I’ve spent a good chunk of hours looking into volatility and I stumbled upon a simple volatility strategy for VXX. The idea is to go long the VXX ETF when volatility is rising and go short the ETF when volatility is falling. This is opposite to the usual approach which is to short volatility on sharp rallies.
I’ve found that short-term trend following can be the best option for trading VXX which is an ETF suffering from inherent structural weakness. Because of the way the contract is constructed it loses value over time. A concept known as time decay.
Following the trend in VXX seems to work well from what I can see. However, it’s risky because VXX is a leveraged ETF with volatile price moves. Also, since this analysis is based on one security backtests, there is the possibility of unreliable results. Curve fitting happens more easily with single stock testing and the future is never going to be exactly like the past.
Volatility Strategy Rules
For this simple strategy we will be using both a long side and a short side.
The long side is much harder to get right in VXX since the ETF has been in a steady downward trend since 2009.
However, it’s important to have a long side in order to balance out our risks. Plus, going long volatility has great hedging benefits.
The buy and sell rules for this strategy are as follows:
- If Close > Top Bollinger Band (20,2) go long VXX on the next open
- Sell on the next open if Close > Open OR after four trading days
- If Today’s Close > Yesterday’s Close AND Today’s Close < 20-day MA go short VXX on the next open
- Cover on the next open if Close > 20-day MA OR after four trading days OR by fixed stop loss at 100%*
Trade Example – Long
In the following chart you can see an example of the long trade setup we are looking for:
As you can see, VXX closes above the top Bollinger Band on the 9th September 2016 so we go long on the next market open at a price of 157.16 (green arrow).
The market closes down that day but reverses higher the next day giving us our first up close. We therefore sell on the next bar open on the 14th September at a price of 162.2 (red arrow) giving us a profit of 3.21% before costs.
Trade Example – Short
In the next chart you can see an example of the short trade we are looking for:
We therefore go short on the next market open at a price of 97.92 (red arrow). The market then dips sharply and moves into profit straight away.
We close our trade four days later on the open of the fifth bar at a price of 88.80 (green arrow). This gives us a profit of 9.31% profit before fees.
Note that the reason we get a signal on the 30th instead of the previous day close is because on the 29th we were already holding a short position which is then closed out on the 30th.
Initial Testing On VXX
On its own, the long side produced a net profit of 98% (9% CAR) on VXX between January 2009 to January 2017 with a max drawdown of 32%. Meanwhile the short side produced a net profit of 475% (25% annualised) with a maximum drawdown of 31%.
Now we have both the long and short rules we can put the two sides together and see what would have happened with the combined strategy.
Volatility Strategy Backtest Results 2009 – 2017
Following you can see the backtest results for our simple volatility strategy between February 2009 and January 2017. This is based on a starting capital of $50,000 and transaction costs of 0.1% per trade.
- Net Profit: 1034.6%
- CAR: 35.89%
- MDD: -33.85%
- CAR/MDD: 1.06
- # Trades: 279
- Win Rate: 58.78%
- Sharpe: 1.18
As you can see, we have recorded some excellent results from this strategy on VXX with an annualised return of over 35% across 279 trades. We have scored a win rate of 59% and made money every single year.
But at this point you may be wondering why I have chosen to only backtest between 2009 and 2017.
The main reason is because I wanted to leave some out-of-sample data with which to test the system on afterwards.
We know that there were big volatility events in 2008 and in February 2018 so it will be interesting to see how the system performs through these periods.
Historical Data For VXX
However, we have a big problem here because VXX was only introduced in 2009. How can we possibly test our system on the 2008 crash when VXX wasn’t around?
Fortunately, there is a solution since it’s possible to simulate how VXX would have traded based on historical futures data and details that are specified in the ETF prospectus.
Even more fortunate is that this has already been done for us and is available to purchase at Sixfigureinvesting.com. In order to run the analysis I purchased this data and loaded it into my backtesting program Amibroker.
So now we have simulated data for VXX from April 2004 we can combine it with more recent VXX data and run a backtest on a much wider date range. This data is adjusted for reverse splits and tracks the IV data within +-0.04%.
Having this data available opens up many more possibilities for building volatility strategies since we can now look back at a much longer data period.
Volatility Strategy Results 2004 – 2018
Bearing in mind that I have no idea how this trading strategy is going to perform during the 2008 period I opened up Amibroker and ran our strategy between 4/2004 – 9/2018 and produced the following results and equity curve:
- Net Profit: 2658.31%
- CAR: 25.86%
- MDD: -42.5%
- CAR/MDD: 0.61
- # Trades: 508
- Win Rate: 55.51%
- Sharpe: 0.92
As you can see, the results from this strategy are pretty good. We have managed an annualised return of over 25% and we have greatly increased our starting capital. Even better, we have made money in 2008 and in February 2018.
We do have a higher drawdown and there is no doubt this is a risky strategy, however, that risk could also be reduced with a smaller position size and with hedging.
Risks To Be Aware Of
I was surprised by the results of this strategy because they show that this system actually made money during the ramp up in volatility in 2008 and it even made a killing in February 2018. Somehow I assumed that any short VXX strategy would have lost money during these periods but I was wrong. This means the strategy could have a lot of value as a hedge to long-only portfolio.
However, there are still plenty of risks to be aware of.
The first is that shorting VXX can be costly and difficult to do. The ETF is typically in the hard to borrow category which means shares are not always available to short with some brokers.
Additionally, spreads on VXX can be wide and the margin interest can be expensive. This cost only increases when the VIX is high. This leads to the unfortunate reality that the trade becomes its most difficult when it’s most likely to succeed. The same can be said for option premiums.
Finally, going short is always a risky trade since losses can be larger than your initial investment. In the case of VXX, the ETF more than doubled in February 2018 which means you would have lost more than 100% of your initial investment if you were short throughout the move. Our backtest may have simply been lucky not to short VXX at the wrong time.
By using a fixed stop loss you can protect some of this risk but it won’t be fail-safe in an illiquid, volatile market. Furthermore, a loss can jump over the stop if there is a overnight gap.
With all that said, I do think it’s possible to make the trade work so long as you watch the market closely and don’t risk too much on your initial trade. You need to watch the market all the time and you need to have balls of steel. If you do that, the biggest risk will likely come from a large overnight jump in the VIX which does not lead to a halted market.
The simple volatility strategy that we showed today appears to work well by riding short-term trends in VXX. I’m not sure if I would trade it as I’m not very comfortable with shorting VXX. But this strategy could be worth exploring further and might be a useful hedge at a low position size.
Notes: Charts and simulations produced in Amibroker with data from Norgate and Sixfigureinvesting. Data is adjusted for reverse splits for simulation purposes. Simulations use transaction costs of 0.1% per trade.
Thank You For Reading
Joe Marwood is an independent trader and the founder of Decoding Markets. He worked as a professional futures trader and has a passion for investing and building mechanical trading strategies. If you are interested in more quantitative trading strategies, investing ideas and tutorials make sure to check out our program Marwood Research.
This post expresses the opinions of the writer and is for information or entertainment purposes only. It is not a recommendation or personalised investment advice. Joe Marwood is not a registered financial advisor or certified analyst. The reader agrees to assume all risk resulting from the application of any of the information provided. Past performance, historical or simulated results are not a reliable indicator of future returns and may not account for real world settings. Financial trading is full of risk and margin trading can lead to financial losses totalling more than what is in your investment account. We take care to present accurate analysis but mistakes in backtesting and presenting of analysis regularly occur. Please read the Full disclaimer.
Thank you to everyone who takes the time to leave a comment. Your feedback, constructive criticism and identification of mistakes is welcome. In order to concentrate on work I may not have time to respond to all comments.