To The Moon

Mathematical Money | October 3, 2026

A few people have asked me to explain how I actually sell puts, so this is the whole thing from the beginning — what it is, how I choose strikes and dates, how to place the order without giving money away, and a live mistake of mine that illustrates the one rule everybody breaks.

If you already sell premium regularly, the first section will be familiar. Skip to the ordering mechanics, because that's where most people quietly lose a chunk of their edge.

What you're actually agreeing to

Selling a cash-secured put means promising to buy 100 shares of something at a set price, by a set date, in exchange for cash today. You set aside the money to honour that promise — that's the "cash-secured" part — and the cash you receive is yours whatever happens.

Two outcomes. The stock stays above your strike and the contract expires worthless, so you keep the premium and repeat. Or the stock falls below your strike and you buy the shares at that price, having kept the premium as a discount on your entry.

That's it. The whole strategy is being paid to make a promise you'd be happy to keep, which brings us to the only rule that genuinely matters.

Only sell puts on things you want to own

Everything else in this post is optimisation and this is the actual rule, so I'd rather state it bluntly than bury it in a list.

If you would not be pleased to own 100 shares of that company at that strike price, do not sell the put. The premium is never enough to compensate for being handed something you didn't want at a price you don't like, and the moment you get assigned on a company you have no conviction in, you'll sell it at a loss within a fortnight and wonder what went wrong.

In practice that means high-conviction names you'd hold for years, where you've done the work, at strikes that represent a price you'd genuinely like to pay. I sell puts on two names, mostly, and I'd be content to own more of both at the levels I'm writing.

What it rules out is the thing beginners find most tempting — selling puts on whatever has the fattest premium. High premium means high implied volatility, and high implied volatility means the market is pricing a big move. You're being paid more because the risk is genuinely greater, not because you found an inefficiency.

Don't catch a falling knife, and here's me catching one

The dangerous setup isn't a stock that's already fallen. It's a stock that's still falling.

A put sold into a downtrend gets run over, because the thing that makes your strike look comfortably far away — recent price — keeps moving. The buffer you thought you had evaporates while you watch.

Here's mine, live, right now. On 18 September I sold 100 MARA puts at the $12 strike for $0.49 each, taking in $4,900. MARA had closed that day at $13.24 after jumping 14% in a single session. The strike sat roughly 9% below the price. That felt like a comfortable margin.

MARA is now $11.23. The strike is breached, those contracts are in the money, and they expire on Thursday.

Two things went wrong and neither was the premium. I sold into a spike, which means I was writing at a local high and mistaking the post-rally price for the real one. And I accepted a 9% buffer on a stock whose normal monthly move is far larger than 9%, which isn't a buffer at all — it's a rounding error with a decimal point.

The lesson I'd draw is not "avoid volatile stocks." It's that your margin of safety has to be measured against how much that specific stock actually moves, not against what feels like a lot in general.

Choosing your strike: what delta actually tells you

Delta has a convenient second meaning for option sellers. A put with a delta of 0.30 has roughly a 30% chance of finishing in the money — of you being assigned.

So the strike choice is really a probability choice. Around 0.15 delta is conservative: you'll rarely be assigned, you collect modest premium, most contracts expire worthless. Around 0.30 is the middle ground, which is roughly where I sit. Above 0.40 you're collecting serious premium and should expect to be buying stock regularly, which is fine if that was the plan and painful if it wasn't.

The honest framing is that there's no free strike. A strike further from the money is safer and pays less, in proportion. What you're choosing is how often you want to end up owning the shares.

Choosing your expiry (DTE), and the theta maths nobody explains

Here's the piece of arithmetic that changed how I do this, and it's the reason I write roughly 30 days to expiry rather than 90. Option premium doesn't scale with time in a straight line. It scales roughly with the square root of time. So a 90-day put doesn't pay three times what a 30-day put pays on the same strike — it pays about 1.7 times.

Which means selling the 30-day contract three times in a row collects substantially more than selling one 90-day contract once, for the same stock and the same risk window. Roughly 70% more, before costs.

That's the entire argument for shorter-dated premium selling, and it's why I write about a month out rather than a quarter. Time decay is not linear; it accelerates as expiry approaches, and you want to be repeatedly harvesting the steep part of that curve rather than sitting through the flat part.

There is a real cost to this. More contracts means more spreads paid, more assignment events to manage, and more decisions — and decisions are where people make mistakes. Someone selling one annual contract makes one error a year. I make considerably more opportunities for error than that.

Take the money before expiry

The corollary of non-linear decay is that the last stretch of premium is the slowest and least rewarding part of the contract's life. Two examples from my own book. On 11 September I sold 100 MARA $11 puts at $0.442, with MARA at $11.98. Seven days later I bought them back at $0.14 — 68% of the premium captured, with two weeks still on the clock, for $2,869. On 18 September I sold three Coinbase $175 puts at $4.85 with the stock at $194. Eleven days later I closed them at $2.04 for $838, which is 58% captured with ten days still to run.

In both cases I could have waited and collected the rest. I'd have tied up the same cash for another week or two to earn the dregs, while carrying the full risk of something going wrong. Taking half to two-thirds and redeploying is almost always the better use of the capital.

Placing the order without donating money

This section matters more than people think and it's the most ignored part of options trading, which is why I've given it more room than the strike selection everybody obsesses over. Never use a market order on an option. Stock spreads are usually a cent; option spreads can be ten or twenty percent of the contract's value. A put quoted at $0.45 bid and $0.60 ask has a mid of $0.525, and crossing to the bid hands over 14% of your premium before the trade has done anything. Do that fifty times a year and it's the difference between a decent strategy and a mediocre one.

Always use a limit order, and start at the mid. If it doesn't fill, work down in small increments — a cent or two at a time. Being patient for ten minutes is worth real money here.

Use the spread itself as a liquidity test, too. If the bid-ask is more than about ten percent of the mid, that strike is thin, and you'll pay the same penalty again when you go to close it. I'd rather take a slightly worse strike with a tight market than a perfect strike I can't exit cleanly.

Timing within the day

Avoid the first thirty minutes. Spreads are at their widest, overnight news is still being digested, and market makers are pricing uncertainty into every quote. You will get a worse fill on identical risk, purely for being early.

Spreads generally tighten as the session settles, and the last hour or two of trading tends to give the cleanest execution. That's when I place most of mine.

One extra wrinkle worth knowing: time decay runs on calendar days, not trading days. A contract sold on Friday afternoon decays over Saturday and Sunday while nothing can happen to the underlying. You collect three days of theta for one day of market risk, which is the closest thing to a free lunch in this whole business.

The short version

Sell puts only on things you'd be glad to own, at prices you'd be glad to pay. Size the strike by delta so you know your real assignment odds. Write about a month out and harvest the steep part of the decay curve rather than holding to expiry. Close at half to two-thirds of the premium and redeploy. Use limit orders, never market, and don't trade the opening half hour.

And measure your margin of safety against how much that stock actually moves. Nine percent sounds like plenty right up until you check what the thing does in an average month.

Two things I'd like to hear from others on.

For those selling puts regularly, where do you sit on delta, and has that changed after you've actually been assigned a few times? I've drifted more conservative over the years and I'm not sure whether that's wisdom or scar tissue.

And has anybody measured what the first-thirty-minutes penalty actually costs them? I avoid it on principle and from experience, but I've never properly quantified it, and I suspect it's larger than I assume.

Stop guessing. Start calculating.

Live to fight another day. 🤙