Why a Poor Man's Covered Call Cannot Be a Runner
18 September 2026 · 5 min read
A poor man's covered call replaces 100 shares with a long-dated call at roughly a third of the capital, then sells a short-dated call against it. The part that gets skipped is that the short call has to be sold close enough to the money to cover the long call's monthly time decay, and that requirement sets the cap. Sell further out to keep more of a rally and the position costs money to hold.
We went looking for a capital-efficient way to hold a directional position while collecting premium. The arithmetic below is what we found instead. Every figure is dated, and the modelled ones say so.
The long leg charges rent
A deep in-the-money call is mostly intrinsic value with a slice of time value on top. That slice is the rent. It decays whether the stock moves or not, and the short call you sell each month is what pays it.
Modelled 2026-09-17 on a 180-day long call at 0.80 delta, 45% implied volatility, flat term structure, with a 30-day short call:
| Short call delta | Credit | Decay | Net |
|---|---|---|---|
| 0.10 | 0.47% | 0.94% | −0.47% |
| 0.15 | 0.80% | 0.94% | −0.14% |
| 0.20 | 1.18% | 0.94% | +0.24% |
| 0.25 | 1.61% | 0.94% | +0.67% |
| 0.30 | 2.09% | 0.94% | +1.15% |
Figures are percentages of spot. Below roughly 0.20 delta the trade has negative carry, which is a polite way of saying you pay for the privilege of holding it.
This closes off the version most people want. "Sell a far out-of-the-money call so I keep the upside" is available, and it costs money every month. A 0.25-delta monthly call on a 45%-volatility name sits about 9% above spot, so a stock that advances 30% over a quarter gets called away repeatedly on the way up.
The safety condition depends on tenor, not depth
The condition that makes the structure safe is that the debit stays below the gap between the two strikes. If it holds, being assigned on the short call cannot lose money, because you can exercise the long call into the assignment.
Write S for spot, K_L for the long strike, K_s for the short strike, X_L for the long call's time value and C_s for the short call's credit. The condition debit ≤ K_s − K_L expands and the long strike cancels out of it:
K_s ≥ S + X_L − C_s
So the binding quantity is the long call's time value, measured against the credit. Going deeper reduces time value, because time value is largest at the money. Going longer increases it. Modelled at 0.80 delta long and 0.25 delta short, as a percentage of spot:
| Long tenor | IV 30% | IV 45% | IV 65% |
|---|---|---|---|
| 90 days | +4.09 | +6.64 | +10.24 |
| 180 days | +2.36 | +4.38 | +7.07 |
| 270 days | +0.85 | +2.41 | +4.27 |
| 365 days | −0.63 | +0.50 | +1.48 |
| 540 days | −3.17 | −2.82 | −3.48 |
Positive clears the condition. The common advice to buy a twelve to twenty-four month LEAPS breaks it at every volatility level we tested. A 90 to 270 day long leg holds.
The term structure works against you
The structure buys long-dated volatility and sells short-dated volatility, so the shape of the volatility curve is a direct cost. We pulled live chains on 2026-09-17 at 11:30 EDT across fourteen large caps. Thirteen of the fourteen were in contango, with the long-dated implied volatility sitting 10% to 42% above the 30-day figure. On the widest of them the modelled return on capital fell from 2.96% a month to 1.15%.
Contango is simply what this instrument costs. It showed up on thirteen of the fourteen names we quoted, so screening for the exception leaves very little to choose from.
Against simply holding the shares
We simulated nine months of monthly rolls over 40,000 paths on a large cap with 30% implied volatility, selling at implied and realising at 25%, so the volatility premium is credited rather than assumed away. Drift 8% a year.
| Mean | Median | Chance of loss | Worst 5% | |
|---|---|---|---|---|
| Hold the shares | 6.0% | 3.5% | 44% | −27.3% |
| Cash-secured put | 5.8% | 7.1% | 11% | −2.7% |
| Poor man's covered call | 13.7% | 3.0% | 49% | −78.4% |
The diagonal's 13.7% average comes from a handful of very large winners. The median holder of it earned 3.0% while a shareholder earned 3.5%. At zero drift the diagonal's median was −17.0% against −1.9% for the shares.
The cash-secured put is the interesting row. It matched buying the shares on average return while losing money in 11% of paths instead of 44%, with a worst-5% outcome of −2.7% against −27.3%. That gap is the whole argument for selling premium, and it has nothing to do with capital efficiency.
One caution on all three rows. The simulation uses geometric Brownian motion, which has no gaps, and a gap is exactly what hurts a short option. Every short-premium figure above is flattered. Holding shares carries its own tail, and that one is understated too, though by less.
What the structure is actually for
The decision reduces to one question: do you expect the company to beat roughly 8% a year? Above that, owning shares wins on average and the diagonal wins if you will tolerate a 49% loss rate and 78% drawdowns to get there. Below it, selling puts wins, and it keeps winning in the flat case that neither of the others survives.
That is a question about the company. No amount of structure settles it, which is the honest limit of this kind of analysis.
Two further costs worth pricing before anyone reaches for it. On a dividend payer you forgo the yield a shareholder receives, and your short call can be exercised the day before the ex-date whenever its time value drops below the dividend, leaving you short shares and owing the payment. And the long leg has to trade: across the fourteen names we quoted, round-trip cost ran from 2% on the most liquid to 22% on the thinnest, which is enough to erase a year of carry on its own.
Everything above is arithmetic on modelled and quoted option prices, and none of it identifies anything to trade. If you want to see how we size the ordinary version of this, what delta to sell and yield on capital cover the ground, and portfolio margin covers the capital side.
This is analysis and education rather than investment advice. Options carry the risk of total loss of the premium or debit paid, assignment can happen early, and the figures above are modelled under stated assumptions that will not match your account. Parts of this research were produced with AI assistance and reviewed before publication. Questions and corrections are welcome via contact.
Common questions
- What is a poor man's covered call?
- You buy a long-dated call deep in the money, usually around 0.80 delta, and sell a short-dated out-of-the-money call against it. The long call stands in for 100 shares at roughly a third of the cost, so the position gives similar upside exposure per dollar. It is also called a diagonal call spread. The trade-off is that your maximum loss is the whole debit, where a shareholder's loss is only the fall in the share price, and the long call expires while shares do not.
- Can you run a poor man's covered call without capping your upside?
- Not at a price worth paying. The long call loses time value every month, and the short call has to cover that decay. Modelled on a 180-day long leg at 45% implied volatility, a 0.10-delta short call brings in 0.47% of spot against 0.94% of decay, and a 0.15-delta call brings in 0.80% against the same 0.94%. Both are negative. The breakeven sits near 0.20 delta, which on that example is a strike about 11% above spot. Selling further out to keep more upside means paying to hold the position.
- How long should the long call be in a poor man's covered call?
- Shorter than most guidance suggests. The safety condition for the trade is that the debit stays below the distance between the two strikes, and that condition depends on the long call's time value, which grows with tenor. Modelled at 45% implied volatility with a 0.25-delta monthly short call, a 90-day long leg clears the condition by 6.6% of spot, 270 days clears by 2.4%, and 540 days fails by 2.8%. The usual advice to buy twelve to twenty-four month LEAPS breaks the condition it is supposed to protect.
- Is a poor man's covered call better than just buying the stock?
- At the median outcome, no. Simulated over nine months on a 30%-volatility large cap with monthly rolls and 8% annual drift, the diagonal returned 3.0% at the median against 3.5% for holding the shares, with a 49% chance of a loss and a worst-5% outcome of −78%. Its 13.7% mean came from a small number of very large winners. At zero drift the median was −17%. The structure carries leverage, so it beats shares in a strong advance and loses badly in a flat or falling one.
- Does a poor man's covered call have dividend risk?
- Yes, and in two directions. A call holder receives no dividend, so on a paying company you give up the yield a shareholder collects. Separately, your short call can be exercised early the day before the stock goes ex-dividend whenever its remaining time value is worth less than the dividend, which leaves you short the shares across the ex-date and owing the payment. Both effects are largest on the high-quality, dividend-paying companies the structure otherwise suits.
Read next
- Technical Thoughts
More Delta Means More Yield, and More Risk. How We Found a Middle Ground
Selling deeper puts pays more premium and carries more risk. Choosing what delta to sell came down to targeting a chance of assignment rather than a delta label, plus explicit rules for which companies may carry it.
- Technical Thoughts
Your Option Book Is a Knapsack Problem (The Hard Kind)
Allocating capital across option positions makes portfolio optimization a knapsack problem, of the multi-dimensional, integer, class-constrained kind that has no efficient exact solution. Here is what our optimizer gives up to answer it anyway.
- Technical Thoughts
A Mechanical Way to Pick Which Stocks to Wheel
Most wheel candidate lists are habit, not analysis. Here's what happened when I made the selection mechanical — scored on diversification, not on premium.