The pot of gold is the trades we made

Vol #212: January 30th, 2025

Edge starts with enthusiasm.

You’re David, they’re Goliath, let’s get this bread. Absent the fanatical belief that you have figured out a way to outsmart the market, your strategy won’t make it past the first losing trade.

0DTE options twinkle with theta pots of gold at the end of a very short rainbow. Obvious mechanics of market participants or rebound patterns scream as the perfect setups. Quants will strongarm any pattern into a signal, and qualts will use their superior macro economic analysis to improve the pricing of liquid global securities.

The markets have many onboarding ramps. We laugh when it’s Fartcoin, but plenty of otherwise serious people want to day trade stocks based on guru Discords or swing 200x leverage on forex pips. Fringe markets like this exist mostly to fleece novice traders enticed by both the simple arithmetic of generational wealth and the low balance required to access it.

The hope is that investors can learn, lick their wounds, and move past this. Butting your capital against the brutal reality of deep liquid markets is a rapid learning experience. The lesson is not that you need more indicators, it’s that the only rainbows come as the result of hard work.

While my first investment was purchasing a basket of low fee diversified ETFs (seriously), my first real trading experience was on a now defunct pip chasing bucket site. Foreign exchange was the ideal proving ground for a few lectures in international relations and a couple weeks of beer money.

The lure of these sites was that you get enormous leverage to trade the tiniest variations in spot currency prices. If there are 1.0456 dollars to a euro, with a ~$10 trade you can buy ~$2000 worth of euros, and make a whopping $2 if the price goes up to 1.0466.

Over the last 24hrs, the EUR/USD rate has ranged from 1.0440 to 1.0380. You can see how easy 20% a day is.

The slide shows and textbooks on market efficiency don’t tell the story as well as a declining balance. Quickly $10 gets stopped out the other way, and whatever pattern you thought you saw - poof. It was probably this guy that just got an order to buy euros.

I wasn’t exactly using technical analysis to trade, but I was intrigued by its mention in the weak form of the efficient market hypothesis. If technical analysis is not predictive, then at the very least we know that all past price levels are incorporated into the current price.

Finding an intersection between my personal obsession and a group project, I set out to study moving average signals on forex pairs. Of the several dozen experiments we ran, some produced returns, but none out of sample. Further, the average of all returns was very close to the risk free interest rate.

That was enough to convince me you needed a firm and structural edge in the markets. So I got a job market making. Traders here don’t draw lines or read 10-Ks, they build a complex system that prices the supply of risk and liquidity, then offers quotes as a service.

A big part of that job is finding and maintaining a fair value for a strip of options. A good fair value means that on average the cost to hedge and inventory a position is commensurate with what you paid/collected for it. Pricing must be cognizant of both the liquidity and fundamental characteristics of the name.

Market makers in aggregate are really really good at this. So good that they make billions of dollars a year doing this for fractions of a penny on each trade. And so good that the current pricing is highly reflective of the future reality.

A good way to test if a market is efficient is by applying a naive strategy. If you can always buy stocks on Friday for less than you can sell them on Monday, then there is a weekend inefficiency. Or more likely its a marginal premium for bearing the risk of markets being closed for three overnights.

Options expiring in less than 24 hours make up roughly 50% of the volume in SPX. Yet they seem so richly priced. 24% IV for today, compared to a 14% IV only a week later.

Powered by this inefficiency and thoughtfully tapping the sign that says VRP, you start selling straddles here. Since 0DTEs were introduced in May 2022, there have been 674 trading opportunities to sell a one day straddle expiring at tomorrow’s close. You would have collected $2.35M worth of premium (total) for an average of $34.85/straddle (per share terms).

The average PnL is losing a dime on those straddles. Ten American cents on a $34.85 premium. The total is -$63.35 ($6335 nominal). On 675 trades, your loss is less than 2x your daily premium collection. There are some big losers - 5 greater than 100 points. The max gains are limited by your price - but there is also an $81 straddle you sell that returns $60.

First order straddles are highly efficient, so perhaps you want to add a volatility filter - sell high vol right? Actually it’s the opposite, the losers start to add up as you include higher volatility level trades.

That positive PnL from being somewhat selective isn’t exactly ringing the cash registers. The capital requirements of short straddles are significant, and bring your best case scenario back down to earth. At 4% interest you’re earning $185 a year on one “share” of SPX, so back of the envelope $465 over the 2.5 years of this study. A little bit better than cash with that kind of risk, and far worse than simply buying SPY? No thanks.

If ATM vol is well priced via the straddle, perhaps there is extra juice in the wings. Skew has arisen in options pricing because we know volatility tends to increase as prices go down, and those options should be priced at a higher implied volatility.

The trade here would be to trade a condor, selling further out of the money options that have higher IV on the downside. A common setup in condors (and the basis of CBOE’s benchmark) is to sell the 20 delta options and buy back something slightly further out of the money. This is functionally the same as selling the “straddle range” under the assumption that it’s good to short the options outside the expected move.

No dice here either. Whether you’re selling 5 or 10 dollar wide straddles, or nudging it 5/10/20 dollars away from the range, the PnL is very tight. In all of these instances the total PnL is between -$50 and -$150. Despite winning ~60% of the trades, the strategy is very marginally negative expectancy.

If you were hoping for a simple winner winner chicken dinner, this is somewhat bad news. Whether you’re attacking the meat or the wings of the curve, there is negative edge (albeit and importantly very small) in a basic strategy. This would only get worse with transaction costs, slippage, etc.

But if you’re hoping to put on trade structures that adjust your risk profile, or capture opportunities from more complex signals and filtering, this is incredible news. Because the markets on average are very good at pricing, dealers can provide tight and liquid markets.

For the vast majority of investment use cases for options, you simply want good pricing on a trade. Just as a stock price should be something like the discounted value of all future cash flow, an options price should reflect the best opinion right now of possible distributions. Take it or leave it if that’s a risk you want to manage or it has positive personal trade offs. At least you’ll get a fair bid or offer.

The first order efficiencies don’t eliminate the possibility of opportunistic strategies. Even those participants without structural edge can generate value by helping to arbitrage nuances and incorporate additional signals. You’ll probably have to dig a little deeper than fundamental information here, everyone else also knows what open interest and earnings dates are. Semi-strong efficient market hypothesis is also likely to hold.

The pot of gold is what brings us all the rainbow road, but it’s the filament thin markets that are the real prize.