The wheel is a short-term capital gains machine

I did not think of the wheel as a tax strategy. I thought of it as an income strategy that happened to have tax consequences, which is a comfortable way of not looking at something.

Then I added up a spring of weekly SOFI cycles and saw the shape of it. Every call that expired worthless was a realized gain. Every lot called away was a sale. Running it weekly, I had manufactured a taxable event roughly every seven days, and not one of them had been held long enough to be taxed at the rate I would have preferred.

Nothing about the wheel is long-term

The rule itself is simple, and it was not the part I got wrong. From IRS Publication 550:

If you hold investment property more than 1 year, any capital gain or loss is a long-term capital gain or loss. If you hold the property 1 year or less, any capital gain or loss is a short-term capital gain or loss.

Now look at what the wheel does to a holding period. A put gets assigned and your clock starts the day you buy the shares, not the day you wrote the put. Then you sell calls against those shares, and the entire point of selling a call is that you are willing to part with them. If the stock cooperates, they leave.

On a weekly cadence, on a stock oscillating in a two dollar band, they leave often. I was never going to reach a year. I was not trying to.

The premium is more absolute still. A written option that expires is short-term no matter how long it was open, which the same publication puts in one sentence:

If your obligation expires, the amount you received for writing the call or put is short-term capital gain.

That is the rule for premium on an option you let expire, which on a wheel is most of them. Close one early or get assigned and the premium stops being a standalone gain and starts adjusting what you paid or received for the shares instead. Different arithmetic, same ceiling: I could not find the patient version of this trade that reaches a year on the premium.

The number I was quietly getting wrong

The wheel produces two cost basis numbers, and the tax one is what matters here. Your break-even keeps falling as premium comes in. Your tax basis does not move at all. The gap between them is not a rounding error, it is a running total of gains you have already realized and will owe on, whether or not the money still feels like it is sitting in the position.

That is the part that surprised me. It is possible to be up on a position economically, look flat on paper, and still owe tax on several hundred dollars of short-term gain, because the premium that lowered your break-even was income the moment the call expired.

The two dates I watch now

Neither is exotic. Both are calendar arithmetic nobody wants to do by hand every week.

When a lot crosses a year. Mostly it never does. But sometimes a put gets assigned, the stock drops, and you stop writing calls because the strikes are not worth writing. Now you are holding shares rather than running a wheel, and that lot has a date after which it would be taxed differently – how much that is worth depends on a bracket I do not know. What I did know is that I was never going to work the date out by hand for every lot.

The thirty days around a loss. Selling at a loss and buying back too soon disallows the loss. Publication 550 again, on wash sales:

A wash sale occurs when you sell or trade stock or securities at a loss and within 30 days before or after the sale you: Buy substantially identical stock or securities, … Acquire a contract or option to buy substantially identical stock or securities.

Read that clause slowly if you run the wheel, and then read it again for what it does not say. It names acquiring a contract or option to buy. Writing a put is the other side of that trade: you are granting someone else the right to sell to you, and taking on an obligation rather than acquiring an option. When I went looking for whether a short put ever counts as substantially identical, I found argument rather than an answer – the case for it rests on a deep in-the-money put being assignment in all but name, not on the sentence above.

Note also that “within 30 days before or after” means the window has two sides, so it can close around a trade you have not placed yet.

How any of that resolves for your account depends on facts I do not have, and I am not qualified to give you the answer. What I will say is that “sell the loser, keep wheeling it” turned out to be a sentence with more in it than I assumed, and it is worth saying out loud to someone who does this professionally before you act on it.

Why the calendar came before the mistake

The honest version of this is not that I got burned and then built a tool. It is that I could see the shape of the mistake from a distance.

The wheel generates events faster than I can reason about them, all of them short-term, some landing inside a window that opens thirty days before something I have not done yet. That is not a problem you solve by being careful in the moment. You solve it by having the dates in front of you before you place the trade.

So the Tax page has a holding-period calendar and a radar for harvestable positions, and neither is clever. They are arithmetic I was otherwise doing badly, late, and in April.

It models wash sales and holding periods. It does not model qualified covered call rules, which can affect the holding period on shares you have written calls against, and it is not a tax filing tool. I would rather name that gap than let a calendar imply it is covered.

If you run the wheel, the wheel tracker is where I watch the cycles, and the same numbers feed the tax view. There is also an options wheel calculator if you want to see the shape of a cycle before committing to track one.