Rolling Down and Out vs Closing: The Mechanics, and When the Maths Stops Working

What a roll actually is

This is an optional part of the Essentials series. Rolling is something a put seller may choose to do and may equally never do — Parts 1 to 6 stand on their own without it. It is here because it is the manoeuvre most often recommended without its arithmetic attached.

Your short put is going against you. The stock has fallen toward your strike and assignment is looking likely.

A roll closes that contract and opens a new one, usually at a lower strike and a later expiration — "down and out". Presented as one transaction by most brokers, it is two: buying back the contract you sold, and selling a different one.

The framing matters more than it sounds. A roll is not a repair. It is closing a losing position and opening a new one that happens to be in the same underlying, and the only reason it feels different is that the loss on the first leg is netted against the credit on the second and never appears as a line item.

The mechanics

A put struck at $50, expiring in two weeks, now trading at $3.40 with the stock at $47.

Leg one: buy it back for $3.40. You sold it for $1.20, so this leg books a $220 loss, then and there.

Leg two: sell a $45 put expiring in seven weeks for, say, $3.80. Credit $380.

Net you receive $40 and your position is now a $45 strike further out in time. Brokers will show this as a "$0.40 credit roll", which is accurate and hides the $220 loss inside it.

You have not fixed anything. You have paid $220 to move your obligation lower and later, and been compensated $380 for accepting seven more weeks of risk on a stock that has already moved against you.

The one condition that makes it legitimate

A roll is defensible when you would open the new position today, on its own merits, with no existing position in the underlying.

That is the whole test. If a $45 put seven weeks out is a contract you would sell anyway — the buffer is adequate against how the stock has actually moved, the liquidity is there, the premium compensates the risk, the size fits your book — then rolling into it is fine. The prior loss is sunk and irrelevant to the decision.

If you would not open that position fresh, the roll is not a strategy. It is a way of avoiding the loss appearing in your records, and the position you now hold is one you have explicitly judged not worth having.

Most rolls fail this test. They are made because the alternative is recording a loss.

Where the maths stops working

Three things break down as the underlying keeps falling, and they compound.

The credit shrinks and then disappears. Rolling down means selling a lower strike, which pays less. Rolling out means buying more time, which pays more. Early on, the time gained outweighs the strike given up and the roll is a net credit. As the stock falls further, you must roll further down to get meaningful distance, and the credit thins. Eventually a "credit roll" requires so much additional time that you are committing capital for months to collect very little.

The debit roll. Past a certain point there is no combination that produces a credit. You pay to move the position. At that moment you are spending money to postpone a loss, which is a different activity from selling premium and should be recognised as one.

Capital lock-up compounds. Each roll extends the commitment. A contract that would have resolved in two weeks now resolves in nine. That capital cannot do anything else, and the return on the position is being calculated over an ever-longer denominator — the annualised figure quietly collapses even as the premium arrives. Return on capital vs annualised return covers why.

The pattern to watch for

Rolling has a specific failure mode, and it is recognisable from the outside long before it is recognisable from the inside.

A seller rolls once, collecting a small credit, and feels they have handled it. The stock falls further. They roll again, for less. And again, now for a debit. Six months later they hold a position they never chose, in a company they no longer have a view on, at a size that has become uncomfortable, having never recorded a loss.

The record looks clean. There are no losing trades in it. There is one very large open position and a lot of capital that has done nothing for half a year.

This is averaging down with the loss hidden inside the accounting. It is the same behaviour that gets equity investors into trouble, made harder to spot because each individual step produced a credit.

The alternative nobody likes

Take the assignment, or close the position and record the loss.

Assignment on a short put is a loss — not a discount, not a strategic entry. Assignment covers why that framing matters. But it is a bounded, resolved loss. It has a number, it goes in your records, the position closes, and the capital becomes available for something you actually chose.

A perpetual roll is unbounded and unresolved. It has no number until you stop, which is why it is comfortable, and the discomfort of recording a loss is doing genuine work: it is the feedback that tells you the entry was wrong. Remove it and your process cannot improve, because your records no longer contain the information it would improve from.

What tends to separate a considered roll from a reflexive one

Four differences show up consistently. None is a rule anyone else can set for a given position.

Whether the decision predates the pressure. A roll made in the moment, on a position that is hurting, is made under the worst conditions available. Sellers who have decided in advance what would justify one are deciding something different from those improvising.

Whether there is a limit on repetition. A first or second roll is a different act from a fifth. Where a limit exists it tends to force the "would I open this fresh?" question to be asked out loud; where none exists, the question often goes unasked.

Whether the roll is a credit or a debit. Paying to postpone is a different activity from selling premium. A roll that costs money is describing the position's condition rather than improving it.

Whether the round trip is counted. Two legs means crossing the bid-ask spread twice, on a stressed underlying where spreads are wider than usual. On a modest credit that friction is a real share of whatever the roll appeared to gain.

How StratosIQ handles it

Not at all, and that is worth stating plainly rather than implying more coverage than exists.

Every contract is scored and recorded on a hold-to-expiry basis. Above the strike is a win, below it is a loss. The published record contains no rolls, no early closures and no post-entry management of any kind, because the score is a judgement about entry quality and mixing management decisions into it would make the figure describe something else.

That is also why the record is more conservative than a managed methodology would report. A strategy that rolls losers indefinitely can show almost no losses for a long time; this one shows them as they happen.

What the scoring contributes is upstream: the safety floor is set so that the situation calling for a roll arises less often. The dominant input is the distance between the strike and a floor built by ShieldIQ from how the underlying has actually behaved, reported as Fortified, Secure, Tight or Exposed, and a contract failing it does not publish.

The full published record, contract by contract and including every loss, is on the performance page.

Worth reading next

Assignment covers the outcome rolling exists to avoid. Why a 90% win rate can still lose money covers why a record with no losses in it should raise questions rather than confidence.