Which puts to buy for tail risk hedging?

[newwurldmn]

There is no such thing as an optimal hedge. The hedge will be based on several factors which may or may not pan out when the market sells off. second everyone will be hedging for a different reason and a different situation.

How did you determine those parameters? Are you sure they are optimal?

 

[dorietrading]

I ran some backtests on SPX (open 1 every day if vix < 20), see results:

 

[john7722]

Would it be a good idea to do the tail risk hedge using leveraged 3x options instead of non -leveraged?

 

[dorietrading]

Would it be a good idea to do the tail risk hedge using leveraged 3x options instead of non -leveraged?

Which symbol would you like to use?

 

[john7722]

UPRO for example

 

[dorietrading]

UPRO for example

Spread is almost 30 cent for the options. So maybe the transaction cost of 3*SPY is less then the spread loss...

The spread in UPRO will be much bigger when there is a big downmove. So i would go for 3*SPY.

 

[Matt_ORATS]

I have read the paper "Capital Asset Pricing Mistakes: The Consistent Opportunities in Tail Hedged Equities" by Chitpuneet Mann, Mark Spitznagel, and Brandon Yarckin.

They define tail risk event as a down 20% move in the S&P 500 over a month. They recommend holding SPY puts with strike roughly 30-35% below spot and 11-12 weeks to maturity.

Are these parameters the best? In other words, which options appreciate the most in case of the tail risk event as defined above?

How to determine it quantitatively? I have some Python skills but no access to the paid data.

Here's how you set up Spitsnagel's:

Universa also will close part of their hedges say down 20% or when the OTM% reaches 90% from 70%:

Not much protection exiting there.
Here's with no exit.

Note: no winning trades.
https://wheel.orats.com/backtest/1LZBvmw7dFqePHsSymh5BbunQojeuCjrg4

Here's one of my favorites ~500 DTE:

https://wheel.orats.com/backtest/1KJnyksBfcu1UKT8dJDWMgrt8uHmBpXY75
And this 120 DTE:

Here's a VIX long call that was mentioned. This is one of the better ones:

Here's the link to this backtest (requires a trial subscription special in my signature below):
https://wheel.orats.com/backtest/1A6EEHRKWzxCNVJSLVFWWr2PB3vJ2kVUhN

I am working on a more complex trade with some promising results:

If you are interested in finding out more about this premium test or others let me know.

 

[Magic]

I use puts which are about 13% below spot, 28DTE. These puts have cost me around 17K per 100K invested from 2013 till now. So approx 2K per year. Biggest win was in week of 18 march 2020, 35K per 100K invested profit.

During low vola you can also buy low delta calendars, to minimize your theta loss.

Dorie, was the 100K per put all invested in the SPY as well? Is there an easy way you can pull up the realized sharpe ratio from inception to the end of 2019 on your portfolio? Thanks.

 

[dorietrading]

Dorie, was the 100K per put all invested in the SPY as well? Is there an easy way you can pull up the realized sharpe ratio from inception to the end of 2019 on your portfolio? Thanks.

No, I'm an option writer. I sell SPX options and buy the puts to lower my delta/vega exposure.

 

[Zwaen]

I also use puts for continuous protection. In my rigorous backtests the longer dated, the better. I chose somewhere between as 1.25-1.75 year as an optimum expiration level. The longer dated puts have lower delta, but higher vega. With this in mind, you can also play with a different number of options versus the capital invested.

I read an interesting study a while ago that it would be better to switch from longer to shorter duration after a vix spike, because you would not be hurt as much by a vola collapse afterwards. When I tested this myself I found otherwhise in the data, strangely.

The charts show the cumulative returns (normalized for 100.000 in spy) for different durations of put options, from 0.50 year to 2.0 year, rolled every 3 months.

SPY 2015-2021:

And last dip/crash more isolated: