Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
Trading: short fut. opt., fut. spreads, div. stocks
Posts: 57 since Feb 2012
Thanks Given: 3
Thanks Received: 27
Thanks for the report SMCJB.
I see you can download historical data there as well.
Has anybody build a regression model with those values? To predict NG future pricing? and if so. is that working?
i remember schwager mentions it in his book (fundamental analysis, Schwager on futures).
My gut says it can't be spot on but it may be usefull for selling options.
The simple answer is that for the further OTM option there is a lack of aggressive buyers to keep prices supported while there are still buyers for the closer to ITM option.
For example, there are still aggressive buyers for Mar CL 87 puts because there is a chance CL will get to that price but less buyers for <80 puts because the chances of getting there with <42 DTE is lower.
Trading: short fut. opt., fut. spreads, div. stocks
Posts: 57 since Feb 2012
Thanks Given: 3
Thanks Received: 27
I can’t shake the thought about the comment in the karen video that delta is about the same as the chance that an option gets in the money.
Perhaps it’s a rough indicator but somehow it doesn’t feel right.
In order to get a more statistical number would the following make sense?
Let’s take a CL put 80 as an example and assume the future was at 96 at the moment it is sold.
So difference between the future & strike = 96 – 80 = 16
Days till expiration is 42.
Now I take the continues oil contract and start with some date (lets take 1/1/2000)
And I test if, if I had sold a put 16 points below the fut price at 1/1/2000, if somewhere in the next 42 days it would get in the money. (so if oil goes down $16 in the next 42 days)
If not I add one to the ‘not-in-the-money’ counter and if it did, I add one to the ‘it-did-get-in-the-money-counter’.
Now I add one day so 2th of januari and do it again, all the way till today.
That would give me about 3500 samples (14 years * 250 trade days) and that way a % of getting in the money.
Would it be fair to take 1 jan 2000 and the next day (since a lot of data would be the same) or would it be better to skip more than 1 day?
Would this method be more realistic (than using delta)? Or make no sense at all
Comments welcome.
I have the largest gain and drop table for CL. Obviously the months that are still trading are not complete.
If you ignore 2008 & 2009, you have max drops of about 30 and max gains of 27-29.
Also remember that you don't need CL to drop 16 to knock you out of the position. A drop of 10 will surely use up your excess. The 90 settled at $1,100. The $80 settled at $140. The 90 SPAN IM is $1768. The 80 is $380. You would need $2,348 to cover the increased IM and premium of a $10 move in CL if volatility stayed the same.
I doubt you will have about $2,500 excess for the 80 put if the IM was only $380.
It will be a waste of time to go back to the early 2000s because when the oil price is in the 20s you won't get any 16 moves.
Looking at CL contract (last 9 months before expiration) for the years 2007-2013 a drop of 16+ happened 2004 out of 17488 or 11.5% in 56 days. And that includes the last 10 months in which have not have any.
A drop of 10+ happened 23.2%. 20+ happened 7.3%. 25+ happened 5.0%. 30+ happened 3.6%.
A drop of 30+ hasn't happened since the CLq2009 contract.
That is counting starting 56 days from the oldest price and looking at each 56 day period. An average of 208 per contract.
For example, if there was a 20 day period where a contract had dropped by more than 25 from the start to the end then that is counted as 20 times for that contract. Each start and stop of the 56 day measurement using that 20 day period had a 25+ drop.
Trading: short fut. opt., fut. spreads, div. stocks
Posts: 57 since Feb 2012
Thanks Given: 3
Thanks Received: 27
Thanks Ron, very interesting.
I guess the “good part” of IB is you start with a higher margin (but I realize that is ‘inverse logic’, we can start a whole psychological thread about that one
Do I understand correctly that in the table, drops are counted double, eg. both clx & clz contract, if it happened while both where active?
Not sure if that would be fair, although the ‘good guys’ would also be counted double.
I suppose you get burned when moving from a low vola period to a high period (like clq 2008 and further in your table) but when those drops continue the vola would be so high, for the same premium you would sell further away making the chance (fut reaches strike) lower. And obviously the other way around moving out of a high vola period is very nice, (if you know)
So I guess creating a chance table would make sense but it would be a lot better if you add (next to $drop & days till exp.) a third parameter, the current volality.
Trading: Primarily Energy but also a little Equities, Fixed Income, Metals and Crypto.
Frequency: Many times daily
Duration: Never
Posts: 5,057 since Dec 2013
Thanks Given: 4,399
Thanks Received: 10,225
9-Jan EIA Release EARLY Estimates.
First estimates coming in, still subject to a lot of changes.
First survey I have seen had only 11 respondents and an average of -150.8, median -155 Range was -126 to -167
Last year same week -201
5 year average same week -131