A deeper dive into our approach to volatility trading

Answering your most frequent questions

The price action yesterday was fairly depressed throughout the day, particularly at the end of the session when intraday realized volatility skyrocketed. Many attributed this to month-end flow.

That's a plausible explanation—nothing like the worst month since September 2022 to dump all the skeletons in your closet and clean up your portfolio. This is way beyond our pay grade and not the kind of stuff we follow, so we don't have an opinion.

At Sharpe Two, we aim to analyze volatility in a practical manner, using data analysis rather than complex mathematical modeling, straddle prices instead of annualized volatility, and only three to five years of data rather than ten.

"OMG! But this is all wrong! How can you pretend to trade volatility with such a poor setup?" (I bet you only have two screens too.)

Yeah, we've heard that before.

Today's post was inspired by our friendly and constructive conversation last week about some elements of our approach. We thought it would be nice to reiterate why we do things the way we do.

As a side note, we're not pretending this is the only way. Absolutely not. In trading, there are multiple ways to skin a cat, and we aim to do so with our limited means as retail traders, proposing some frameworks for other retail traders and investors with limited capacity. If you feel like hedging your book with third-degree Greeks and a deep understanding of dealer positioning, we'll assume you already know how to find good trades, have amazing infrastructure support around you, and that your math skills have put you at the top of the food chain. Good for you, and you may not need to read Sharpe Two.

Why do you seem to only look at straddles?

At Sharpe Two, we aim to approach volatility trading with the highest degree of practicality. In that sense, the cleanest way to measure how the market is pricing what will likely happen soon is to examine at-the-money straddle prices.

Why not look at the implied volatility and its annualization, then?

Implied volatility and its annualization also use the information contained in options prices. Through different mathematical approximations and estimations, you end up with a nice number that is handy for comparing things across the board.

For instance, it is much easier to compare the market's view along the expiration cycle of a specific product or to put different tickers in perspective. This method is great at scaling. But like any scaling method, it may compress some details contained in the rawest form of information—the option prices.

Our favorite approach to trading volatility is identifying the part of the curve where the variance risk premium is the most stretched. To do this, we compare the price of straddles to an average of the most recent moves in the underlying. A concrete example: when determining if a 14 DTE straddle is a good trade, we will compare the current straddle price against an average of the last 14 days' moves observed in the underlying (no overlapping). Is it perfect? Absolutely not. Does it do the job? We think so.

VRP is mostly here in EWZ, except during the “exceptional” time before the election in October 2022.

As a retail volatility trader, you are unlikely to say, "I sold 16% in 30 days in SPY and bought 14% in IWM to hedge." Your book should be much simpler anyway, with more practical problems like the margin required for a given straddle. Once again, prices have more immediate information for your trading style.

Talking about margin, why only straddles? Aren’t they super risky?

When you think about it, you already know your maximum exposure when you put on a straddle. The worst that can happen is things move extremely against you, and you are long or short the number of contracts you sold times the number of stocks per contract. It is relatively easy to hedge, despite popular belief and those who have a vested interest in having you take safer positions (yes, brokers, we're looking at you).

You will also avoid nasty surprises where a sudden rise in implied volatility will challenge your portfolio while the market is still far from the strikes you've sold. April was another great reminder—there's nothing wrong with selling 20 delta strangles; the only problem is that retail traders tend to do it in size, blinded by an illusion of safety and the desire to maximize profit. Then volatility spikes, and we are margin called.

As a consequence, we never make any recommendations other than selling straddles. We also don't spend time identifying which strike is the most expensive from a Vanna/charm perspective. It's a nice optimization but won't beat a good trade. As a retail trader, finding a good trade through VRP analysis at-the-money will get you there in 95% of cases.

This is all Sharpe Two is about - finding good trades and the simplest way to implement them.

Why do you trade only ETFs? Do you ever trade stocks?

We trade a wide range of ETFs across many geographies and asset classes. This always gives us something to look at, regardless of the current market focus. However, we do not trade individual stocks.

In an ETF, the market inherently prices a higher premium for the basket over the sum of all the individual components. In practical terms, if you were to add the volatility of all SPY constituents, you would find 14.5, but the actual volatility of SPY is 16. We won't discuss why here, as this will be the subject of another article.

So, it is already easier to harvest volatility on an index than on an individual stock. It is also less risky—the basket component tends to aggregate and diminish the volatility you could see on a single name. Why would you want to get exposed to unexpected corporate actions or earnings announcements?

As a side note, let us insist on this: We are not saying there is no money to be made trading volatility in individual names. In fact, we know many successful retail traders doing just that. It is just not something we are focusing on.

Why do you show only three years of data?

This question inspired this article after someone read the analysis on the overnight trade.

We rarely review more than 3 to 5 years of data because we are now in a completely different regime. We can't say, on the one hand, that rates are now up, the stock/bond portfolio is broken, and 0DTEs have completely changed the options landscape, yet we keep looking at data from 10 years ago.

It's the same market but a different modus operandi—a nontrivial nuance. We understand that people would like to see how a specific strategy would perform over extreme periods of stress like COVID-19, but we like to challenge that view: what is the added value?

COVID-19 was a very specific moment in time, and if things were as crazy again, there would most likely be no strategy backed by normal market conditions (like the variance risk premium) that would be applied in these times. Instead, we would do something completely different. So what's the point?

Let's take an example: had Israel bombed Tehran over the past three weeks, who in their right mind would have casually kept putting VRP trades on?

Except for giving an "illusion" of security (the biggest loss is the loss you haven't seen yet, anyway), a long-period backtest doesn't add much value other than demonstrating the presence of market inefficiency over time.

For instance, you can observe the variance risk premium over 20 years with blips in the curve here and there. The same goes for the equity risk premium—stocks go up with occasional dips.

Do we need more than that to understand that when VRP is there, this is a great trade?

However, when trying to determine whether the market is giving us good odds, we are not convinced that a long-dated backtest is helpful.

We also dislike them because they encourage overfitting. It is tempting to find a solution or setup that would have "survived" COVID-19 or the regional bank crisis.

However, these events are already part of history. It's best to forget about them and accept that something equally bad will eventually arrive with different market effects and consequences that your backtest can't apprehend.

At the end of the day, trading is the "art" or "science" of dealing with incomplete information, and a backtest traps us into thinking we have complete information. The quicker we accept that we are blind, the faster we can be successful retail traders.

We are not saying this approach is the best. However, we think it's working for what we are trying to do. More importantly, it forces us to think hard about a cause rather than falling into a data mine, thinking we found gold when it was coal.

Be sure to follow us on Twitter @Sharpe__Two for more of our insights. If our work resonates with you, don't hesitate to share it with others who might find it helpful.

---

Data, charts, and analysis are powered by Thetadata and Dataiku DSS.

Have access to our indicators using our API.

Book a consultancy call to talk about the market with us.

Contact at info@sharpetwo.com.

Disclaimer: The information provided is solely informational and should not be considered financial advice. Before selling straddles, be aware that you risk the total loss of your investment. Our services might not be appropriate for every investor. We strongly recommend consulting with an independent financial advisor if you're uncertain about an investment's suitability.