Cash-Secured Puts and Covered Calls, Screened by the Numbers

The two strategies

Both strategies sell an option rather than buying one. The seller takes cash up front, called the premium, and accepts an obligation lasting until expiry. The premium is yours the moment the trade fills, in every outcome. What varies is whether you also transact in the shares, and at what price.

A cash-secured put means selling a put with a strike below the current share price while setting aside enough cash to buy 100 shares at that strike. If the stock closes above the strike at expiry, the put expires worthless and the premium is your entire result. If it closes below, you can be assigned, obliging you to buy 100 shares at the strike even though the market price is now lower; your effective cost basis becomes the strike minus the premium. The cash is committed in advance rather than borrowed, which is what makes the position cash-secured. The maximum loss is substantial: shares falling to zero leave you holding worthless stock you paid the strike for.

The symmetry is useful: a cash-secured put commits you to buy lower, a covered call to sell higher. Each converts an uncertain outcome into a certain payment today, sized in proportion to how uncertain the market believes that outcome to be.

Why delta is the screen's center of gravity

Delta is the sensitivity of an option's price to a one dollar move in the underlying: a 0.30-delta call gains roughly thirty cents per share when the stock rises a dollar, and a −0.30-delta put gains roughly thirty cents when it falls a dollar. Our filters use absolute delta, so only the magnitude matters.

What puts delta at the center of a premium-selling screen is its second, looser reading: absolute delta is widely used as a rough approximation of the probability that an option finishes in the money. On that interpretation a 0.20-delta put is loosely described as having about an 80% chance of expiring worthless, letting the seller keep the full premium. The framing also makes the central tradeoff legible. Higher delta means a strike closer to the current price: more premium, greater chance of assignment. Lower delta means a strike further away, less premium, better odds the contract simply expires. There is no free position on that curve.

Our pre-filter accepts absolute deltas from 0.05 to 0.55, deliberately wide, and a narrower preferred band of 0.15 to 0.35 drives the ranking; contracts outside it are penalized in the score rather than discarded. That band is a conventional middle ground. Below roughly 0.15 the premium is often too thin to survive the bid-ask spread; above roughly 0.35 assignment becomes an outcome you should expect to manage. The conservative profile narrows the pre-filter to 0.05–0.25 and the band to 0.10–0.20, favoring distant strikes with thinner premium. The aggressive profile widens the pre-filter to 0.20–0.55 and moves the band to 0.30–0.45, accepting materially higher assignment probability for larger premium.

One caveat deserves emphasis. Delta-as-probability is an approximation, not a measurement. It falls out of Black-Scholes, which assumes prices follow a lognormal distribution, whereas real markets produce far more extreme moves than that model expects and those moves cluster. The approximation therefore understates tail risk: the losses concentrated in that remaining 20% are larger, and more correlated across positions, than the number implies.

Annualizing the yield, and why that number misleads

The headline number on each scanner row is an annualized yield:

Annualized yield = (premium per share ÷ capital per share) × (365 ÷ DTE)

For a cash-secured put, capital per share is the strike price, because that is the cash you must set aside. For a covered call, capital per share is the underlying share price, because the stock you are committing is what the premium is earned against. Premium is the mid price, (bid + ask) ÷ 2, whenever quotes are live, and the last traded price when they are stale.

A worked example, using illustrative numbers chosen for arithmetic clarity rather than drawn from any real contract. Suppose a put with a $50 strike is quoted at a $0.40 mid with 7 days to expiry. The period return is 0.40 ÷ 50 = 0.80%. Scaling by 365 ÷ 7 gives roughly 41.7% annualized. That is exactly the sort of figure that should make you suspicious.

The suspicion is warranted, because annualizing a seven-day return quietly assumes you can repeat that trade about fifty-two times a year at the same yield. You cannot. Implied volatility rises and falls, so the premium behind this week's 0.80% may not exist next week. Assignment interrupts the sequence entirely: once you own the shares you are no longer collecting that premium, you are holding a stock that just fell. And the causality runs the wrong way for comfort, because the highest annualized yields appear precisely where the market is pricing the greatest uncertainty.

Treat annualized yield as a comparison tool across contracts, not a return forecast. It makes a 7-day contract and a 45-day contract commensurable on one axis, and says nothing about whether either is worth trading.

The liquidity filters and why they matter more than they look

An options quote has two prices: the bid, what a buyer will pay you, and the ask, what a seller will charge you. The gap between them is the bid-ask spread, a real and immediate cost that comes out of your result before the market has moved at all.

When the market is open we require that (ask − bid) ÷ mid is no greater than 0.35 and that volume plus open interest is at least 20. When the market is closed or quotes are stale, meaning older than 30 minutes, we fall back to the last traded price, back-solve implied volatility from it, and require open interest of at least 20. Separately, implied volatility above 3.0, or 300%, is rejected outright.

Be clear about how permissive these thresholds are. A 35% cap sounds restrictive until you express it in dollars: on a contract with a $0.40 mid it permits a 14 cent spread, more than a third of the premium collected. Options spreads are routinely far wider in percentage terms than equity spreads, so the cap sits where it does to avoid discarding most of the market, not because 35% is an acceptable cost. Requiring volume plus open interest of 20 is likewise a very low bar: it removes the completely untradeable and essentially nothing else.

Clearing these filters means a contract is not obviously broken. It does not mean it is genuinely liquid. Check the live spread in your broker's chain and decide whether the premium still makes sense after paying it.

IV and IV Rank

Implied volatility is the amount of future movement the market has priced into an option, derived by solving backward from the option's price for the volatility input that would justify it. High IV means options are expensive, which is attractive to a seller, but they are expensive because participants genuinely expect a wider range of outcomes. You are being paid more because more can go wrong. Premium-selling screens naturally surface high-IV names, and it is easy to read a large premium as an opportunity rather than as a price for risk.

Raw IV is hard to interpret on its own, because volatility levels are specific to each security. A biotech that habitually trades at 90% IV is not unusually stressed at 90%; a utility at 45% almost certainly is. IV Rank places today's reading inside that security's own recent history:

IV Rank = ((current IV − 1-year minimum HV) ÷ (1-year maximum HV − 1-year minimum HV)) × 100

Here HV is historical volatility, the standard deviation of log returns multiplied by the square root of 252, measured on a rolling 20-day basis across the past year. The result places current implied volatility on a 0 to 100 scale against the range the underlying has actually realized, which distinguishes a stock that is always volatile from one that is unusually volatile right now. What IV Rank does not do is explain why. An elevated reading caused by a pending court ruling and one caused by general market stress produce the same number and call for very different judgments.

How the ranking score works

Once a contract clears every hard filter it is scored so candidates can be ordered. The composite score is:

Score = (annualized yield × (1 − band_penalty) + IV bonus) × DTE fit

The score starts from annualized yield, its primary driver. A band penalty reduces that yield when delta falls outside the preferred band, so a high-yielding contract at an uncomfortable delta is discounted against a slightly lower-yielding one sitting where we want it. A small IV bonus of min(IV, 1.0) × 0.10 is added, mildly favoring richer volatility while capping the contribution so extreme readings cannot dominate. The result is multiplied by a DTE fit factor, max(0.15, 1 − (|DTE − target| ÷ 14)²), which decays quadratically as the expiry drifts from the target and never falls below 0.15. The default target is 7 days, with a search window of 14 days either side, clamped to the overall 1–62 day range.

Worth stating explicitly: there is no minimum dollar premium threshold. Selection is driven entirely by the liquidity filters, the delta band, and this composite score, so a contract paying a few cents can appear if it is liquid, correctly positioned on delta, and close to the target expiry. Whether that premium justifies the spread is a judgment the screen leaves to you.

What the screen does not tell you

A ranked list is persuasive in a way that is not always earned. Several things that matter a great deal to a premium-selling trade are absent from our calculation.

It does not know your tax situation. Premium and assignment are taxed differently across jurisdictions and account types, and every yield figure we display is pre-tax.

It does not know whether you actually want to own the stock. That is the real precondition for selling a cash-secured put, and no scanner can evaluate it. If you would be content to own the shares at the strike, assignment is an acceptable outcome. If you would not, you are holding a position whose most likely bad outcome is one you have no plan for.

It ignores upcoming earnings and dividends. Both materially change assignment risk. An earnings release inside the contract's life can produce a gap far larger than the delta-implied odds suggest, and it is often exactly why the premium looked attractive. On the covered call side, an approaching ex-dividend date raises the chance of early assignment, because a call holder may exercise to capture the dividend. Neither date is checked by the screen.

It does not model early assignment. The contracts we screen are typically American-style, so the buyer can exercise at any point before expiry rather than only at the end. The delta-as-probability shorthand describes the position at expiration; assignment can arrive sooner, on terms you did not choose.

The risk-free rate is a fixed assumption. We use 4.3% rather than reading a live yield curve or matching the rate to each contract's maturity, and that input feeds the Black-Scholes delta calculation. The effect is small for short-dated options, but it is an approximation rather than a market observation, and it drifts out of date as rates move.

Where the numbers come from

Option chains and implied volatility come from Yahoo Finance. Delta is not taken from that feed; it is computed locally with Black-Scholes from the supplied implied volatility, the fixed 4.3% risk-free rate, the strike, the underlying price, and the time remaining. The scan runs Monday through Friday at 9:35 AM and 3:35 PM Eastern, and on demand when stored results are more than 30 minutes stale. Quotes move continuously between those runs, so a row that looked appealing at the morning scan may be priced quite differently by the time you act on it.

For the full inventory of data sources and known gaps, see our data and methodology page. Terms used above are defined in the glossary.

Continue reading