.sigma's butterfly book

market returns, negative & positive, displayed on a histogram, resemble cauchy distribution not normal distribution. this actually may help your case of placing flies ATM because in a cauchy distribution, you get more narrow, higher peaks in the middle, and heavyyy tails on both sides.

also, fun fact, if you turn all market returns to positive numbers and turn it into a histogram; you get pareto. check the tails, they're heavy and they have allll the effect on the "mean"; if a mean even exists....
 
Amahrix said:
market returns, negative & positive, displayed on a histogram, resemble cauchy distribution not normal distribution. this actually may help your case of placing flies ATM because in a cauchy distribution, you get more narrow, higher peaks in the middle, and heavyyy tails on both sides.

also, fun fact, if you turn all market returns to positive numbers and turn it into a histogram; you get pareto. check the tails, they're heavy and they have allll the effect on the "mean"; if a mean even exists....
More...

Hey Ama, thanks for the post.

I agree yes market returns do not resemble a normal curve, I don't think anyone has thought that since the days of Bachelier lol but I think it was Osbourne who first discovered
this phenomena with the Cauchy distribution. I agree with you, it makes sense
intuitively, you returns should cluster right around break-even, while having some big outliers positive/negative.

In regards to this helping my case, I'm not too sure. First off I don't just place flies ATM. Yes, usually I do but its not the only way I structure these set ups. I use volatility and expected moves as well for pin risk plays. Spray and pray baby.

Cauchy distro looks like a Low IV distro in the body, and high IV distro in the wings. The lower the volatility the more expensive the fly will be, and the higher the vol the cheaper.

I'm not sure what you mean by the "mean" not existing. Its a statistical equation, measuring dispersion of data, its to measure relativeness and averages. I know you know this but yeah lol.

Or, do you mean in regard to the Cauchy distribution compared to the normal? The Cauchy doesn't obey the law of large numbers, the avg location never converges to any fixed number. Meaning highly "unstable average". Cauchy mathematically also has no finite moments, I.e., mean, variance, but can be normalized.Its very interesting.

The only thing the Cauchy and normal distro have in common is their modes (ii think)
 
.sigma said:
Or, do you mean in regard to the Cauchy distribution compared to the normal? The Cauchy doesn't obey the law of large numbers, the avg location never converges to any fixed number.
More...

yea thats what i mean. and i see what you're saying. hope you're right!! :)

I learned from Mandelbrot+Nassim about the Cauchy. Never heard of Osbourne. Will look into it for fun.

Thanks.

perhaps you're on to something and i only wish you the best of (luck)? :p <3
 
Last edited:
Amahrix said:
yea thats what i mean. and i see what you're saying. hope you're right!! :)

I learned from Mandelbrot+Nassim about the Cauchy. Never heard of Osbourne. Will look into it for fun.

Thanks.

perhaps you're on to something and i only wish you the best of (luck)? :p <3
More...

I could never find shit about Osbourne! THe only source I have is the book "Physics of Wall St."
 
expected gain.PNG
 
.sigma said:
For example, if I'm thinking of setting up an iron butterfly, and plan to leg in would I sell the ATM straddle first? Then buy the OTM strangle? Does it depend on current IV of the strikes in choosing which leg in to order entry into?
More...

Selling the ATM straddle is not recommended - as you mentioned, firstly cos of margin, and more importantly, you will have a naked position which could turn real nasty, very quickly. (Even a big up day like today, when the SPX has risen 5% can leave you in pain).
Legging an iron fly would be done with either the put side first or the call side first, depending on the underlying. Eg. if the market has had a down-day, and you expect a bounce, sell the put credit spread first, and if/when the bounce comes, then sell the call credit spread. I don't leg in to open an iron fly, but sometimes leg out when closing (closing one side and effectively converting a IF to a credit spread). Not too often though, and it's usually been with TLT due to it's somewhat mean-reversion tendencies.
 
.sigma said:
Also about legging into flies. I'm not too sure to go about this.

For example, if I'm thinking of setting up an iron butterfly, and plan to leg in would I sell the ATM straddle first? Then buy the OTM strangle? Does it depend on current IV of the strikes in choosing which leg in to order entry into?

If we're talking natural 121's would you first buy the long call spread? Then sell the short call spread? Or vice versa? Or neither? Instead just sell the two ATM calls, then leg the wings? Of course if you sell uncovered your margin is way different.. And someone like me doesn't sell naked ever.. I should read RA combo to fly conversion thread again! But I'm curious what have you guys done or think about legging flies?
More...

I have been exploring the idea of legging into butterfly spreads myself when it appears the market has a short term directional bias. If one is a consistently profitable intraday trader, it seems to make sense to enter a vertical spread that would benefit from the anticipated price move and to hold off on entering the other vertical until the outlook of the underlying either changed or it is near end of the trading day.

Glorious is the long butterfly done at a net credit, although for trade tracking purposes, it is probably best to account for this type of trading separately.

The downside to doing this is increased trade management requirements and potential frustration when calling the market wrong.

I will probably end up legging in a third or so of my 'fly trades in the future, given anticipated time constraints watching the market.
 
BeautifulStranger said:
If one is a consistently profitable intraday trader.
More...

…..then one wouldn't need to leg into verticals, he could just trade delta1 and scalp all day... unless he was hedging or something. But I know what you implied, and I agree, if one has a not bad sense of market direction one can leg into these verticals.
 
You must log in or register to reply here.
Back
Top