final profit/loss is greatly sensitive to the threshold selected for rebalancing.

It has been nearly one year since my previous article, and during the past one year the NUGT-DUST double short strategy was massively profitable. In this article, I seek to use a different methodology to investigate the feasibility of the double short strategy. If one were to short both sides of 3x leveraged funds for exactly one year, what is the likelihood of profit and the range of profit possibilities that could result? Given that NUGT and DUST have been trading for about 5 years, this gives 1079 possible start dates that this strategy could have been implemented in the past.


Let's look at the statistics of this strategy applied to the NUGT-DUST pair over 1079 possible start dates (from Dec. 8th, 2010 to Mar. 24th, 2015). The following data shows the total return percentage of the strategy. Note that the total return values are exclusive of trading commissions and short interest fees, given that these can be highly variable. When trading CFDs instead of stocks, financing can be as low as 3% on both long and short sides.

NUGT-DUST double short, 1-year holding period

  • Number of trades: 1079
  • Mean profit: 27.3%
  • Median profit: 32.0%
  • Standard deviation: 29.0%
  • % Positive trades: 86.1%
  • Maximum profit: 67.4%
  • Maximum loss: -81.2%

At first glance, this strategy appears to have worked very well for the NUGT/DUST double-short pairs trade in the past. 86.1% of 1-year trades were positive, and the average gain was +27.3%, with a median of 32.0%. However, there was great variability between individual results, as indicated by the standard deviation of 32.0%.

The following chart shows a histogram of 1-year return percentages, grouped into 10% intervals. Click to enlarge

However, it should be noted that during the gold crash of 2013, there was a sustained period in which strategy would have sustained heavy losses. During an unfortunate 225-day period from (ending dates) Feb. 27th, 2013 to Jan. 16th, 2014, the average return was -11.9%, and only 41.3% of trades were profitable.

A plot of 1-year total return as a function of trade end-date is shown below. A clear period of underperformance (marked in red) during most of 2013 can be observed.

Overall, I would have to say that this double-short 1-year holding strategy has been quite successful over the past 5 years. This has no doubt been aided by the gyrations in the price of gold mining stocks that have transpired over this period. Moreover, there was no rebalancing required with this strategy, saving on trading commissions and bid-ask slippage.

However, how would this strategy fare if the underlying index were to move in a sustained direction over time? To investigate this, I applied the same strategy to the Direxion Daily Financial Bull 3X Shares ETF (NYSEARCA:FAS) and the Direxion Daily Financial Bear 3X Shares ETF (NYSEARCA:FAZ), which are the 3x-leveraged versions of the Financial Select Sector SPDR ETF (NYSEARCA:XLF). Their price history since the financial crisis market low (Mar. 6th, 2009) are shown below.

FAS Total Return Price Chart

FAS Total Return Price data by YCharts

The overall statistics are shown below.

FAS-FAZ double short, 1-year holding period

  • Number of trades: 1523
  • Mean profit: 4.6%
  • Median profit: 4.1%
  • Standard deviation: 24.7%
  • %Positive trades: 60.9%
  • Maximum profit: 40.4%
  • Maximum loss: -227.3%

Clearly, the double short strategy has been not as effective with FAS/FAZ, although both mean and median 1-year returns are positive. 60.9% of trades were successful. A histogram of the returns are shown below.

Click to enlarge

Finally, the returns of the strategy as a function of ending date is shown.

We can see from the chart above that the worst time to apply this strategy was just after the market low, with a horrendous maximum loss of -227.3%.


While the strategy outlined in this article (short both sides for 1 year) is comparatively simpler than the approach described in my previous article (automatic rebalancing when one side exceeds a threshold weighting), and also avoids trading commissions and bid-ask slippage, it is still vulnerable to same risk. Namely, that this strategy performs best when the underlying security is volatile, but performs poorly when the underlying security moves in one sustained direction.

Obviously, if we could identify in advance which securities would be volatile, and which securities would not, then this strategy could be a winner. Unfortunately, predicting volatility may well be as difficult as predicting security prices. If anyone has any ideas on which securities (for which both bull and bear 3x-leveraged ETFs are available) might be the best to try this strategy on, do share in the comments section below!

For an alternative viewpoint, Seeking Alpha author Jonathan Kinlay, also a hedge fund manager, has discussed the theoretical aspects of the double short strategy in a recent article.

Disclosure: I am/we are short NUGT, DUST.

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.