Play it Again Sam
Vol #157: December 7th, 2023
Memory is poignant, soothing, and uniquely personal.
It’s the caricature that wakes you up in the middle of the night to relive the time you forgot to zip your fly at school, followed by the vanquishing thought of bright copper kettles. The song that reminds you of her.
As we go further back in history, the rosiness grows proportionally. Zits melt away after months, and halos appear after decades. How you remember that moment might even be impacted by your current situation. (Here’s looking at you, GARCH).
As a kid I thought the coolest story my grandfather would tell was about meeting Walt Disney in an elevator at ABC. I still think that’s pretty cool. As far as wisdom goes, a nugget I kept from him was the idea of story archetypes. I don’t remember when or why it was introduced, but I have a vivid memory of Bob explaining this lens.
It’s a lesson cribbed from Kurt Vonnegut, who posited that there are but 8 shapes of stories. “Boy Meets Girl” is every rom-com where Tom Hanks and/or Meg Ryan bumbles in then out of love before arriving at happily ever after. “Man in Hole” is the protagonist who finds her life disrupted only to vanquish the demons and emerge stronger.
As with most rubrics, once you see the pigeonholes, it’s easy to slot the plots into their spot. Most television series these days fit into the category of “Which Way is Up”, a narrative style that constantly introduces ambiguous situations with all the gray morality of real life. It makes it easy to keep writing season after season, so long as you don’t jump the shark.
Looking at the markets, we tell ourselves many kinds of stories to understand how things should work now and more importantly how they will work in the future. Tribal war paint is donned as investors separate themselves into camps of value or growth believers. Technical analysts perform their rain dances, and in options land the dealer gamma trackers worship at the altar of 0 DTEs.
Each of these value systems looks at the same corpus of information, selects what they believe to be important, and comes to distinctly different conclusions. We focus on what we know well, though we’re at risk of finding facts to support the theory as often as the other way around. Either way, it makes a market.
This isn’t all bad. The 200 year old logic of David Ricardo says we should specialize on our comparative advantage to elevate the outcomes for all. But even further than that, there’s only one person that needs to sleep at night with your portfolio, so you get to choose the bedtime story.
Data can be manipulated for valuable insights, but also bent until it screams and lies to you. Nowhere is this more observable than backtesting, the ultimate re-writing of history in an effort to predict the future.
It gets easier and easier to compile a backtester these days with the help of ChatGPT. My first hack at this five years ago took me a few weeks to build that vicious wheel, while this time I built a better and more robust tester in back to back half days. This availability makes the risks even more acute. Something like an accidental look-ahead bias, where you make a current decision based on information that would have only been known ex-post, will quickly pollute your suspected alpha.
The motivation this time was to understand recent earnings behavior. We’re almost at the end of the third quarter reporting, and the resultant moves in stocks have been larger than usual. Options volumes rhythmically explode around earnings time - often 2 or 3x, but sometimes 10x - because there is such a concentrated blob of variance and opportunity. If you want a nickel to become a dollar, what better time than right before a big known event?
Markets are relentless in their pursuit of efficiency. Institutional vol sellers have eviscerated the variance risk premium over the past decade, and tail buyers have found the right amount of convexity to own. The forces are well enough balanced that implied vol over any medium duration has become very well priced. The 2022 bear market was unique in that long puts or long VIX calls didn’t work so well because you were paying for exactly what was realized - great for the academy, not so much the naive portfolio hedger.
Since the mid 2000s the idea of an earnings risk premium has been elevated. The trade is betting that short term implied volatility over-prices the risk of a big move, as investors bid up the cost of protection. Sell the prem-o, collect the dream-o. Vol gets annihilated after the announcement, and your short straddle or long butterfly collapses into value. That’s a reasonable narrative, and it fits the profile of someone who likes to fade the news. The magazine cover indicator is as much a personality as a trading strategy.
There’s an equally valid perspective that we want to own the gamma in this situation, as being long tails costs a bit in the short term; but extreme moves, black swans, and tail events happen more often than we expect. This opposing camp of Taleb devotees will find masochistic comfort in their decaying theta.
What’s funny about either of these stories, is that they’re both correct and valid approaches for options markets trading strategies. So long as it’s not financial advice, I can endorse either one of them.
The most important thing here, is that regardless of which strategy you choose, it aligns with how you think about the markets. (And risk management, always risk management.) This will ensure that you can maintain the discipline to execute it over the inevitable highs and lows that will occur.
I’ve seen stats that suggest that recently the earnings premium has been inverted, and on average long straddles into earnings net about 4% per cycle. That means if you buy the straddle for $10, it’s worth about $10.40 at the end of earnings. Not bad for a quarter's work.
My own tests roughly confirm this, and you can temper the returns by turning up or down the risk budget. Putting 80% of your capital on the line each quarter (spread amongst the top 50 most liquid names) nets you close to 12% with a max drawdown of 16%. Console yourself through those drawdowns on the bosom of long convexity.
If you want to believe a different narrative about the markets, that earnings vol is overpriced and gets crushed, there’s still a positive expectancy strategy. With long straddles, even if they’re expensive, you know what your worst case scenario is. Short straddles have an unbounded component, which is why we like to use defined risk butterflies to get short volatility around earnings.
Buying in the wings, which protects your ATM shorts, has a wet blanket effect on most butterfly strategies. Setting the wings across a wind range of distances has recently and consistently ground out a small loss. Again, +1 for the academy, markets are roughly efficient. You can’t have your short vol cake and eat protection too.
Where I think this strategy gets fun is turning the hockey stick upside down, and being “long” the hole in one. With a butterfly, the maximum value happens at the exact center strike. The long call spread is worth its maximum and the short call spread is worth the minimum - both spreads happen to have the same strike as the short leg.
Around earnings, this is very unlikely to happen - which makes the tight ATM butterfly very cheap. We don’t know which direction it’s going to move, but it’s almost certainly going to move beyond this very tight range. Except Walmart in May of 2023 when the pre and post earnings prices were within 4 pennies of each other, exactly at the $150 strike. That’s how pennies explode in value, even if you have to give away >10% of the maximum value to close before the end of the day.
We’re still “selling” the notion of a big move, but we do this by throwing thousands of pennies out there and cashing in on the whopper. Most of the time the strategy treads water. It sees small draw downs of only 6%, but has a bunch of triples, a handful of home runs, and the above: a bottom of the 9th down by 3 Grand Slam. If you want to really sell the news, it’s best to set up a portfolio that is positively exposed to the extremes of nothing happening.
The story that earnings are surprises or that they’re non-events can both be true over time. Different premia are captured and managed in ways that are suited to the appetite of the investor. They’re not winners at exactly the same time, but that’s also what makes them a valuable and orthogonal portfolio component. If you could tell when THAT was going to happen, please give me a call.
Our capital allocations are inevitably a reflection of ourselves and the stories we want to paint about the future writ broadly, and personally. Trading for today or investing for the long term requires the support of well aligned beliefs. We all want to make money, but we also want a good story to tell, and that’s a very personal question.