Option on Levered ETFs (Part 1)

“They” say human labor will be irrelevant by 2027. By then, any business you can think will be solved by capital (electricity and tokens) before you brush your teeth in the morning.

You either get rich in the next year or join the permanent underclass.

So we aren’t shocked that the hottest fads in investing is pure return fuel:

double, triple, even 5x levered ETFs
options (record volumes, with nearly 2/3 of SPX options in 0DTEs!)

You’re not gonna break outta that underclass clipping 10% per year amirite?

I’m not here to judge whatever you think you need to do. The Moontower SOP, as always, is:

It's Dangerous to Go Alone - by Andrew R. Jones

I give you tools, you use them as you see fit.

We need to talk about the collision of these trends:

options on levered ETFs

We can hack together a pricing framework. I say “hack” because we use a few building block concepts, combinations of arithmetic and logic, to construct a model. The process fuses several ideas I present often, reinforcing your comprehension of the basics, while alerting you to common misunderstandings.

We’ll do this in 2 parts.

Today, we cover four essential concepts:

1. Distance in return space

2. “Vol bonus”

3. Derivatives on derivatives

4. Option vega and convexity

Next week, we arrange the basics into a model for finding a fair price for an option on a levered ETF.

Essential Concepts

1) Distance

A $100 stock that goes up 10% 2 consecutive days is now $121.
A $100 stock that goes down 10% 2 consecutive days is $81.

In return space, $21 on the upside is the “equivalent distance” to $19 on the downside.

2) Vol “bonus”

I’ve written as much as can be written about vol drag/volatility “tax”/volatility “drain”, whatever you want to call it. It’s the “chop”. Up 10%, down 10% yields a cumulative return of -1%.

Trending leads to a vol “bonus”.

Imagine a 2x levered ETF on the $100 stock from earlier. Let’s say the ETF also starts at $100.

◾Reference stock goes up 10% 2 days in a row..

$100 —> $110 —> $121 or a 21% cumulative return

🟩2x levered ETF goes from $100 —> $120 —> $144 or a return of 44%

If you sized your ETF position half as large to have the equivalent risk or beta exposure you will have outperformed an equivalent risk holding over 2 days, although indifferent after just 1 day.

The vol bonus comes from trending.

If you step through a random walk where a stock can either go up or down X% you find that in most paths, the ones where we chop or recombine frequently (the paths fenced in red), your compounded return is less than if you simply added up all the daily returns and applied them to $100. Be careful: volatility doesn’t change your expectancy, only the distribution of return.

The infrequent, large vol bonuses offset the frequent vol taxes to keep the expectancy the same, even though you “usually” suffer a vol tax.

3) Derivatives on derivatives

A levered ETF is a derivative with a fair value. At any given moment, its NAV should be equal to:

NAV at the start of the day * (1 + reference asset’s return * leverage factor).

This is approximate since you should also deduct one day’s worth of expense ratio and the fund’s trading costs divided by the share count, however these values will have negligible impacts on daily fair value calcs for most users. The point is that a levered ETF is a derivative just like all ETF values are derived from an underlying basket, future, or security.

We want to price options on the levered ETFs which means we need a volatility surface. The reference asset’s option chain will provide a consensus vol surface. We pull a page straight from the arbitrage trader’s handbook — use a liquid market to price a closely related market after adjusting for the differences. This will provide a fair value of the levered ETFs options relative to the reference asset’s option prices.

4) Option values, vega, and curvature

We will cover one last “basic” before trying to reason our way towards a model for pricing options on levered ETFs.

Consider an at-the-money option.

💡Technically I should say at-the-forward but we can assume a RFR of 0% in which case ATM = ATF

If we double the vol, what happens to the option price?

It doubles.

The approximation for an ATF call:

ATF call ~ .4 * S * σ * √T

where:

S = stock price
σ = implied volatility annualized
T = fraction of a year until expiry

If c doubles, the call doubles. It has a linear dependence on vol.

💡Visual derivation of the ATF option approximation

Look at the approximation again.

What’s the vega of the ATF call?

Before you go searching for a Black-Scholes calculator, just recall that vega is the change in option value for a 1 point change in vol.

If σ increases by 1 point, the ATF call increases by .4 * S * √T

If we are pricing a 1-year ATF call on a $100 stock, if vol increases by 1 point, the ATF call goes up by $.40

Therefore, by definition:

ATF vega ~ .4 * S * √T

💡Example: Consider a 1-year ATF call on a $100 stock. If vol increases by 1 point, the ATF call goes up by $.40

Notice that the vega itself has no dependence on the volatility.

This is only true for ATF options!

Out-of-the-money option vega DOES depend on the vol level. It’s not hard to understand this intuitively.

Consider the 60 strike put on a 16% vol, $100 stock expiring in 3 months. 16 vol is roughly SP500 vol. The 3-month, 40% OTM put is probably worthless.

💡16% is a 1-year standard dev. 16%/√4 = 8% quarterly standard deviation. That put is 5-standard deviations OTM. Save me the Taleb-stanning that option is worthless.

A worthless option has no vega. If I raise the vol from 16% to 17% that put is still worthless. It has no sensitivity to the vol.

But that can’t be true for all levels of vol. If we 10x the vol to 160% then that put is now only 1/2 standard deviation OTM. It’s most definitely NOT worthless.

💡Re-framing volatility as time: Increasing the vol by a factor of 10 is algebraically equivalent to increasing time-to-expiry by √100. Even if we stick with 16% vol, a 40% OTM put on the SP500 with 100 years til expiry is clearly valuable

So we know that the 60 put acted like it had 0 vega when vol went from 16% to 17% but when vol 10x we expect the option to have a non-negligible valuation. Somewhere along the line, this option “picked up vega” or sensitivity to volatility.

While the ATF option is a purely linear dependence on implied vol, the OTM option’s dependence ranged from 0 to some positive number. This is the source of vol convexity.

💡See Finding Vol Convexity

A far OTM option might have no vega. But as implied vol increases, that option’s strike “becomes closer” to the ATM strike. I mean if vol is 200% there’s really very little difference between the 70 strike and the 60 strike in terms of standard deviation.

As vol increases, OTM option vega increases. If you keep jacking up the IV, the vega eventually peaks. The maximum vega an option can have is the ATF vega, which has no dependence on vol level at all.

I computed European-style option values using Black-Scholes for a range of strikes for 3-month options on a $100 stock using 20% vol at each strike.

Then I doubled the vol to 40% on each strike and computed the difference in option values:

The ATM options have the most vega and will be the most sensitive to vol so naturally they go up the most.

But remember, at 20% vol some of those OTM options would be close to worthless but when vol doubles, an option like the 80 put went from $.04 to $1.16! On the chart, it gained $1.12.

If vol doubles instantly, your ROI on an ATM option is 100% but doubling the vol on an OTM option leads to comical ROIs:

At the extremes, you get divide by zero errors as the return is infinite on a previously worthless option.

💡Clarification: I’m measuring changes in the OTM option on each strike. In other words, the “extrinsic” option. So for the 90 strike we are using the put, but the 110 strike, the call. The price changes will be the same for the call or put on the same strike, but the ROI due to a vol change shouldn’t be muddied by the instrinsic value which is constant on our examples, so we use the OTM option.

We’ve covered the essentials, next week we’ll answer:

How do we translate the reference asset’s vol surface to the levered ETF?
What does this mean for strike selection?

“trader” is a uselessly broad term

Moontower.ai Announcement

The only sale moontower.ai has per year is underway. A 20% discount applied to all new plans as well as extensions for current subs.

Starter plans get access to the Discord, while Pro and Enterprise plans include Discord as well as free subs to this substack. Sign up here 🚀

Friends,

Erik had Euan Sinclair on the podcast last week. Euan is a straight shooter, so his interviews are not just info-dense but fun to listen to.

🙈Shameless testimonial

Back on April 10th, during the Liberation Day market chaos, Euan texted me:

The [moontower.ai] app has been super helpful. when this sort of thing happens, having a top down tool is the only way to keep up.

He’s since told his own community (emphasis mine):

Kris A looks at vol markets very much in the way a professional vol trader does. You really need to be comfortable thinking in IV terms rather than price terms, but once you get there, this stuff is gold. I really think you could use only this site and trade successfully as a professional. Kris is a good guy, very helpful and again, totally legit. As you know, there aren’t many resources like that.

Just saying bruh. The only sale of the year is going on now (20% off) and applies to Starter, Pro, and Enterprise plans.

Get 20% off your moontower.ai sub

Anyway, back to Euan’s recent knowledge-drop.

🎙️Euan Sinclair on Retail vs Institutional Trading | The Outlier Podcast (YouTube)

I list many of the key moments from the interview later in this email but there’s one in particular I want to zoom in on:

🔎Institutional vs. Retail Trading – Fundamentally Different Jobs

Euan on the track record misconception:

If you’re working at a trading firm you don’t get an account like that… you don’t know how much risk capital is being allocated to it. So when someone’s like what’s your track record the answer is really ah I see you don’t understand what professional institutional trading is.

Kris: I’ve written about this more extensively here from the perspective of an employee, running a market-making biz, and as a HF PM in When it’s normal to have no idea what your returns are (13 min read)

Euan on the value of a seat:

There’s a lot of money they get given because they’re at the institution, not because of anything they do. The seat and the franchise has a lot of value. Whereas a retail guy has none of that.

Even if he acknowledges I’m making all this money because I’m making markets to customers and the markets are wider, that’s not you. That’s Goldman.

For a lot of them they wouldn’t have the first clue where to start because they’re a piece of a team, right? They might be at the end of a chain of sales groups where they have to make a market, but they’re making a market to a structured desk, which is then decomposing it, talking to a sales guy, pitching it to the customer, and then it comes back to him.

You take a professional out of his seat, say there’s a million bucks, sit in front of your IB screen, make some money. Most of them can’t do that either. It’s just it’s a very, very different job.

Kris: There are levels to this as well. I got my fund job after pitching Parallax on how I’d build out a cross-asset commodity options trading business (you can see the seeds of moontower.ai in the inspiration for what that plan encompassed). At the fund, we had no sales desk. It was basically, here’s a phone, a computer, access to capital, now run along and make more capital.

I had to give brokers a reason to call me even though I wasn’t going to be in the habit of crossing bid-ask spreads. I had no insight into physical markets or fundamentals that could make crossing the spread consistently worthwhile. It was just a matter of composing the data the way I wanted to see it (“measurement”) and combining those cards with the players’ actions (the orders in the broker and electronic markets) to construct option portfolios that had value.

When it came time to hiring, this concept was clearly foreign to some candidates (usually coming from banks with sales desk and captive flow) OR the candidates understood this would be the nature of a buy-side options job but since they were never actually forced to turn water into wine had only naive conceptions of how you do this. “You just sell vol when it’s higher than realized, right?”. Sir, we run vega-neutral books, shoot for absolute returns, and expect to do really well when this stupid bull market hits a hiccup…you’re gonna need to try harder than that.

But now as a retail trader, I don’t have the scaled cost-structure and I don’t see any voice flow so I’m missing the bet/call/fold actions that I was accustomed to. The retail trader’s advantages are liquidity is easier to get and they aren’t forced to trade anything they don’t want to. The first advantage is real but is the first victim of success and the second is a consolation prize for not mattering.

It’s harder to be a retail trader. But that juxtaposition is not quite right. The retail vs institutional trader distinction is apples and oranges. Retail trading is more like being an entrepreneur. That’s a better comparison.

I say more on that in this thread.

☂️Other ways the term “trader” is asked to do too much

I’m not even addressing all the ways “trader” is used. Some brokers call themselves traders. Execution traders take orders from PMs but are still called traders. These roles all certainly overlap with the job descriptions of traders who act on their own ideas but the broad connotation of trader is closer to “bettor” and in that sense the customer of the broker and the source of the thesis.

About 20 years ago, after my 3-year non-compete expired, I interviewed with a large multi-strat to be an excecution trader. The job paid 2x what I was making.

[It was the heydey of HFs getting way overpaid for beta-like risk/reward. Lot of fat on the bone as they were not moneyballing their talent the same way prop shops were accustomed to where regardless of your 1-year performance you are only worth some spread to replacement level trader. You don’t have an unbounded call option on a great year because they will tell you your seat was worth more. The goalposts move in a way that keeps your comp at a level which balances “I won’t flip the desk over in my bonus meeting” with “this is probably more money than I’d get if looked elsewhere”. A true eat-what-you-kill deal fixes this but they likely have much tighter risk limits.]

Anyway, I’m interwiewing for this execution trader job on a small team. I think it goes great. I feel like I show well.

Nada.

I backchannel with a friend who worked there. “What they say about me?”

“You’d be bored.”

Meanwhile I’m like, yea, so what? Being bored at a job paying a million bucks per year sounds like a Me problem, not your problem. I’ll take my chances.

Anyway, there’s lots of different kinds of traders.

Fwiw, I think the types of traders best embodied by the word trader are the sociopaths featured in World For Sale.

[Random personal note…during that job interview, I was left alone in the conference room for a bit. My dad called. This was either Star-tac flip or Nokia phone days, I can’t quite remember. I picked up to discover my grandmother had just passed out of the blue. I had to do my own sociopath act to get back to the interview without letting on. Note to self, don’t pick up a phone call during an interview.]

More insights from the interview…

Hedging

If you go to a firm like a Jane Street or a SIG, their idea of hedging isn’t so they don’t lose money. Their idea of hedging is so they take away all the risks that they don’t want so they can isolate the exposure to the one thing they do want.

Transaction Costs

With trading, you become a better trader by taking your hands off the keyboard and letting things play out because every time you touch the keyboard, it costs you money… You might make 5,000, but whether you make 5,000 or lose 5,000, you’ve got that 50 bucks always. And it adds up.

Trading Psychology – Act Like a Professional

If you pull that shit as a professional trader, you lose your job… Instead of reading all this nonsense about trading psychology, it’s just act like a goddamn adult. Come to work, do your job, don’t be a [expletive] and then go home.

Risk Management – Small Losses vs. Catastrophic Risk

Retail is too averse to small risks and leave themselves open to catastrophic risks… [Professionals] don’t think of losing a million dollars in a day as a risk. That’s just what happens if you’re tossing a coin.

The Sequence of Returns Problem

People will do a very careful analysis. They’ll go over five years of data, you know, thousands of data points, come up with a strategy, and then they’ll trade it for two weeks and go [expletive]. But if I told you, I’ve got a trading strategy and I’ve tested it on two weeks of data, you’ll be like, what the [expletive].

Position Sizing and the Yield Hunting Trap

People decide they want to make a certain amount. You can’t make a certain amount. You can only make what the market’s giving you… If you go basically yield hunting and try to make the same amount of money each time, you are going to blow up. I guarantee it.

Price Matters More Than Structure

Everyone thinks their edge is in the structure. So they become price-insensitive around that structure [he picks on iron-condors]… There’s always a price where you should buy stuff too.

Zero DTE Warning

Don’t trade zero DTE… The amount of money you can make is pretty small because the premiums are pretty low and the costs relative to the amount of money you can make are huge.

Volatility Rank vs. Variance Premium

If you’re just looking at the vol of a single stock through time, selling it when its vol gets high is usually the wrong way round… Usually implied vol is understating, you know, implied vol might do this when the realized V is doing that.

[moontower.ai users understand this in their bones because it’s embedded in the calibration of fair value]

Event Premium – Earnings and Announcements

Selling options right before earnings is a good trade. Selling biotech options right before FDA announcements is the same thing… Whenever there’s an event and you don’t know what the outcome’s going to be, but you know when the event is, sell V.

Biotech Sizing – Accept Individual Losses

You have to accept you’re going to get a 10SD loss. You can’t expect to do that risk-reward thing on one stock… You have to do that over a whole bunch of stocks and say on average, if I do this on a 100 stocks, I’m not going to get a 10s.

the arbitrage reflex is more profitable than the opinion reflex

A traditional way to think of a stock price is the expected value of its future prices weighted by their probability and discounted to the present. Ignoring the cost of money, in a binary world a $100 stock could be fairly priced if it was 50/50 to be worth $200 or $0. It is also fairly priced if there’s a 20% chance of it going to $500 and an 80% chance of $0.

There’s this vocal VC named Keith Rabois who aggressively cheerleads his companies. I don’t know the guy. His online persona exudes many standard deviations of F-U confidence. Sounds par for the uber-rich these days, but he’s extra fun because he’s pugnacious. And got baited into a silly pride bet.

Here’s a tweet from investor @compound248:

We wouldn’t talk about this in the Masochists section because this is fairly basic financial reasoning. The type that really needs to obvious to everyone in a society is flirting with a simulcrum of the movie Rat Race. But it’s appropriate to spell out the opportunity here in gambling terms.

Keith is offering an even-money bet, his $100k to Compound248’s $100k on the stock multiplying by 31.5

If you think in odds:

Keith is offering even money on a 30.5-1 odds proposition

That might be more clear when you think of Keith buying the stock. If he buys $OPEN he risks losing the stock price or 1 bet and if the stock goes up 31.5x he wins 30.5 (because the 1 bet or amount of cash he spent for the stock is not part of his win or return).

It’s similar to how a stock 2x’ing is 100% return, or 10x’ing is a 900% return. A stock that 10x’s paid 9-1. You risked S, you won 9S if S is the stock price.

Normally when you buy a stock, you get paid dollar for dollar as it moves times the number of shares you have. If the stock doubles you make S in profits which is how much you risked when you bought it. You are paid in proportion to the move.

Keith needs a heroic move to simply get paid even money. His proposition is an arbitrageable violation of how return works. I don’t know anything about Compound248’s outlook on $OPEN by him taking the bet. He could be bullish or bearish. When you hear the proposition, your mind shouldn’t go to “Is $OPEN a good or bad investment?” because what you should do doesn’t depend on this assessment.

Keith’s offer is free money regardless of your outlook.

He’s laying 30.5-1 odds where the max loss is $100k.

So you solve for “How much OPEN do I need to buy to make a $100k profit if it pays 30.5-1?”

It’s simply:

1/odds * bet size

1/30.5 * $100k = $3,280.21

I need to buy $3,280.21 worth of shares. Since the stock was $8.48 that’s just about 387 shares.

If the stock goes to 0, you lose the $3,280.21, but Keith hopefully pays $100k. If the stock does go up 31.5x, you break even.

You could also structure the hedge so that at a stock price of 0, you break even. You buy $100,000/$8.48 or about 11,792 shares. If the stock hits Keith’s bogey you get paid 30.5 on your $100k and you happily peel off 1 bet size to him as a tip. Any share quantity you buy between 387 and 11,792 is a guaranteed win.

An amusing post-script to this story is HF manager Benn Eifert requesting $10mm of action on this proposition. Of course, Keith said no — he’s confident not stupid. Keith said he did the bet with Compound248 just to shut the “troll” up or something.

I don’t understand how rewarding a troll with the easiest money I’ve ever seen is anything but encouraging future trolls, but maybe this is why Keith is rich and I’m writing on the weekend.

Moontower #293

Friends,

The largest firms in the world’s capex doesn’t lie. They’re all-in on AGI. Since the so-called hyperscalers are also a historically disproportionate share of the SPUs, they are dragging the entire economy into that bet.

The “dog with a mouthful of bumper” outcome is they are right and actually eat the whole economy, only to be murdered by their own bunker guards. The irony of using Sama’s ChatGPT to query this:

OpenAI’s fundraising appetite and valuation requires AGI to be a thing, whereas a company like META will survive as a going concern and may even get stronger despite the capex blowing a big hole in their balance sheet temporarily if AGI is not a thing. Yann LeCun, META’s chief scientist and Turing award recipient, is leaving the company. He’s critical of the “scaling hypothesis” which claims that if you just keep throwing compute and data at the models, they will become “intelligent”. He doesn’t see how causal reasoning would ever emerge from pattern-matching even if it’s the most superior pattern-matching the world has ever seen.

This debate feels like mommy and daddy are fighting while the family’s future hangs in the balance. The naughtier kids are using the distraction to raid the cookie jar, while the market’s spirits are a thermometer on who the rest of the kids want to side with.

In a Carlota Perez framework, the prospect of AGI would suggest we are still early in the installation phase, but if the common knowledge surrenders to” AGI is never spawning from the current soil” then we are probably already in the deployment phase but not every middle-manager has taken Prompting 101 yet. In the first case, the calls aren’t hot enough, and in the second, you want your stocking stuffed with puts. Either way, we’ll take the electricity. But Santa, gas not coal, please.

Below the sci-fi (is that science fiction or science finance?) theatrics, we have the mundane business of actually using these things. Last week I shared the listicle voting app thing with some resources for those who want to make stuff with LLMs. This week Khe had a detailed post describing a CRM he built:

Vibe coding? More like “whack-a-mole” coding: The uncomfortable truth around my Pebblr CRM app (link)

Here’s another fantastic article that I think captures nuance between AGI breathlessness and very real capabilities of LLMs today.

Is Vibe Coding the Future of Skilled Work? (Scott H Young)

It’s written from a relatable perspective since I’m using these things so much, while easily admitting they are enablers rather than substitutes. And isn’t that a sweet spot? Like you can do more but still matter? Well for now I guess. We’ll see what happens. If the puts pay off maybe that’s bullish for humanity.

I’ll share some vibe-coding projects here and there.

Here’s my Car Lease Embedded Option Calculator.

Some background for the uninitiated:

Money Angle

Here’s a tweet from investor @compound248:

Keith is offering an even-money bet, his $100k to Compound248’s $100k on the stock multiplying by 31.5

If you think in odds:

Keith is offering even money on a 30.5-1 odds proposition

It’s similar to how a stock 2x’ing is 100% return, or 10x’ing is a 900% return. A stock that 10x’s paid 9-1. You risked S, you won 9S if S is the stock price.

Keith’s offer is free money regardless of your outlook.

He’s laying 30.5-1 odds where the max loss is $100k.

So you solve for “How much OPEN do I need to buy to make a $100k profit if it pays 30.5-1?”

It’s simply:

1/odds * bet size

1/30.5 * $100k = $3,280.21

I need to buy $3,280.21 worth of shares. Since the stock was $8.48 that’s just about 387 shares.

If the stock goes to 0, you lose the $3,280.21, but Keith hopefully pays $100k. If the stock does go up 31.5x, you break even.

I don’t understand how rewarding a troll with the easiest money I’ve ever seen is anything but encouraging future trolls, but maybe this is why Keith is rich and I’m writing on the weekend.

Money Angle For Masochists

Tomorrow, we are going to launch the annual Black Friday/Cyber Monday discount for moontower.ai. It’s the only sale we offer during the year (current subscribers will be able to extend at the discounted price as well).

The app’s selling point, what makes it different, is its “point of view” on vol. It really starts from “what’s normal in the options market now” vs what sticks out. The analytics are geared towards answering that question because that’s how you find contradiction.

If sunscreen and umbrellas are simultaneously expensive, it might be because there’s a sunshower expected — but do the odds make sense? Before you could even reason about that you needed to know that sunscreen and umbrellas were both expensive in the first place, otherwise you wouldn’t even consider the question. Questions are where opportunities live.

The price of options is summarized by properties of vol surfaces, which in turn, can be compared to each other. We do the measuring and comparing to point out where the questions are. Options are not as simple as point spreads. Strikes themselves are fixed but the stock price changes, time passes. That same contract’s properties morph like natural landscape seen through a time-lapse camera. We are your guide in this wilderness.

We surface what types of trades look relatively attractive from the vol trader’s vantage point. You can think of that as a solid hypothesis based on the data, but from there, you can adjust based on your knowledge or opinion of what is going on in the name.

We don’t make guesses about the future. We don’t say do X and you will make money. It’s obvious to anyone who has ever taken risk that handicapping the future is not enough to make money. Think about it. How confident are you that the current price of SPY is not the all-time high? Probably 100% and rightfully so. The “SPY to be up 1 cent” one-touch option would be priced at 99.99999% percent. The knowledge is replacement-level not value-add.

You don’t sign up for Bloomberg because it tells you how to make money. You sign up because it helps you see*. And seeing in the correct terms is a prerequisite to profitable decisions. It’s the base of the pyramid upon which you layer the rest of your process.

That’s what we’re solving for in the options niche. The vol trader’s lens.

I made these vids this past week to offer a concise description of some of the key tools for seeing like an option trader:

*When you sign up for analytics, you usually do so knowing what you want to see. But also, there’s a built-in education. You learn what matters to others as well. Digital real estate is not scarce but deciding how to fill it is still constrained by taste and demand. The option pricing software that I used in market-making was also full of clues about best practices because professional users drove the features. What’s interesting when you have an analytics product that serves both retail and enterprise is the features can be a weaker signal about what matters. B-to-B vs B-to-C.

Stay groovy

☮️

Moontower Weekly Recap

Posts:

the art of paranoia

This is a fun one.

A good friend and mentor from the pit sent me this (lightly edited and hyperlinked):

I was catching up on moontowers and reading your cotton story. Made me think of one that you can use if you ever cover interest rates.

I started in trading with silver options. I was quite meek as I had never clerked. I was just backed and thrown in the pit at age 22. Anyway, a couple of years in, silver had rallied from around $4 to $7. The whole pit was short long dated $4.50 calls to a spec who was holding the long. These things were easily exercisable as interest rates were high. Great short to have.

Broker [badge redacted] (you may remember him) offered the synthetic put at zero. No one knew what he was doing. I bid 2 ticks under and he said sold.

So I said “1000”. I think the biggest trade in that pit was like 200.

Anyway I took 1000, exercised them and made 10 grand before commissions. The carry was probably north of 50 ticks and he increased his shorts by around 800 as all the other locals got hit with my exercises. The other locals were pissed.

So I came in through a broker and bid 2 ticks under the next day. Locals hit me. I got lucky because I was clearing [redacted prime broker] and they neglected to put in my exercise. They delayed it by mistake (and gave me the interest as I recall.)

Anyway the locals thought they found someone who would hold these things and just wanted a synthetic put for a credit. I kept doing the trade for a week.

[Redacted broker] got it right by the second day. I think I made another 5 or 10 grand before the locals figured it out and stopped doing it.

Anyway. I hope if you relay this no former silver option traders subscribe. They still don’t know it was me!

There’s a lot going on here!

Option pricing mechanics, pit dynamics, deduction.

Let’s start with the option pricing.

I don’t want to deny you the opportunity to figure that part out for yourself.

In the following list, I’ll start with important clarifications but as the list unfolds the material is more of a hint than just reference information.

Like a quiz show, buzz when you understand why “locals” [ie the other traders] were willing to” sell 2 cents under” and why my friend was willing to buy that level.

Clarification and hints…

1 silver option references 1 silver future. This is typical in commodities whereas equity traders are used to an option having a multiplier of 100
1 silver future references 5,000 troy ounces of silver. So if silver is $10 an oz, it’s a $50,000 contract
The minimum increment, or “tick”, in silver options is .001 or 1/10th of penny. Since it references a 5,000 oz contract, a penny is worth $50 and a tick is worth $5.
This story happened many years ago. Look at those silver prices: “Silver had rallied from around $4 to $7. The whole pit was short long dated $4.50 calls to a spec who was holding the long.”
Silver options are American-style meaning you can exercise them anytime.
“Random assignment”: when an option is exercised, any contract in that series held short is equally likely to be assigned regardless of the clearing firm or account. See rules.
When a broker quotes a “synthetic put” (and yes synthetics are not just identities but directly traded orders!) the convention is to bid or offer as a “differential to intrinsic value”. When my friend bid “2 ticks under”, it’s understood that he’s willing to pay 2 ticks under “intrinsic value”. If the futures are $7 and the broker sells the $4.50 call 2 ticks under than the trade package is:
- broker sells the $4.50 call at $2.498 to my friend
- brokers buys the future for $7.00 from my friend
- Make sure you understand this: to buy a synthetic put it to buy call and sell the future “1-to-1” (meaning for every option you buy, you sell 1 future.) Another way to express that is you hedge on a 100 delta. A synthetic put is assumed by everyone to be a simultaneous package of “long call, short future” in the same way as a straddle is call + put on the same strike or strangle is call + put on different strikes. “Synthetic put” is a real tradeable thing not just the name for an option identity.
While futures are subject to initial and maintenance margin, option premiums are settled in full. In other words, the premium is not itself marginable. That is normal but worth stating because there are some option markets where the premium is marginable (ie WTI options on ICE)

At this point, you should be able to understand my friend’s side of the trade. If you don’t want to bother, stick around, I’ll spell it out soon enough but also that part should feel really obvious to anyone that’s ever owned an American option.

The harder question is:

Why were the broker and the other locals willing to sell him the synthetic put 2 ticks under?

Your last hint is a line already tucked into the story:

The whole pit was short long dated $4.50 calls to a spec who was holding the long. These things were easily exercisable as interest rates were high. Great short to have.

This is fun stuff. Let’s unpack it.

Why did my friend buy the synthetic put 2 ticks under?

It’s an American-style option. He buys the $4.50 call for $2.498 and sells the future at $7.00.

He exercises the call immediately, effectively paying $6.998 closing out the future he sold at $7.00. He makes 2 ticks or $10 actual dollars x 1,000 contracts. $10k profit (before exchange fees which probably claimed ~ 25% of that). Call it $7,500 for a minute’s work and no risk.

How could my friend’s buy be so good if he also says:

The whole pit was short long dated $4.50 calls to a spec who was holding the long. These things were easily exercisable as interest rates were high. Great short to have.

The key:

The carry was probably north of 50 ticks

50 ticks = 5 cents

In other words, the interest on $2.50 of intrinsic option premium was 5 cents or about 2%

💡While a market-maker will be hedged with futures, you only need to post margin to maintain that leg. You can satisfy the collateral requirement with T-Bills, so you really are capturing the the interest on the short option premium without having it offset by hedge funding which is close to zero.

💡In the story, there’s 5 cents of carry. If we knew the DTE, we could back out the interest rate prevailing back then. If we knew the interest rate, we could back out the DTE. But we have neither. We just have the recollection that the call had about 50 ticks of carry.

Think of the typical local’s position:

The whole pit was short long dated $4.50 calls to a spec who was holding the long.

The locals are hedged. They are short the deep in-the-money calls and long futures against them on a 100 delta. In other words, they are short synthetic puts. The puts on that strike are worthless however if the spec never exercises the calls, the locals will get about 50 ticks of interest. It’s like selling a worthless put for a nickel.

💡If you’re so Taleb-pilled you can’t imagine a put being worthless, just pretend you could buy the actual $4.50 put for a tick or the $4.75 put for 3 ticks. There’s such a thing as arbitrage bounds. Options 101 stuff.

Amateurs will say things like “that’s worthless” because they believe a price “can’t get there” but when I say it’s worthless I mean by arbitrage. Like the value of the put on that strike is dominated in such a way that if you could sell it at a positive value there’s free money on the board. An option trader thinks in a matrix of arbitrage relationships when they examine a chain.

You can see why the locals were willing to sell the put 2 ticks under. They didn’t think they would get assigned! They assumed they were going to collect 50 ticks on a riskless position.

The mistake was in thinking my friend wouldn’t exercise the calls. Perhaps they thought he needed to cover risk and if they were all short the calls he was just closing. But that doesn’t make sense to me since he was a “meek” trader before then and certainly wouldn’t have been short 1000 calls.

That’s the most puzzling part to me. I understand why they might believe a customer like the spec would not exercise the call optimally, but when a market-maker buys them you need to update. “This guy who trades options all day is buying an American-style option below intrinsic, I wonder what he’s going to do?”

I don’t want to be too snarky because “street smart” is probably one of the most salient features of a pit trader. I’ll assume I’m missing a detail that makes their decision justifiable.

The rude awakening

“The other locals got hit with my exercises. The other locals were pissed.”

This is the random assignment. When my friend exercised his calls the free interest gravy train slowed down as suddenly some of the locals got assigned which closed their positions (the short call goes away and the long future they were hedged with is liquidated to the exerciser).

💡In random assignment, you “expect” to be assigned on your pro-rata portion of the OI held by all shorts. How many you actually get assigned on can vary from this theoretical expectation because the sequential 1 by 1 assignment process is memoryless just like dealing from a deck of cards with replacement each time. Just because you got tagged on the last one doesn’t change your odds of getting tagged on the next one.

Let’s keep unpacking the story.

Poisoning the well

My friend masked his behavior to blend in with customers and used a broker to “bid 2 ticks under” the next day. The locals didn’t think it was a market-maker going through a broker. The locals were reasonable in not suspecting a local to be hiding themselves behind a broker order for 2 reasons:

Paying commission which would eat into the slim margin even further.
This behavior is known as “poisoning the well”. This type of pro vs pro crime was considered bad form. Just like in poker how the pros try to avoid each other and just eat the fish. Of course from the outside, the norm is considered anti-competitive.

Let’s address both of these.

Regarding #1:

My friend may have worked out a deal with the broker. After all, the broker stands to make a thousand bucks or so for no real effort and if the order is contingent on a reduced commission rate they’ll probably go for it.

[I used to do this all the time. Wet the brokers’ beak. We like to talk about nerd stuff and math here, but trading is a business like any other business. Giving out orders is currency. Traders who pay lots of commission magically get the first call. Whether it’s commodities or equities brokers can solicit the other side rather than bring an order to the pit and instead just cross it without worrying about the order being broken up (ie bettered) they can double-bill. Brokers are fiduciaries and their business is competitive, get too many bad fills and you risk losing a customer, but there’s a lot of leeway in discretion. I’m using a broad brush, the rules and details vary across asset classes, but the mechanics rhyme everywhere.]

Regarding #2:

Prisoner’s dilemma.

How much is it worth to defect?

Will we return to stasis and you got away with one or did you really poison the well?

Will anyone find out?

Does any of this change if you don’t really fit in with the club and can’t hope to be part of it?

And then there’s pure disagreeability mixed with self-confidence. I was at Max’s soccer game last weekend and the ref in a moment of deprecating humor said “They told me when I was a kid you need to either be smart or likeable. And I’m definitely not smart”. I don’t think this particular buddy ever cared about going along to get along. But he’s also super-smart.

I’ll add my own perspective to the game theory stuff. I did nothing but watch markets get tighter, more competitive and more ruthless over time. The edge you never would have settled for becomes something you’d kick grandma down the stairs for before you know it. Someone is going to defect.*

This is chained to another observation. In a landscape where there’s excess profit (ie too much reward per unit if risk) there is always someone sandbaggin’. They aren’t showing how smart or fast they are because they will harvest at this level before it tightens and not have to tip off to others what’s possible. I know this firsthand because of a friend that runs a brilliant trading operation in a niche space in which he never shows the market how fast he really is. Imagine my lack of surprise when I read about this “don’t show” tactic from HFT-er Liquidity Goblin’s Let’s Pretend We Have An Edge (paywalled).

A fortunate mistake

I got lucky because I was clearing [redacted prime broker] and they neglected to put in my exercise. They delayed it by mistake (and gave me the interest as I recall.)

Anyway the locals thought they found someone who would hold these things and just wanted a synthetic put for a credit. I kept doing the trade for a week.

[Redacted broker] got it right by the second day. I think I made another 5 or 10 grand before the locals figured it out and stopped doing it.

My friend was able to milk the ruse for a bit longer because of an error! The prime broker failed to process the exercise which reassured the locals briefly that they were not getting picked off. Sounds like it gave him at least another day to buy more synthetic puts for a credit before the locals got wise to the game.

Knowing is half the battle

Let’s put a bow on this like an 80s cartoon ending lesson.

Being short those synthetics to a single counterparty who doesn’t realize the calls should be exercised is good fortune. But once someone wants to buy those deep ITM calls opening, you know that they know what they’re doing.

I’ll tell you from personal experience that anytime someone wants to trade a deep option or a reversal/conversion where there is little or no open interest your guard goes up. You check your funding assumptions. You think harder about what your option model is saying about the value of an American vs European style rev/con. The difference between the 2 represents the value of the early exercise option for that strike. Those are modeled via trees simulations and difficult to debug as opposed to closed form equations for European Black-Scholes.

Only the paranoid survive.

*I refer you to this excerpt from A Former Market Maker’s Perception of PFOF:

SIG wasn’t know as the “evil empire” on the Amex just because of the black jackets we wore. They understood the risk-reward was completely outsized to what it should be 25 years ago. They were amongst the first to tighten markets to steal market share. They accepted slightly worse risk-reward per trade but for way more absolute dollars. They then used the cash to scale more broadly. This allowed them to “get a look on everything”. Which means you can price and hedge even tighter. Which means you can re-invest at a yet faster rate. Now you are blowing away less coordinated competitors who were quite content to earn their hundreds of percent a year and retire early once the markets got too tight for them to compete.

SIG was playing the long game. The parallels to big tech write themselves. A few firms who bet big on the right markets start printing cash. This kicks off the flywheel:

Provide better product –> increase market share –> harvest proprietary data. Circle back to start.

The lead over your competitors compounds. Competitors die off. They call you a monopoly.

deworsification?

Let’s start here…

Essential Wisdom from Twenty Personal Investing Classics (6 min read)

Elm Wealth distills investing lessons from 20 classics, cross-referencing them with James J. Choi’s paper Popular Personal Financial Advice versus the Professors (Journal of Economic Perspectives, Vol. 36 No. 4) that analyzed advice across 50 best-selling personal-finance books, revealing where popular wisdom aligns (or clashes!) with financial theory.

This stood out to me:

Bear in mind that the first twelve books listed above are written for the broad audience of all investors in developed market economies, with particular focus on US investors. It is well-known (and disturbing) that the financial literacy of this audience is, on average, quite low – as evidenced by a mean score of 56% (yes, that would be an “F” if not graded on a curve) on the below five-question quiz, known as the “Big Five” test by researchers. A survey conducted in 2021 found that less than one third of respondents answered at least four of them correctly, the threshold researchers define as “high financial literacy.” At least as concerning as the low test scores is the fact that the scores themselves have fallen dramatically between 2009 and 2021. If you decide to read some of these books, don’t be surprised to find a good deal of the advice proffered seems blindingly obvious if you come to them with above-average financial sophistication.

Umm, the quiz:

With the growing zoo of money distractions (crypto, gambling, prediction markets) and the results on that quiz, I’m guessing a large swath of society is gonna feel like there’s a financial tapeworm in their wallets.

I tweeted this a few days ago:

Let’s focus on evergreen financial hygeine. These were common themes I saw in the books Elm Wealth selected:

Diversify. It’s the rare free lunch: combine two assets with equal expected return and volatility, and your portfolio’s risk-reward improves.
Minimize frictions. Avoid fees, taxes, and excessive trading.
Reduce touchpoints. The fewer chances to act on emotion, the better your long-term results

Just to piggyback on the diversification bit. It sounds trite, but you’ll hear some people push back against it with the buzzword “deworsification”. When you’re as smart as Warren Buffett maybe you can use this word. But Buffet himself recognized Ed Thorp was a genius despite Thorp’s strong conviction in diversification. [And vice versa, by the way. Thorp’s recollection of hanging out with Buffet when they met in their 30s is pretty heart-warming. Game recognizing game. Apparently, after dinner Buffet showed Thorp a toy he really liked — non-transitive dice. Think of them like roshambo. A beats B which beats C which beats A.]

Here’s the cold-ass truth. Not diversifying is incinerating money. Looking back at your concentrated outcome and saying “see” is not proof of anything but survivorship. In fact, you can prove mathematically that diversifying is a free lunch.

What would you rather own?

Portfolio A: a single stock with an expected return of 10% and 30% vol

Portfolio B: an equal-weight portfolio of 2 stocks where each has 10% expected return and 30% vol but are 70% correlated

Stuff you can read if this is not clear:

The proof is sitting there in market prices too. A diversified portfolio of inferior credits will have a higher rating than the bonds in the basket. A higher rating means a higher price (lower yield).

The nuances of that are better understood today than they were in the heyday of CDO-squared.

See the GFC through a quant’s eyes

Of course, diversification always means you left something on the table in hindsight. You sign up for FOMO. But this is the nature of every decision. If you get crushed, you wished you traded zero and if you win, you wish you traded more. Results alone tell you nothing about the quality of your shot.

To complete the point:

“If you invest and don’t diversify, you’re literally throwing out money,” stated Jeff Yass. “People don’t realize that diversification is beneficial even if it reduces your return.”

Why is this the case? “Because it reduces your risk even more,” added Yass. “Therefore, if you diversify and then use margin to increase your leverage to a risk level equivalent to that of a nondiversified position, your return will probably be greater.”

The modern pod shops are another triumph of diversification, which takes us to the next section…

I am impressed by the multi-managers on the whole. They continue to generate positive returns with outstanding Sharpe ratios and thus far don’t capture the same downside as conventional 60/40 portfolios.

I think of them as the holy grail marriage of deep security research that you would have associated with a long/short fundamental manager plus the quantitative risk management and attribution metis that prop trading firms trading their own money have accumulated through the decades.

Many investors, usually from the cheap seats, want to hate on them because they don’t match the SP500 plus their pass-through fees are multiples of typical fees. Not to mention, hedge fund managers are just natural villains to normal people who maintain a Richard Scarry worldview about which jobs are valuable (eh, like any competitive profession, some of them are decent and some of them are vampires).

Regardless, this quote from Byrne Hobart conveyed something I never found the words for, so as soon as I read it, I had to clip it.

From Why Does Volatility Matter?:

If the portfolio you’re looking at is 100% net long conventional asset classes, and if you think it’s absurd to pay high fees in order to match the S&P with less liquidity—lucky you! You’re part of society’s financial shock absorber, a middle class or above saver in a rich country with functioning capital markets. But if you’re in that position, there’s a very real sense in which joking about how the S&P has outperformed complicated multi-manager setups year-to-date is a form of financial punching-down. They have a different benchmark, and a harder job. And they’re doing you the very generous service of ensuring that the next time you buy the S&P 500, the price of every single component reflects the collective attempt by thousands of professionals with massive data and analytics budgets who are all trying to push the price 1% closer to optimal.

I want to share another post from Andrew who I introduced last week.

So-Called “Bonds” in Prediction Markets

Great subtitle:

Rare events teach slowly

Moontower #292

Friends,

I really liked this listicle by Sasha Chapin:

50 things I know

I was gonna do my usual thing of sharing which items from the 50 resonate most for me.

I’ll give you one.

11. I know that talent doesn’t feel like you’re amazing. It feels like the difficulties that trouble others are mysteriously absent in your case. Don’t ask yourself where your true gifts lie. Ask what other people seem weirdly bad at.

But what I want to know is which of these land for YOU.

This was an excuse to do a project.

Here it is…go vote:

https://sasha.moontowermeta.com/

I have been doing a lot more vibe-coding recently. This project was an excuse to make something that required a database backend to record the likes and user sessions.

I used AI to then document the process which you can read about here:

General environment and stack:

👨🏽💻Setting up Git + Vercel + Terragon

Specific for the sasha voting page:

☕1-page guide

🔍Detailed tutorial

You can find a new portal here and in my signature down below:

🤖Resources to Get More Out of AI

Related and timely — this week Khe documented making a CRM which is the next-level up because it uses authentication (ie username/password to login):

I coded an entire CRM from scratch

If there are any resources you love as you explore, feel free to share!

Money Angle

Essential Wisdom from Twenty Personal Investing Classics (6 min read)

This stood out to me:

Umm, the quiz:

I tweeted this a few days ago:

Let’s focus on evergreen financial hygeine. These were common themes I saw in the books Elm Wealth selected:

Diversify. It’s the rare free lunch: combine two assets with equal expected return and volatility, and your portfolio’s risk-reward improves.
Minimize frictions. Avoid fees, taxes, and excessive trading.
Reduce touchpoints. The fewer chances to act on emotion, the better your long-term results

What would you rather own?

Portfolio A: a single stock with an expected return of 10% and 30% vol

Portfolio B: an equal-weight portfolio of 2 stocks where each has 10% expected return and 30% vol but are 70% correlated

💡Stuff you can read if this is not clear:

The nuances of that are better understood today than they were in the heyday of CDO-squared.

🔗See the GFC through a quant’s eyes

To complete the point:

“If you invest and don’t diversify, you’re literally throwing out money,” stated Jeff Yass. “People don’t realize that diversification is beneficial even if it reduces your return.”

The modern pod shops are another triumph of diversification, which takes us to the next section…

Money Angle For Masochists

Regardless, this quote from Byrne Hobart conveyed something I never found the words for, so as soon as I read it, I had to clip it.

From Why Does Volatility Matter?:

If the portfolio you’re looking at is 100% net long conventional asset classes, and if you think it’s absurd to pay high fees in order to match the S&P with less liquidity—lucky you! You’re part of society’s financial shock absorber, a middle class or above saver in a rich country with functioning capital markets. But if you’re in that position, there’s a very real sense in which joking about how the S&P has outperformed complicated multi-manager setups year-to-date is a form of financial punching-down. They have a different benchmark, and a harder job. And they’re doing you the very generous service of ensuring that the next time you buy the S&P 500, the price of every single component reflects the collective attempt by thousands of professionals with massive data and analytics budgets who are all trying to push the price 1% closer to optimal.

I want to share another post from Andrew who I introduced last week.

So-Called “Bonds” in Prediction Markets

Great subtitle:

Rare events teach slowly

Stay groovy

☮️

Moontower Weekly Recap

Posts:

why poker is used to train traders

This video is the best articulation of why poker is used to train traders.

From the description:

Jerrod Ankenman, professional poker player, co-author of The Mathematics of Poker, and Quantitative Researcher at Susquehanna, explores the connections between poker and trading. Jerrod, who began his career playing poker and went back to earn his PhD, explains how concepts like probability, expected value, risk management, and game theory apply in both the card room and financial markets. Poker and trading both demand strong mathematical thinking, disciplined decision-making, and the ability to manage uncertainty under pressure. Jerrod shares lessons from his poker career (which includes a WSOP bracelet win) that translate directly into quantitative finance and trading strategy.

If you are a trader this video despite not being technical is alpha. It tells you where to look if you pay attention.

A few ideas I’ll re-emphasize:

SIG treats poker as a structured way to train probabilistic thinking. Jerrod structures the flow of the video as a parallel between 3 concepts in poker and their analogs in trading.

Ante
Decision practice
Interpreting outcomes

You’ve heard this before — both poker and trading require making decisions with incomplete information. But a more subtle point is about speed. The goal in both is the same:

“Make the decision now that’s as close as possible to what you’d decide if you had infinite time and information.”

Both poker and trading have an information structure of what’s private and what’s public. But also what’s behavioral. Examples of the trading version of these ideas:

Private info: proprietary models, internal data, trader knowledge sharing.

Public info: news, filings, Fed releases.

Behavioral info: order flow and price action. In other words, what others are doing.

This last point gets a lot of emphasis and maps strongly to poker because at its core, it recognizes that trading is an adaptive, adversarial game not a physics problem. It’s an evolving pattern recognition and categorization problem.

You can’t model every opponent individually, or every trade uniquely, because you shrink the statistical power of your findings. You must intelligently group similar situations into profiles. Familiar poker examples: “tight-aggressive regulars,” “recreational loose players”. In trading, profiles can account for who (ie retail), periodicity (time of day, time of month, and so forth), or why (index rebalance, hedging mandates).

The art is balancing taxonomy and specificity to have enough data to be statistically meaningful, but close enough to be relevant.

A few other powerful ideas:

The big risk isn’t volatility.

It’s being wrong about your edge. The market wiggles are the nature of the game, that’s not a risk.

[I’ll take a moment to repeat myself — if you blow out because the market moved, you were committing malpractice. Being aware that the market will do things amounts to no more than a toddler understanding of risk. Volatility shouldn’t keep you awake at night. That the exchange might cancel your trades or that they may ban orders should.

From Investopedia on the response to the Hunt Brothers’ silver squeeze:

Federal commodities regulators introduced special rules to prevent any more long position contracts from being written or sold for silver futures. This move stopped the Hunts from increasing their positions by temporarily suspending the fundamental rules of the commodities market. With longs frozen and shorts free to pile in, the price of silver began to slide.

From my floor days, I can tell you there’s lore about who knew the valuable bit of info that you were only going to be allowed to do opening trades on the short side. Exchanges were run by the traders and brokers before they went public. This is the weird black or grey swan stuff that bosses worry about.

A company going bankrupt? That’s a line item in portfolio_shock_analysis.xls, not something that makes you cry in public to your investors.]

Back to Jerrod. A big risk is being wrong about your edge. It’s a risk because edge hides behind low signal-to-noise. This is one of the great teachings of poker. Short-term results are noise. He explains that in Limit Hold’em, even a high edge hand has only .02 big bets worth of expectancy vs a standard deviation of 2.5 bets.

[Kris: In investing language, a .008 Sharpe for one trial. The SP500 has a daily expectancy of about 3 bps and 100 bps standard deviation for a daily Sharpe of .03. The poker hand has almost 4x the noise of the daily SP500 return.]

Since poker teaches that you will make the right decision and still lose money, it trains you to emotionally decouple decision quality from result quality.

This is a ceaselessly profound concept. Not because it’s so clever, but because of how it resists internalization. It’s easy to understand, it’s hard to apply the understanding to how we receive the world. Fooled by randomness might be a tired title, but it’s never been stronger as we are bombarded with data.

The risk of being wrong about your edge is insidious because the relative efficiency of markets means it’s hard to make excess returns, but it’s also hard to lose too much doing sensible things. The problem is when sensible things aren’t adding value beyond randomness, but you think they are. You’re wasting your life tossing coins.

[Unless you like action for the sake of action. In that case, you’re understimulated. Go take a risk in the name of actual growth or something.]

The link between speed and skill

Jerrod notes that you don’t have time to “go to the lab” mid-hand or mid-trade. Edge requires building mental shortcuts and intuitions that perform well under time pressure.

This feels easiest to imagine in the world of sports. I’ve heard elite athletes talk about how the big jump from say college to pros is “just” the speed of the game. It’s not that they are doing new things, it’s that they must be able to do the same things faster without losing precision.

I remember reading a profile many moons ago about Jason Kidd who was known for his passing and court vision. I got the sense that he could see a split second into the future. Being able to make and execute a decision just a tiny bit faster compounds into outlier results over time. The long-term ROI on having your intuition slightly better tuned works to disproportionate effect.

This echoes. We play the mock trading games and some people are just a tad faster every time. Maybe when new info hits the game they refresh their market quickly, not necessarily making a perfect 2-way, but it’s biased in the direction of the asymmetry. Ricki Heicklen discussed this with Patrick McKenzie. If you are trading “the sum of the siblings in the room” and someone’s count is revealed, do you Bayesian update in the right direction and in a proportionally coherent way? When you see someone do this consistently, you know they’re clocking differently than the others (and I’m excluding the clueless whose updates are logically incoherent to how they processed a similar situation in the opposite direction).

In competitive scenarios, if you can debug your thinking in the moment, you’re too slow. When making a market for a broker, you need to hear the order, intention, what’s not said and how the trade looks vs the framing of the option chain in seconds (and this of course assumes your tools are already designed with this workflow in mind, showing you the info you need when you pull up the ticker). This is obviously not happening if you need to step through expected value trees. There’s no substitute for reps if you need to decide faster than the speed of system 2 axon to dendrite sex.

Feedback loops to build that intuition

Jerrod is blunt. The best way to learn in poker and trading is post-mortem discussion. Go over the tape with your team. Chat scenarios. A great feature of trading is if you love it, you want to talk about it, so this doesn’t feel like work.

When I was at Parallax, I used to carpool with 3 other traders. Shout out to Steve, Ben, and Rob — I still wish we livestreamed those rides. The commute in the morning was your typical sports or current events banter (ok fine, gossip). But the ride home was all play-by-play of trading scenarios from the day. What happened, what would you have done there, etc?

While Jerrod treats discussion as the primary way to learn (I agree — trading is an apprentice craft) he does acknowledge a role for books.

[Even though I recommend some books, they are more for describing the overall epistemic landscape or inspiration. 99.9% of what I know about trading comes from discussion or experience. I either learned how to price or think about something from someone else or after discovering first-hand a new way to lose].

He mentions that most poker books are wrong. He offers a strategy for figuring out which ones are good. But he also encourages reading the bad books because it reveals how your opponents might think. That bit reminded me of an old post I wrote:

Twitter Reminds Me Of The Trading Pits

[Random: I was hanging out with a trader from my cohort who now runs education for a big prop firm in Chicago. He was re-learning poker because a lot of the stuff we learned 25 years ago is now considered wrong. I’m not surprised, since 1 year of poker information in the online, poker-on-ESPN, poker-celebrity-giving-TED-talks era likely generated a decade worth of info from 20th-century poker.]

In sum, SIG is using poker to build the same mental circuitry that trading relies on in an enclosed, fast-feedback petri dish. It’s speedrunning experiential learning in a low-signal environment so the requirements of a successful trading career seem less alien. If trading were as easy as “just study and you’ll get good grades”, motivation and time would be sufficient ingredients. With trading and poker, you could have infinite time but if you don’t know how to learn, you’re pushing on a string.

The overfit sauna

The phrase “studies show…” is almost always followed by bs. To understand why, I’ll point you to a post by David Epstein.

David is the author of Range, The Sports Gene, and the new book Range Adapted for Young Readers which I bought for my 12-year-old son and nephew.

I’ve been a long-time reader of David’s letter and this post is both useful and timeless.

Everything in Your Fridge Causes and Prevents Cancer

It’s a reminder that outlier studies and results in general make headlines, but are statistically inevitable if you do enough studies.

Excerpt:

It wasn’t every sauna enthusiast who reaped the supposed protective effect against dementia; it was specifically those who used a sauna 9-12 times a month. Sauna bathers who hit the wooden bench 5-8 times a month — sorry, no effect. And those who went more than 12 times a month — again, no luck.

That should raise a caution flag in your head.

When only a very specific subpopulation in a study experiences a benefit, it may indeed be that there is some extremely nuanced sweet spot. But it is more likely that the researchers collected a lot of data, which in turn allowed them to analyze many different correlations between sauna use and dementia; the more different analyses they can do, the more likely some of those analyses will generate false positives, just by statistical chance. And then, of course, those titillating positive results are the ones that end up at the top of the paper, and in the press release.

Here’s the point I want to hammer home: when you see a tantalizing health headline — like that saunas prevent dementia — keep an eye out for indications that the effect only applies to specific subgroups of the study population. Even if the headline is very authoritative, revealing nuggets are often buried lower in the story.

I want to stress that you shouldn’t assume the sauna results can’t possibly be true. But when you see Bears-undefeated-in-alternate-jerseys type conclusions — and someone is claiming one thing causes the other — you should hold out for more evidence.

This doesn’t just happen in health news. Investing/trading is another area where making a mountain out of a statistical molehill is rampant. Unless you are specifically studying a phenomenon that you’d expect to be discontinuous (binary, “phase change”, threshold cutoff) you should be wary of any signal from a specific range of an otherwise continuous function.

I’ll take a simple example from Kris Longmore’s article explaining how month-end rebalance trades work. The post is titled How Wealth Managers Pay You To Trade. He writes (emphasis mine):

How I’d Test This

So here’s the hypothesis: if we can identify which asset outperformed during the first part of the month, the underperformer should outperform as we approach month-end, when rebalancing pressure is likely to be greatest.

The first step is simple. Pull daily data for SPY and TLT going back as far as you can get it (I used data from 2007). You can get this from Yahoo Finance – nothing fancy.

Then ask a straightforward question: If I know which asset outperformed during the first 15 trading days of the month, can I predict which will outperform during the last ~7 trading days?

Why 15 days? Because it’s roughly two-thirds of a trading month, and it gives us a reasonable window to identify the outperformer before month-end rebalancing kicks in.

Could you use 10 days? 20 days? Sure. But 15 seemed reasonable and shouldn’t really matter much. If it did, then that would be a big red flag. We want stuff that’s fairly robust to the actual implementation details.

Back in my floor days my biz partner was incubating a futures trend strategy and he’d have me look at the backtest results. I’m no scientist, but I knew enough to realize that if the signal depended on a particular value of the parameter (ie the exact amount of what defined a “breakout”, the number of lookback days, etc), then the result was overfit.

It’s the same idea as David’s sauna therapy study.

When you are in a competitive domain where many people are constantly mining, “too good to be true” discoveries should be met with extra skepticism.

A current example of this is the so-called Mississippi Miracle in which both the left and right appear to have an axe regarding the childhood literacy improvement in Mississippi schools. It checks the box of “domain where many people are constantly mining” so interventions that show huge returns deserve a lot of skepticism. You can count on Freddie deBoer to deliver that, but I think the pushback in the comments section of his post show the complexity:

There Are No Miracles in Education