Vol trading discussion

M

maggie12333333

I browsed around reddit randomly and found this post recently in reddit:

https://www.reddit.com/r/path%3D%252Fr%252Foptions%252Fcomments%252F7rbtfp%252F

This is a post quite a while ago but I found it interesting therefore I created post here for discussion.

This is the volatility smile graph that he shared:
https://i2.wp.com/www.marketcalls.in/wp-content/uploads/2010/08/Volatility-Smile.jpg

As he mentioned in one of his reply he suggested to buy the 5600 strike put and sell the 5700 strike put when there is a "kink" in the volatility smile, in order to take advantage of vol spread between the two.

What I don't understand is I suppose the trade should be delta hedged by futures. I.e. if the underlying stays at the middle (say 5650) during expiry which the short 5700 strike put will incur loss of 50 and the long 5600 strike put will be expired worthless. The trade will incur a loss of $50 per contract which is a substantial loss for a tiny gain of the vol spread. How to deal with this risk?

If delta hedging is involved (e.g. open short futures position when price under 5700 and unwind the short futures position when price under 5600) what happen if futures keep being in and out of the strike (i.e. oscillating around the strike level) which significantly increase the hedging cost?

If someone have any thoughts on this please share. Thanks so much in advance and I really appreciate if you can help with my questions.
 
"I'll answer any questions you have (including politics, religion, video games, bodybuilding, volleyball) except uber specific details on our particular models."

Is the core of trading system based on religious principles?
 
Last edited:
Well, the original author explained it this way:
“This is incredibly complicated stuff that isn't going to translate well over Reddit. Vol scalping at our level requires you to have a pricing model and correlation matrix. There's some decently high level math involved in this and not something I can really get across here but basically you sell the things that cross your theos and buy the ones that go below.”


So it’s not simple, and you have to spend time developing vol models and getting experience trading them.

I do something similar in a simpler way by trading ratio spreads: selling more of the high-vol options than buying those with lower vol. But I also use models that tell me the proper ratios and strikes to use to keep my risk in check. And even then, this is not a big money maker because best opportunities don’t show up often. That guy may just have more evolved and advanced models, if he is still making lots of money using that approach.
Your issue though, may be to try to copy someone vs develop your own models and trading methods. Because once a simple method can be copied by everyone then it usually no longer works because everyone will try to sell the same options, thus making them cheaper until they’re too cheap to sell. This is actually how VIX itself went to levels below 12 or even 11 in recent years - everyone learning how to sell volatility and copying everyone else.
While you need to be able to outsmart everyone else, not just ask how to copy them.
 
guru said:
Well, the original author explained it this way:
“This is incredibly complicated stuff that isn't going to translate well over Reddit. Vol scalping at our level requires you to have a pricing model and correlation matrix. There's some decently high level math involved in this and not something I can really get across here but basically you sell the things that cross your theos and buy the ones that go below.”


So it’s not simple, and you have to spend time developing vol models and getting experience trading them.

I do something similar in a simpler way by trading ratio spreads: selling more of the high-vol options than buying those with lower vol. But I also use models that tell me the proper ratios and strikes to use to keep my risk in check. And even then, this is not a big money maker because best opportunities don’t show up often. That guy may just have more evolved and advanced models, if he is still making lots of money using that approach.
Your issue though, may be to try to copy someone vs develop your own models and trading methods. Because once a simple method can be copied by everyone then it usually no longer works because everyone will try to sell the same options, thus making them cheaper until they’re too cheap to sell. This is actually how VIX itself went to levels below 12 or even 11 in recent years - everyone learning how to sell volatility and copying everyone else.
While you need to be able to outsmart everyone else, not just ask how to copy them.
More...
I'm just at the beginning of this topic, but this is how I would break it down: current Volatility is significantly outside the historical Volatility Distribution for the strike/option. Is this how these Volatility models work?

What lvl of significance do you use for it (0.01?). Do you go with the distance to the theoretical price on reversal and calculate the potential gains when it reverses as a threshold? What do you use as reference metric wise for the potential profits? Do you go with the average volatility=current volatility/price- average options price or do you go with the end tail 99% as a reference to calculate the expected Gain from the trade? What is the threshold, where you pull the trigger? What is the ROI of these strategies?

Seeing as ppl always go with a calendar when doing these, there must be a threshold in order for the price to hit gamma convexity to make these trades profitable. Do you trade the same strike in different periods and go Long the one with a longer expiration? Or do you choose a Long Strike that is undervalued in period X and go short the overvalued strike at expiration y? What is the min expiry for the Long Leg?
 
Last edited:
Atikon said:
I'm just at the beginning of this topic, but this is how I would break it down: current Volatility is significantly outside the historical Volatility Distribution for the strike/option. Is this how these Volatility models work?

What lvl of significance do you use for it (0.01?). Do you go with the distance to the theoretical price on reversal and calculate the potential gains when it reverses as a threshold? What do you use as reference metric wise for the potential profits? Do you go with the average volatility=current volatility/price- average options price or do you go with the end tail 99% as a reference to calculate the expected Gain from the trade? What is the threshold, where you pull the trigger? What is the ROI of these strategies?

Seeing as ppl always go with a calendar when doing these, there must be a threshold in order for the price to hit gamma convexity to make these trades profitable. Do you trade the same strike in different periods and go Long the one with a longer expiration? Or do you choose a Long Strike that is undervalued in period X and go short the overvalued strike at expiration y? What is the min expiry for the Long Leg?
More...


I don't use calendars for basic vol smile arb, just same date for both/all strikes.
Everything else you wrote/asked doesn't relate to anything I'm doing. All I see is free $500 bucks and I take it (as an example), while my only safety is empirical evidence of what worked in the past and why, and experience staring at options chains for thousands of hours (incl writing down some rules & formulas). Others may do this differently.
 
guru said:
I don't use calendars for basic vol smile arb, just same date for both/all strikes.
Everything else you wrote/asked doesn't relate to anything I'm doing. All I see is free $500 bucks and I take it (as an example), while my only safety is empirical evidence of what worked in the past and why, and experience staring at options chains for thousands of hours (incl writing down some rules & formulas). Others may do this differently.
More...
So you go long short straight without defined risk/reward or do you just trade the spread depending on where you think the reversal will be based on your observations(?or historical distribution?)? I don't get how my post doesn't relate to it, when you mentioned that you look at what worked historically.
 
Last edited:
I think the poster intended to do a hedged risk-reversal which is more easily done on a 1 to 1 basis.

Why go thru a multi-lot put spread and the necessary framework for dynamic hedging of deltas on a Reddit post?

He is playing the Pied Piper. ET is all you need. Everyone gets debunked. At least, within 2 business cycles.
 
maggie12333333 said:
I browsed around reddit randomly and found this post recently in reddit:

https://www.reddit.com/r/path%3D%252Fr%252Foptions%252Fcomments%252F7rbtfp%252F

This is a post quite a while ago but I found it interesting therefore I created post here for discussion.

This is the volatility smile graph that he shared:
https://i2.wp.com/www.marketcalls.in/wp-content/uploads/2010/08/Volatility-Smile.jpg

As he mentioned in one of his reply he suggested to buy the 5600 strike put and sell the 5700 strike put when there is a "kink" in the volatility smile, in order to take advantage of vol spread between the two.

What I don't understand is I suppose the trade should be delta hedged by futures. I.e. if the underlying stays at the middle (say 5650) during expiry which the short 5700 strike put will incur loss of 50 and the long 5600 strike put will be expired worthless. The trade will incur a loss of $50 per contract which is a substantial loss for a tiny gain of the vol spread. How to deal with this risk?

If delta hedging is involved (e.g. open short futures position when price under 5700 and unwind the short futures position when price under 5600) what happen if futures keep being in and out of the strike (i.e. oscillating around the strike level) which significantly increase the hedging cost?

If someone have any thoughts on this please share. Thanks so much in advance and I really appreciate if you can help with my questions.
More...

To lock in "kinks" in the vol smile you have to delta hedge to lock in the IV diff anomaly. The hedge is done once...and then done again the opposite way when you exit the trade. To minimize the vega/gamma risk you need to ratio the spread appropriately.

On Reddit, he points out that because the CBOE prioritizes retail orders over "professional" MM bid/offers, you will get filled on market orders hitting the bid/lifting the offer before the MMs if you have a platform/system that allows you to join live bid/ask markets. This is a nice advantage, but a huge undertaking with regard to risk management, proper hedging, utilizing a robust trading platform with reliable real-time quotes, and big capital if you are going to mass quote and manage a whole book/options portfolio.

I would not recommend any individual retail trader to take down the size he was recommending, especially in something like the VIX. If you don't know what your doing, or your "joining" quotes are too slow to pull out during big vol moves, you can get run over pretty bad and get stuck with a position you don't want. Also, if you are coattailing the bid/asks of the MMs to get on orders you can fall victim to market manipulation and spoofing by the MMs if you're not paying attiention. So mass-quoting just because you get priority over the "professionals" is NOT a good idea for the individual retail trader. You're better off staying small, being nimble, and picking your spots.
 
You must log in or register to reply here.
Back
Top