Long vol through short ITM VIX puts

[guspenskiy13]

Hi everyone,

I was wondering, if anyone has utilized this strategy, either naked or as a spread...I was searching for information, but did not find anything specific yet...

It seems that selling to have a decent return/risk profile, especially when /VX is cheap and has plenty of DTE...

As I've looked at VIX settlement prices - there are only 4 instances, when the settlement was sub-12.
Then, as an example (which might be an exception from the rule and a huge mispricing, given extremely low vol we are experiencing right now)...

02/24
/VX Mar @12.3 w/ 35 DTE
VIX Mar 12.5P @1.00

Selling 12.5 put would cover losses up to 11.5 settlement. I mean, with 35 days to go and continuation of extremely low vol - we could exit on any small increase in /VX and give up 0.05 - doesn't matter.

Overall, to sum it up:
in any instance, when an ITM VIX Put has plenty of DTE and can cover losses up to $12 or low 12s settlement - it could be sold as a very low risk/decent reward way to go long volatility. The only premium, which we could pay is the Roll Yield.

I've also briefly looked at calendar puts, which is more neutral (Term Structure/Skew) play...got an impression that it would require a lot more data...meanwhile, I'm thinking if it would make sense to relate the VIX Put prices to VX, DTE and Roll Yield...

Even looking at this right now, when /VX May gets sub-15 w/ 64DTE and 12P quotes @0.35 bid...I understand, that there is technically "no floor", but given the amount of time/current low vol/positive skew of VIX...seems as a decent bet.

Thank you,

I highly appreciate any comments/input.

 

[water7]

you can backtest & livetest this concept :]

start experiments with simple parameters (price, DTE, contango) and make your own buy/sell indicator and develop from there

or you can just long vol (and trade the skew) just because it seems a decent bet
it is up to you..

:]

 

[guspenskiy13]

start experiments with simple parameters (price, DTE, contango) and make your own buy/sell indicator and develop from there

Price of the put option is the main variable. What makes it expensive?
I suppose the overall level of the term structure. Spot VIX or M1 is meaningless; when the further months fall significantly - that's when vol is truly cheap...

M3 dips to 14 with 63 DTE here in Feb...that was cheap...managers actually went long on that opportunity, first significant increase since idk when...maybe election...

 

[sle]

You gonna find that your short VoV vega and VIX gamma were will dampen the effect of your long delta. Also, the low strike put IVoV is fairly low so you get a vega convexity hit when a shock happens.

 

[El OchoCinco]

Hi everyone,

I was wondering, if anyone has utilized this strategy, either naked or as a spread...I was searching for information, but did not find anything specific yet...

It seems that selling to have a decent return/risk profile, especially when /VX is cheap and has plenty of DTE...

As I've looked at VIX settlement prices - there are only 4 instances, when the settlement was sub-12.
Then, as an example (which might be an exception from the rule and a huge mispricing, given extremely low vol we are experiencing right now)...

02/24
/VX Mar @12.3 w/ 35 DTE
VIX Mar 12.5P @1.00

Selling 12.5 put would cover losses up to 11.5 settlement. I mean, with 35 days to go and continuation of extremely low vol - we could exit on any small increase in /VX and give up 0.05 - doesn't matter.

Overall, to sum it up:
in any instance, when an ITM VIX Put has plenty of DTE and can cover losses up to $12 or low 12s settlement - it could be sold as a very low risk/decent reward way to go long volatility. The only premium, which we could pay is the Roll Yield.

I've also briefly looked at calendar puts, which is more neutral (Term Structure/Skew) play...got an impression that it would require a lot more data...meanwhile, I'm thinking if it would make sense to relate the VIX Put prices to VX, DTE and Roll Yield...

Even looking at this right now, when /VX May gets sub-15 w/ 64DTE and 12P quotes @0.35 bid...I understand, that there is technically "no floor", but given the amount of time/current low vol/positive skew of VIX...seems as a decent bet.

Thank you,

I highly appreciate any comments/input.

Just a cusrosry look, I find it hard to believe the MAR 12.50 put was @ $1.00 on that date looking at historical prices.

 

[water7]

Price of the put option is the main variable. What makes it expensive?
I suppose the overall level of the term structure. Spot VIX or M1 is meaningless; when the further months fall significantly - that's when vol is truly cheap...

M3 dips to 14 with 63 DTE here in Feb...that was cheap...managers actually went long on that opportunity, first significant increase since idk when...maybe election...

- you use "technical level" in VX3 and attempt to time the next cyclical move
there's nothing wrong with that..
people can use VX1,2,3,4,5,.. to express their view

- it's better if we can test this concept into more detailed trading plans
doing a lot of experiments is more useful than hypothetical thinking

- you already talked about roll yield in the first post, maybe you want to incorporate the data here?

- why do you want to short ITM VIX put? instead of long ITM/ATM VIX call?
what is your expectation of the skew?


thanks

 

[sle]

The motivation for the trade is pretty clear - he wants to be long vol while not paying excessive risk premium. This is also known as "have the cake and eat it too; and fuck the baker while at it".

Theoretically, it's probably possible to find a situation when VIX is cheap at that forward while vol of vol is expensive, but I have only seen that once or twice. As I said, the key problem is that your vol of vol exposures will be a powerful offset to the long vol delta.

Ps. Do we want to have a more detailed conversation about vol of vol dynamics?

- you use "technical level" in VX3 and attempt to time the next cyclical move
there's nothing wrong with that..
people can use VX1,2,3,4,5,.. to express their view

- it's better if we can test this concept into more detailed trading plans
doing a lot of experiments is more useful than hypothetical thinking

- you already talked about roll yield in the first post, maybe you want to incorporate the data here?

- why do you want to short ITM VIX put? instead of long ITM/ATM VIX call?
what is your expectation of the skew?


thanks

 

[water7]

The motivation for the trade is pretty clear - he wants to be long vol while not paying excessive risk premium. This is also known as "have the cake and eat it too; and fuck the baker while at it".

Theoretically, it's probably possible to find a situation when VIX is cheap at that forward while vol of vol is expensive, but I have only seen that once or twice. As I said, the key problem is that your vol of vol exposures will be a powerful offset to the long vol delta.

Ps. Do we want to have a more detailed conversation about vol of vol dynamics?

you just spoiled the cake twist

way too early..

 

[guspenskiy13]

You gonna find that your short VoV vega and VIX gamma were will dampen the effect of your long delta. Also, the low strike put IVoV is fairly low so you get a vega convexity hit when a shock happens.

What if we take the assumption of holding until the expiry?

- you use "technical level" in VX3 and attempt to time the next cyclical move
there's nothing wrong with that..
people can use VX1,2,3,4,5,.. to express their view

Not quite 'technical' IMO...I've pulled some /VX data recently; /VX price vs. Trading Days until the Expiry...
The numbers are quite interesting, especially for extreme readings for 120 Trading-DTE...from both sides: long and short..

Which makes sense. Because when your median settlement is ~15.8, with a high concentration in the 14.5-15 range...And there is a /VX with 120 trading days to go, which trades at 23s...It makes a lot of sense to short it...
On the other hand, when the same DTE contract trades sub-16...the risk is different...

90th percentile /VX values at high DTE have 100% short profitability, with an average gain of 4-5p.
10th percentile /VX values at high DTE have 90% long profitability - mostly indicated the end of vol cycle..

What I am saying, is that low VIX values mean nothing really in terms of "cheap"/"expensive". It's all about the term structure; the M4+ contracts - where the calls are bought or futures contracts are sold.

When M6-M7 is at 23 - there is plenty of "juice" and nice R/R to short it. When M6-M7 is sub-15 - the R/R of a short doesn't look attractive at all.

I really do believe, that /VX should be viewed as a function of: 1) absolute price; 2) trading days until expiry; 3) roll yield. I have incorporated the first two - still working on the third.

- it's better if we can test this concept into more detailed trading plans
doing a lot of experiments is more useful than hypothetical thinking

- you already talked about roll yield in the first post, maybe you want to incorporate the data here?

I will post some actual data in regards to /VX prices at high DTE - and their performance. As I've said, extreme values on both sides have very high probability.

- why do you want to short ITM VIX put? instead of long ITM/ATM VIX call?
what is your expectation of the skew?

Because IMHO, it has greater probability and I can do it with greater size. I've shorted small /VX May 12P @0.4. The losses are covered up to 11.6. On 10 contracts; I gain $400 for settlement >12; I lose the same $400 for settlement @11.2. If you look at the distribution of VIX settlements, there were only two below 11.5.

Seems as a decent bet.

 

[sle]

What if we take the assumption of holding until the expiry?

Then, unless your are buying really flat term structure, the roll down will kill ya

This strategy is very similar to selling TLT puts when you felt you wanted to be long bonds as a form of protection. The key difference is that usually the forward drop as well as the skew is in your favour on TLT and against you on VIX. So that brings us back to the original issue of the cake ownership...