We(I) need one!
Any ideas for a topic?
Also, we need another emilov to play with.
We(I) need one!
Any ideas for a topic?
Also, we need another emilov to play with.
How about something along the lines of...
The importance of picking the right direction of price movement, for winning with option plays, for mug punters.
eg if you pick the right direction do you always win??
In the interests of promoting more option discussion on this forum, I'll start with;
Calendar spreads - I just don't get them.
To clarify, by Calendar spread I mean the traditional 'long' one where you buy the far month and sell the near month, both at same strike, which is usually at the money.
A cal spread is supposed to make its money by the greater theta of the near month outweighing the far, but this actually only applies if the options are at or near the money, and any decent move from the strike takes it into loss. In fact the payoff profile is so pointy near the point of max profit that its never going to happen and the profit range so narrow its not going to finish within it very often. The only way a long calendar shows a decent profit is if the underlying finishes virtually unchanged, in other words it wants little or no volatility.
A long calendar is supposed to be appropriate for times of low IV right, the thinking being that when IV is low you are going to be paying a lower absolute amount for the spread, and that an increase in IV will affect the far month more than the near month, increasing the value of the spread. in other words we are positive vega, in other words it wants an increase in Implied Volatility. This is the alternative way to make money
Now my problem with the calendar is that the two 'wants' in bold are mutually exclusive; they are conflicting; you cant have both , or even have one with no effect on the other.
IV just does not rise by any significant amount while the underlying stays virtually still. The only way you are going to get your "increased IV" wish is if actual volatility increases, and if the underlying moves by any decent amount then no amount of IV increase on options that are by now way in the money or out of the money are going to make a profit out of this trade.
On the other hand suppose you get the "no volatility/movement" wish and the underlying does nothing till near expiry and it finishes right at the strike. Now it is a dead cert that IV will have dropped some more in the meantime (unless it was already at the lowest it could ever be, which is hard), so when you go to sell the far month it isnt worth what it was supposed to be. So your max profit really isnt even as good as the max profit in the ridiclously point payoff diagram you started with, and the same thing has reduced the effective profit range as well.
Apart from a contrived scenario where the underlying experiences decent volatility and happens to finish near the strike when the music stops, I dont see how this could ever have a positive EV.
Discuss. (or just rip into me instead)
I don't trade straight out calendar spreads much for pretty much the reasons you outline.
However 3 points.
- Implied Vol doesn't necessarily follow stat vol.
- You don't have to let the front month option expire
- You don't have to trade the ATM stread
For instance, is there is a situation where the underlying can stay pretty quiet with rising IV - earnings anyone? There is no law that says you have to let the front month expire before taking a profit, or that you have to let the announcement happen before trading out of it.
Another way to us the calendar spread is as a low risk directional trade with +vega and +theta. An OTM put calendar for example.
Once again you don't have to wait till front month expiry to close the trade and collect a profit.
Just some thoughts.
true IV doesnt nec follow stat vol, in fact it does fluctuate around a bit on its own, but the magnitude of these fluctuations is not usually enough to make it profitable. To put some numbers to it, if IV =20% when opening, it may well drift back up to 21 or 22, but even if you catch the top of that move I dont think that would be enough to overcome time decay, trading costs and bid ask spread if any, at least by enough to show a respectable profit. If IV went to 25% that would be enough to generate a profit, but i am saying changes of that magnitude dont happen in absence of actual vol, or at least not very often.
if a temporary IV increase is expected due to earnings report or similar, would not a straight vol play (eg straddle purchase) be better?
You don't have to let the front month option expireCould you explain with maybe an example of what you mean? I havent crunched every scenario but would have thought that since IF you find yourself in the position the options are at the money and the spread is in profit , the nearer to expiry the more the difference in theta opens up, the more incentive there is to let the near month expire, and that closing early reduces max profit some more?
You don't have to trade the ATM stread. Another way to us the calendar spread is as a low risk directional trade with +vega and +theta. An OTM put calendar for example.
agree totally. an OTM calendar is a completely different animal which is why i was careful to refer only to ATM spreads.
I am interested in the idea of using long delta strategies to capture seasonal volatility in commodity options. I have been using my long delta model on QQQQ, SPY, XLF, GDX, XLE with some success (thanks to much help from mazza and wayne), and while examining its potential in commodity markets I noticed some signals seemed to match up with what I remembered of commod seasonality.
This seems like it could work ok (theoretically), because usually if traders expectations of normal seasonal occurences aren't met, there is a good chance of volatility as spec positions on the underlying are unwound/crushed?
So we would have a leg up on the normal commod traders, who have to trend or fade with their idea of what seasonality would bring. We just know at X time of year, expectations of supply/demand will shift rapidly and should long delta position accordingly in advance.
So, perhaps as an example, position long delta at the start of Jan to close out the end of Feb, and same at start of June to close out the end of July. Obviously filtering opportunities with an eye to vega risk.Corn: This market's seasonality can be divided into three time periods: late spring to mid-summer; mid-summer to harvest; and post-harvest. The most pronounced seasonal trend in corn is the decline of prices from mid-summer into the harvest period. Prices are often near their highest level in July because of factors associated with the old crop and uncertainty over new crop production. Even in years when a price decline begins before mid-July, it can continue after mid-July if the crop outlook is favorable. Harvest adds large supplies to the marketing system, which normally pressures prices to their lowest levels of the crop year. Prices usually rise following harvest. However, the "February Break" is a well-known phenomenon whereby corn prices usually show some degree of decline during the month of February.
Position in advance for Jan expectations by going long delta start of December, position in advance for Nov expectations by going long delta start of October?Cotton: Cotton is a market where the "trade" has very heavy participation and seasonals tend to be a function of heavy deliveries issued against the expiring futures contracts--December, March, May, July, and to a lesser degree, October. In November, the market tends to recover from harvest lows, and then in January the market tends to back off to lower levels.
Just a thought. Not sure if this is the sort of suggestion you had in mind for the thread. I really like the idea though. If traders seasonality expectations are right, we might only make a tiny little bit of money. But if they are wrong, we stand to gain bigtime. By only positioning long delta during times when we know real supply and demand is hitting the market (due to harvest or whatever) we avoid our chances of the price going nowhere for the duration.
Disclosure: Long cash, gold, stocks.
If setups are available, steep tenor skews [>10% bpts] to flatten + falling stat vols --> calendar > long straddle for earnings playsif a temporary IV increase is expected due to earnings report or similar, would not a straight vol play (eg straddle purchase) be better?
Short gamma outweighs theta, even close to expiry. Loose qualifier - I personally prefer to offset once it touches neutrality.IF you find yourself in the position the options are at the money and the spread is in profit , the nearer to expiry the more the difference in theta opens up, the more incentive there is to let the near month expire, and that closing early reduces max profit some more?
Anyone who has tried trading atm calendars as "income" plays, will have experienced at one point, any theta gains over weeks, wiped out by one day's movement. Hence Wayne's point about offsetting the short, prior to expiry.
It can be modeled quantitatively by plotting the partial derivatives against each other. You'll observe the curvature is different.
I see no issues if you're using similar parameters to your index/equity model.
Sorry if this has been covered in another thread, But I couldn't find an answer anywhere.
When it comes to index options, Is each index point worth $1 or $10. I have read conflicting infomation from different sources.
Glad you like the idea. Things have developed very nicely since we spoke last.
The above is just an example, I haven't tested it out, and thinking about it now, there must be some differences between options on cash and options on futures which I probably didn't consider.
Perhaps this difference would be a good topic to discuss? I am certainly interested to learn about it.
Disclosure: Long cash, gold, stocks.
I have been reading to try and expand my knowledge on put options ( until now I have only ever sold covered calls )
I noticed in one of the texts I was reading it mentioned that ACH pays interest on the margins it collects from puts.
My question is does commsec (or any other brokers) pass any of this interest on to the customer.
mazzatelli, thanks for your reply.
do you mean like this;?
This is an attempt to plot gamma v theta as you suggest. the pink and blue lines are for a 100 XYZ call, XYZ=100, int rate=5% Vol=15%, using an option model. for comparison purposes gamma is rebased so it equals theta at 28 calendar days to expiry.
The turquoise and yellow lines represent the net gamma/theta of a Call calendar spread commencing at 28/56 calendar days to expiry ie four weeks apart.
Only trouble is I am not seeing any indication that gamma outweighs theta using this method. am i doing it wrong, or perhaps you are meaning spreads that start further out in time?
By gamma risk I suppose we are really referring to a risk of 'movement away from current price', as in a long calendar or short straddle it hurts us either way (assuming the current price is at the strike we like)
i guess i have taken a differnt approach in the past in that instead of loooking at the gamma number, i tend to look at the (historical) actual distribution of returns for that stock/index over the corresponding period and seek to determine whether closing or running it is the highest EV, and that is what i base my decison on.
In most case i seem to get the result that it is better to run it, in other words the cost to close in preimum outwieghs the risk of a movement outside the current profit zone, which has led me to a rule of thumb that;
if i find myself short an option at the money with short time (say 1 day up to 2 weeks) to go it is better to let it run to expiry (unless under threat of early assignment) and
if i find myself long an option at the money with short time to go i am better off taking the premium available back at that point rather than 'hoping' for a favourable movement bigger than the premium
This rule of thumb seems to be opposite of what yourself and wayne are saying, which is something of a worry; if i have got it wrong i need to know about it.
In the case of a calendar spread, it seems to me a moderate move away hurts us more than it would if short a straddle/iron fly, so in that case i can certainly see (now) why it would be better to close a long calendar as the underlying converges or crosses the strike. still not sure about the other situations though..