Sunday, 27 April 2014

Gold - A better way to go long? (Part 2 - Upside Risk)

In Part 1 we looked at some basic stats for GLD options and developed an upside bias. However, outlook does not always translate into reality! One of the advantages of structuring positions using options vs. solely equities or futures is the ability to profit even if the underlying does not move or move in the opposite direction versus your original bias. Before we start thinking about designing a position around our bias, let's think about risk!

If we buy 100 shares of GLD, what is the risk? Theoretically, if spot gold fell to $0/oz, GLD shares will be worthless. So the theoretical downside risk is the full amount you paid for the shares. (the downside risk is actually greater than selling an naked at the money put!) Of course spot gold falling to 0 is an extremely unlikely scenario. One may argue, in exchange for this downside risk, you are also holding a position with theoretically unlimited upside. Unfortunately that is equally unlikely. 

Of course, gold can rise to 10,000$/oz, but that would take some extraordinary loss of confidence in the current financial and monetary system. Never say never, but I don't see this scenario playing out in the near future (2-3 months). So the idea is to design positions around realistic trading bands of gold, while structuring risk exposure on both sides so that:

a) We can receive credit upfront for the entire structure, ensuring if GLD doesn't move or moves slightly the wrong direction, we still have an opportunity to profit. 

b) so we can manage the position in separate legs, giving us flexibility to extract profit from certain parts of the position as GLD moves around

The first step is to determine the upper trading band of GLD in the timeframe we want to structure the position. Since we are trading off this potential event in Ukraine, let's see what happened to GLD when sanctions were first announced a few weeks ago. 

According to Reuters:* March 6: Crimea's leadership votes to join Russia. U.S. President Barack Obama orders sanctions on those responsible for Moscow's actions in Ukraine.

Looks like it went from a low of 128.21 on Mar 7th to a high of 133.69 on March 14. IV30 rose from 14.35 to 16.65 during the same period. This means in a period of 7 calendar days, GLD moved 133.69 - 128.21 or $5.48. Based on the Mar 7 IV30 of 14.35, GLD moved a little more than 2 standard deviations (using 7 days but the IV30 volatility)! Okay, that's quite a move in a week. But GLD did retrace and a month later on April 7th, it has fallen to 124.91 at the close, well within the 1 standard deviation band priced in the options. Let's keep that in mind, we should be aware GLD can go on sharp rallies in a very short period of time. This will become important as we actively manage and adjust the position (if needed) as GLD moves around.

Let's look at the min to max within the last 90 days. GLD traded at a low of 119.46 on Jan 30 to a high of 133.69 on Mar 17. So in 45 calendar days it moved up $14.23. IV30 was priced at 16.17% on Jan 30th. This is a upside move over 2 standard deviations. Okay, we know GLD can move 2 standard deviations within 45 days with how the IV30 is priced. This all happened quite recently, within the past 90 days. 

Going off on a tangent here... I like to base my trading decision on analysis of fairly recent data. This is an arbitrary decision but in my opinion, the market today is different than the market in the past (eg. data from 2010 would be more representative of market conditions today than say.. data from the 1920s). So you won't see me going back 8 years and pulling out all the GLD data and analyze my way through every extreme move in order to establish a trading band. If I decide to set up a longer term position with LEAP options, then maybe i'll go back a few years, but for positions structured 40-60 days in the future, I tend to look back for the past 90-180 days. Once again, this is a personal preference. 

Back to the trade... let's look at the June monthly options (by Monday open 4/28 they'll have 54 days to expiration). Since the structure I have in mind will have short option positions (as well as long), I would like to keep the position gamma low. This will lead to slower (and lower starting) delta on the position as a  whole (of course if we are right about GLD direction it'll also lead to less profit, but it's always better to be cautious than believe you are right about the direction of the underlying price). In additional I would like to have positive theta, in case we are wrong and gold stays flat for the next little while. The whole thesis is to structure a position where we can profit

a) from rising GLD prices
b) GLD staying flat
c) potentially profit from rising IV or falling IV on certain legs of the structure


c) less risk relative to straight out holding GLD shares or Gold futures.

Before we think about positions, let's check the option skew in the June series. 

The June skew is parabolic, with further OTM calls and puts priced at higher IV. It means we can buy closer to the money options for relatively cheaper IV and sell OTM calls for higher IV. Remember, we always want to trade with the skew whenever possible while expressing our directional bias for the underlying.

Let's summarize the analysis up until this point: 
a) GLD moved up 2 STDev when US sanctions were announced the first time
b) Min to Max move in the past 3 months was over 2 STDev over a 45 day peroid.
c) we want to buy lower IV and sell higher IV calls 
d) we think IV isn't significantly under-priced or over-priced from Part 1 based on HV30
e) we have an upside bias for GLD because of potential conflict in Ukraine 

2 Standard deviation range based on Friday closing price and IV30 is 111.24 to 139.62 in 54 Calendar Days

2.33 Standard deviation range is 142
2.6 Standard deviation range is 144 

Personally I would be comfortable with setting up the position with a break even at the 2.6 Standard deviation range. If prices were actually normally distributed the probability of it moving up more than 2.6 standard deviations is less than 1%. In reality, prices are not normally distributed and there's significant tail risk. However, this is a level of risk I can tolerate comfortably. If you are more risk averse you can certainly decide on a higher break-even point.

The position I have in mind given the analysis we have done is a forward call 1:-2 ratio. It will let us take advantage of skew and start off essentially delta neutral. Since the market is not open right now, we will need to revisit tomorrow. But if the price remained around where it is right now, this is a potential position.

we'll go into depth about this position and how we can cap off the upside risk and maximize margin usage in Part 3. We will also structure the downside position in part 4 and look at the overall GLD structure. 

No comments:

Post a Comment