Use A Broken Wing Butterfly To Capture VXX’s Down Drift

Posted By on Nov 12, 2012


In September I outlined a strategy for hedging against market volatility for little cost; even being paid to do so. This involved buying a longish term VXX call backspread and taking off if VXX hadn’t spiked within 3-4 weeks of expiration.

(NB the VXX underwent a 1:4 split in October and so the prices quoted in the article should be multiplied by 4 for a direct comparison with today).

Anyone employing this strategy before the recent volatility should have done well. And indeed the following months, with the fiscal cliff looming, may provide other good, short term, examples to exploit this volatility.

However there is a frustrating issue with this strategy: most of the time VXX doesn’t peak and the backspread just sits there losing 5%-10% of its value until it is taken off 3-4 weeks before expiration.

Indeed, for most of the two months since my original article, followers of the strategy would see such an outcome. As that article stated in the absence of volatility it drifts down, due to the ‘contango’ effect of the VIX term structure (see the article for details), only spiking up quickly if the VIX moves sharply up.

I was therefore asked by several of our subscribers whether instead we should try to profit from this drift, rather than wait for a volatility spike which may never come. The strategy would also need to avoid significant loss should further such spikes occur.

Now would be a good time to consider such a trade. If the recent post election volatility is short lived, the VXX should come down from its recent spike, and the effect of the drift would be magnified.

In addition the VIX term structure is pretty flat at present:

Trade Date         
Expiration Date     
VIX
   Contract Month
11/9/2012
CASH
18.61
11/9/2012
17-Nov-12
19.03
1
11/9/2012
22-Dec-12
18.58
2
11/9/2012
19-Jan-13
19.30
3
11/9/2012
16-Feb-13
20.18
4
11/9/2012
16-Mar-13
20.68
5
11/9/2012
22-Jun-13
22.19
6
11/9/2012
21-Sep-13
22.96
7
11/9/2012
21-Dec-13
24.05
8
11/9/2012
18-Jan-14
24.00
9
11/9/2012
21-Jun-14
24.78
10

Source: CBOE

The December futures are almost exactly the same value as the cash rate and lower than the November contract. This backwardation is unusual and, historically, is short lived; we should expect the structure to revert to its usual state of contango (when the December futures are higher than November’s) after which VXX should fall due to the usual negative roll yield described in the original article.

However should further volatility stay high or even rise, due to the impending fiscal cliff say, the VXX may rise further, and we need to be hedged against this.

The trade to consider is a VXX put broken wing butterfly. For those not familiar with this strategy it is similar its more common cousin, the butterfly, in that it comprises two at-the-money short puts with one in-the-money and one out of the money put. The difference is that there is an unequal gap between the three sets of options.

Let’s illustrate this with an example with VXX at $37.18:

  • Buy 25 VXX Dec12 29 Puts
  • Sell 50 VXX Dec12 34 Puts
  • Buy 25 VXX Dec12 36 Puts
  • Total Credit: $1,375
  • Net Margin: $11,125

(NB This is the amount we will trade in our portfolio. You can, of course, trade a smaller amount)

Here’s the profit and loss at expiration:

 

 

 

 

 

 

 

As you can see we get to keep the $1,375 credit if VXX stays above 36 and so, in particular we still make a decent return (12% on risked capital in just over a month) should the current volatility continue, or increase.

However the real money is made if VXX drifts down 5-15%; up to $6,375 if VXX is at 34 on 22 December.

We start to lose money below 31.5. But remember the trade thesis: VXX spikes up or drifts down. It doesn’t tend to crash. For that to happen VIX would in turn have to fall significantly in the next 40 days, something that would only happen in a significant market rally (which would benefit other parts of our portfolio anyway).

We will therefore have a rule that should VXX touch 31.5 at any stage we will remove the trade. Depending on when this happens this would result in at most a $2,000 loss on the trade. This is within our usual ‘never lose more than 2% of our portfolio ($2,000 in this case) in any one trade’ stop loss and represents an unlikely scenario.

Anyway, we now have a choice on how to play VXX: wait for the big one off gain at minimal risk with the call backspread, or exploit the security’s underlying downward drift with the broken wing put butterfly.

So which one to choose? Well, that depends on your opinion of volatility over the next few weeks. Personally I think the current worries over the fiscal cliff are overdone, and any election uncertainty is bound to dissipate. Hence my choice of the BWB.

Whatever you decide, we have constructed trades which offer significant profit, with managed risk and known exit points. And these are attributes of good options trading.

 

Tags: , , , ,

2 Comments

  1. Hi Chris,

    The VXX is already down to 34.79 this morning. Is this within normal fluctuations for this trade since the goal is be be down to 34 by dec 22?

    Mitch

    Post a Reply
    • Mitch,

      No it isn’t: it was a large, unusual, move. VXX doesn’t tend to fall 6% like that.

      It’s good in that it’s moved directly under the maximum profit curve. But annoying in that it does bring the downside scenario, hitting 31.5 before 22 Dec, more into play and producing a small paper loss.

      (It also means that I’ll have to change the strike prices of the trade for my planned reproduction of the article on Seeking Alpha. But we won’t be changing the trade for the performance checking).

      Anyway, we’ll stick with it, only doing anything (i.e. taking it off) if it hits 31.5

      Chris

      Post a Reply

Submit a Comment

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>