When your short put is deep in the money heading into expiration, you have two real choices: take assignment and own the stock, or roll the put out in time to avoid it. Neither is automatically right — the answer depends on whether you actually want to own the shares at that price.
Let's make this concrete. Say you sold a $170 put on AAPL when it was trading at $175, collected $2.50 in premium, and now AAPL is sitting at $163 a week before expiration. Your put is $7 in the money. You're staring at a potential assignment of 100 shares at $170, which means you'd be buying stock that's currently worth $163. That's an unrealized loss of $700, partially offset by your $250 in premium — so you're actually down about $450 per contract. Now what?
The case for accepting assignment
If you ran the wheel correctly, you only sold that put on AAPL because you were genuinely willing to own it at $170. That was the deal you made. So if nothing has fundamentally changed about the company — no earnings disaster, no sector collapse, no reason to think $163 is the beginning of a much bigger drop — then taking assignment and moving into the covered call phase is exactly what the wheel is designed for.
Here's the thing people miss: rolling costs you something. To roll that $170 put out 30 days, you'd probably buy it back for around $8.50 and sell the next month's $170 put for maybe $10.00, netting you $1.50 in credit. That sounds fine. But you've now extended your risk exposure by another month, you're still obligated at $170, and if AAPL keeps drifting lower to $155, you've just delayed the pain and made it worse. Rolling to avoid assignment can turn a manageable loss into a much bigger one if the stock keeps moving against you.
Accept assignment when: the stock is above your cost basis adjusted for premium, the thesis is intact, and you're comfortable holding through a recovery. AAPL at $163 with a $167.50 effective cost basis (after premium) isn't a disaster. You can sell a $167 or $168 covered call the next morning and start grinding it back.
The case for rolling
Rolling makes sense in a specific situation: when you're wrong about wanting to own the stock at that price. Maybe AAPL dropped because of something you didn't anticipate — a major product recall, a significant guidance cut, an analyst downgrade based on real data. Or maybe you simply don't have the capital to absorb 100 shares right now without it crowding out other positions.
Rolling also makes sense when you can roll down and out for a credit. If you can move from the $170 put to a $165 put 45 days out and still collect $0.75 in net credit, you've lowered your obligation by $5 and bought yourself time. That's a legitimate trade. What you want to avoid is rolling for a debit just to delay the inevitable — you're paying money to kick the can down the road, which usually ends badly.
The other time I'd roll without hesitation: if you're within a day or two of expiration and there's a clear technical support level nearby that the stock hasn't broken yet. Sometimes you get a small bounce back through your strike in the final hours. Rolling to next week buys you that chance without committing to a full 30-day extension.
The question you need to answer first
Before you touch anything, ask yourself: "If I didn't have this position, would I buy 100 shares of this stock right now at my strike price?" If the answer is yes, take assignment. If the answer is no, roll or close.
That question cuts through the noise. A lot of traders roll reflexively because assignment feels like losing. It isn't. Assignment is just the second phase of the wheel. The loss already happened when the stock moved against you — rolling doesn't undo that, it just changes the form it takes.
One more thing worth mentioning about mechanics: if you're going to roll, do it as a single spread order — buy the expiring put and sell the new one simultaneously. Don't leg into it. Trying to buy back your short put first and then sell the new one separately exposes you to the bid-ask spread twice and risks getting a worse fill on the second leg if the stock moves between orders.
What I'd actually do
On a stock like AAPL that's down 7% but hasn't broken any major support and has no new fundamental problems, I'd take assignment. I'd wake up the next morning, look at where the stock is, and sell a 30-day covered call at or just above my cost basis. If AAPL is at $163 and my effective cost is $167.50, I might sell the $168 call for $2.00 and keep working it. That's the wheel doing exactly what it's supposed to do.
Save the rolling for situations where your thesis has actually changed, where you can improve your strike meaningfully, or where you genuinely cannot afford the shares right now.
Today's action: pull up any short puts you have that are currently in the money and ask that one question — would I buy this stock at this price right now if I had no position? Your answer tells you everything you need to know.