Hey there, I need some clarification on double calender spreads leading into earnings.
1) Is it best to open the calender spread at the height of IV, or as early as possible?
2) If the option I am selling is expiring this coming week, week of earnings, and the leg I am buying is expiring next month, what is the pros/cons of the long leg being closer to expiration or further out.
3) The position comes off as being long vega. Obviously after earnings there is a vega crush. The further my long leg is from my short leg, the greater the vega on the position. How is specifically vega being calculated here? Is it the difference between the two legs?
As of right now the short leg is displaying an IV of 47%, and the long leg an IV of 28%. Vega is at 757. The P/L chart looks nice and all with its high peaks but I know once the vega crush comes around, the chart will look completely different. I am just trying to figure out how different. This is a position that is risking $7,300; in case you want to tell how relevant that vega amount is to the total position size.
In my head, it is my understanding that as of right now there is a 19% point difference in IV between the two legs. For every point difference that is lost between the two legs, then I can expect a ~ $757 loss. Am I correct in thinking this? But if it is like this, well I know that the short leg will drop IV the most, and the long leg by a bit..so the difference in IV points between each other would be smaller....leading to a big loss from vega.
One other thing. I notice that people tend to open a third calender spread just in case the stock barely moves after earnings. Why not sell a call & put spread right near the money? Since IV the day before is so large..any vertical spread right near the money is heavily skewed towards movement..and you can literally get something like 1:10 risk/reward if the stock ends up not moving the next day.
1) Theoretically you would like to sell the front month option with relative high IV and buy the lower IV option in the back month. Therefore as "high as possible". The greater the front to back month differential the more lucrative at a time when the front month vol is in a high percentile.
2) Generally speaking, the "nominal value" (assuming you sell an OTM option) closer to expiration will be less expensive than the same strike at a later expiration. Relatively it might be more expensive due to IV differential. When closer to expiration an OTM option's theta will be greater (time decay), thus losing extrinsic value more swiftly.
On the bright side an OTM option with a longer duration could be relatively cheaper due to a lower IV, the option will be less sensitive to time decay, and one has the ability to structure another trade around that specific option.
Note that there are a quite a number of factors to consider such as the current price of the underlying, IV levels and your future outlook of the underlying. That being both directional and volatility wise.
3) "This is a fascinating subject and one that has no easy answer, as I
imagine you have begun to realize.
First, for the benefit of those who are not sure what vega weighting
means, suffice to say that it is not possible to accurately compare the
vega of one option to the vega of another option. So, if you are
trading calendar spreads and you sell the front month and buy the back
month, both the short and long option will have its own vega. If you
want to know what your "position vega" is you cannot simply add the two
vegas. They reason is because the two options have independent and
different implied volatilities; i.e., you're comparing apples to oranges
and not apples to apples when you add those two vegas together.
The short answer is that there is no reliable vega weighting method
available to the average retail trader. The models that do exist are
proprietary models developed by the larger financial institutions and
are the product of rather significant investment of resources in both
financial and human capital. In short, you won't find a vega weighting
model in any software that you could purchase for use as a retail trader
and unless you possess the mathematical talent and insight to create one
on your own it probably is not something you will be able to put
together over the weekend in an Excel spreadsheet.
In the past we have had Mark Sebastian from OptionPit on to talk to us
about vega weighting. Mark is a pretty sophisticated guy when it comes
to options volatility subjects and he offered a "quick and dirty" type
of calculation for weighting one vega to another. That formula offered
some benefits but was not the type of solution you could rely upon in
The problem is that implied volatility in one option contract is not
tied to the implied volatility of any other contract. They can rise and
fall independently of the other. If there is a spike in implied
volatility in the front month at-the-money contract it does not mean
that there will be a corresponding spike in the back-month at-the-money
contract. Sometimes there may be a corresponding rise in the two, but
not necessarily. They move independently of the other and the only
thing they really have in common is the underlying security.
Drilling deeper, the real problem is that implied volatility is not
real. It does not exist; at least not in the physical world. It's a
mathematician's method of quantifying human psychology. Now, that's a
dubious quantification. Option prices are not determined by
Black-Scholes or any other option pricing formula. Those formulas
explain the prices, but they do not set the prices. Option prices are
set just like prices on anything else in a free market; i.e., supply and
So, how much demand is there for the front month at-the-money contract?
How much demand is there for the back-month? Who's willing to sell
those contracts and how many are they willing to get short?
Black-Scholes cannot quantify supply and demand. If demand for the
front month option is at a fever pitch (i.e., a lot of people are trying
to buy it) but the next month the interest is in selling the same
strike, thereby creating a significant supply, the pricing formulas will
explain that imbalance in supply and demand between the front and back
months through implied volatility.
Consider Nuance Corp. (NUAN). The front month February at-the-money
contract currently as an IV of 73.3% and the March at-the-money contract
has an IV of 47.0%, as per OptionVue. What about April? April's
at-the-money is at 43.4% and July is at 40.5%. Why the spike in IV in
the front month? Well, NUAN reports earnings tomorrow. There is also
some speculation that NUAN is a buy-out target. Whatever the reason,
the demand for NUAN calls in the front month is sufficiently high to
skew its implied volatility much higher than any other month.
Deere & Co. (DE) also has an earnings date coming up. It's front month
Feb. at-the-money IV is 28.5%, the Mar. is 24.7%, and Jun is 26.7%.
With NUAN we saw that massive spike in the front month and a successive
decrease each month we went out. DE has no massive spike and while IV
in Mar. is lower than Feb, Jun higher than Mar. NUAN and DE are very
different companies in very different markets. Quantifying and
explaining the difference in supply and demand for these options in a
formula that can be programed into a computer is no small task.
On the other hand, I know I can look at the DE and NUAN option chain and
recognize the patterns that are present. I can intuitively draw
conclusions about what is going on based upon information I have about
these two companies. I can conclude that there is concern/interest
surrounding NUAN's earnings conference that may effect short-term price
movement but that this concern seems less pronounced in subsequent
months. But, I don't know how to accurately weight the vegas of those
options to determine my true vega exposure if I were to get long the
FEB/APR call calendar.
I recommend that you simulate a similar the trade in the TOS platform use and use the "Thinkback" future located under the 'Analyze' tab this way you can see the what the difference in vega does to a similar position in a binary event. Note that vega is not not the only factor that is affecting the pricing of your position. I would be interested as well to see what happens with the vega in relation to the front to back month vega differential.
Last edited by Cogito ergo sum; July 16th, 2013 at 05:16 PM.
The following user says Thank You to Cogito ergo sum for this post:
Great, thank you for the response. On Monday I actually ended up opening up a calender spread on Google. This is how the position looked like at today's close:
Please register on futures.io to view futures trading content such as post attachment(s), image(s), and screenshot(s).
Google ended up trading about $10 lower than the entry level on the calender spread. Looks like even with the rise in IV for the short legs, burning of theta was enough to post a minor gain heading into tomorrow.
Google reported earnings today and ended up missing. Aftermarket price was $873. Looking at my positions, the short puts are ITM and with Google at $873. That would come to $17 premium on that leg if Google does expire right at that level. Considering the current premium is $9.40 on the leg..that would be nearly $2,820 loss. The short calls..there is $1,000 premium that I will end up capturing. Only question remains is how much I will make on my long puts and lose on my long calls. Since there is still plenty of theta value in those options..those are more difficult for me to calculate. Optimally..it is clear the closer I can have Google trade to $890 the better..or at least to $880.
For an added test, on my Ameritrade account I decided to open another calender trade on Google..and this one was opened minutes before the close. So when IV on the short options were at their very highest.
Here is the actual position:
Please register on futures.io to view futures trading content such as post attachment(s), image(s), and screenshot(s).
It will be interesting for me to see the differences in both spreads. In contrast to my earlier calender spread, the puts are much further out. As long as Google trades above $860 tomorrow, then for the short positions I will take in the $2.65 and a juicy $6.75. So..$265 and $675. Then once again..it is a matter of how my long options end up looking. My long puts will obviously grow in value, but the long calls will lose value. In this spread..it seems that I should have 3 winning legs..so coming out in a gain will not seem like an issue.
I will update this thread tomorrow on both positions. Should be a good for learning. Both positions by the way is with actual money..no paper money. Which is why the new calender from today is such low value..just there for educational reasons.
I see. The thing is, the IV on the long legs does not come off as being high. I forgot to document how high the IV was when I first opened the position, but right now it is at 24.91%. Looking at the leaps..IV tends to be about 22-23%. So I can predict that the IV on my long legs should drop to that same level. Considering I opened the longs days before IV began to really rise..I could have very well opened them when it was a 23/24% IV. So I cannot see myself taking too much of a hit on IV for my longs.
I just went back to ThinkBack to see what the IV was on the long legs and this caught my eye. So the longs were opened when the IV was at 26.42%. The day the IV on the shorts was 54.82%.
Next day on Tuesday: Longs - 25.81% ||| Shorts - 59.57%
Next day on Wednesday: Longs - 25.56% ||| Shorts - 67.48%
Next day on Thursday: Longs - 24.86% ||| Shorts - 80.34%
So..after the position was open, every day after that, the IV on the short legs would rise, but the IV on the long legs would drop. Considering a short position is short vega..the increase in IV on the shorts was filling in more premium against me. Since my long position is long vega, the decrease in IV on the longs should have been taking away value from my options.
Looking at data from about one month ago, the IV on the current expiration date of my longs is sitting at 24.05%. Currently my longs are back at this same IV..so I do not see how I can have too much value sucked out of my longs come tomorrow morning.
Have some more data to work with here. April 18th, the last day before Google reports.
Short leg IV - 94.52%
Long leg IV - 27.43%
The very next day:
Short leg IV - Cannot get a value for this. But the very next week options, IV there dropped from 40% to 23%.
Long leg IV - 21.70%
Here is the other variable coming in, Google went up that day. IV drops when a stock moves higher. Today Google missed earnings, so if anything, that should keep some decent IV in the longs; or at least not take a crazy bit out. Considering those longs went to 21.70% on a +4% day..had it been -4% day..we could probably be looking at a 23% IV or so. Which fits the calculations I made in my prior post. Big difference is that instead of the longs dropping from 27.43% to 21.70% like it did in this example..instead I should be looking at 25% drop to 22%.
By the way..I set up this ThinkBack example with similar strike levels compared to my actual positions right now (% wise), and this spread in April was in the green. On $1,100 risk; it came out at $350 gain. This is with the short call legs going ITM by $15. Which is practically where Google finished trading today in the aftermarket against my short put.
The following user says Thank You to ForeverTrading for this post: