How to Backtest a Covered Call Strategy (and the Wheel)

To backtest a covered call strategy, write the rule as one explicit sentence (ticker, strike as a delta or a percentage out of the money, days to expiration, roll schedule, what happens on assignment), run it on split- and dividend-adjusted history for the underlying, charge about 0.1% per trade on both legs, and compare the result against simply holding the same shares over the identical window. The comparison is the whole test. Premium collected in isolation always looks good; premium measured against the upside the cap gave away is the number that tells you something.

Call writing is the most popular income structure retail investors actually run, and it is also the one most often adopted on vibes. The premium arrives every month, it feels like free money, and nobody checks what happened during the quarters the stock ran 20% past the strike. A backtest is the cheap way to see both halves of the trade before you put six figures of stock behind it.

Why a covered call backtest is different from a stock backtest

A plain stock backtest has one moving part: when you are in and when you are out. A covered call backtest has four, and each one moves the answer. The strike distance sets how much premium you collect and how often you get called away. Days to expiration set how many times a year you repeat the cycle. The roll rule decides whether you close early, let it expire, or roll up and out when the stock rallies. And assignment decides whether you keep compounding the shares or keep getting knocked out of them at the strike.

That is why two covered call backtests on the same ticker over the same 20 years can disagree completely. One sold 10-delta calls 45 days out and barely gave up upside; the other sold at-the-money calls weekly and effectively converted a growth stock into a bond with equity downside. Neither is wrong. They are different strategies wearing the same name, which is why the parameters have to be written down before the test runs.

How do you backtest a covered call strategy?

Specify the five inputs, run them on long adjusted history with costs, and benchmark against buy-and-hold. Most of the work is in the specification, because an unspecified rule silently gets the most flattering interpretation. Here is what has to be pinned down:

Once those are fixed, the mechanics are the same as any strategy test: a long enough window, honest costs, and a benchmark. Twenty years or more is the right target because it forces the rule through 2008, 2020, the 2022 drawdown, and at least one melt-up, and call writing behaves very differently in each. The reasoning behind that window length is covered in how long you should backtest a trading strategy.

Do covered calls outperform buy and hold?

Usually not on total return over long horizons, and the reason is structural rather than bad luck. Selling a call caps your gain at the strike plus the premium, while the downside is only cushioned by that premium, not removed. Markets deliver a large share of their long-run return in a small number of explosive months, and a call-writing program is short exactly those months. What call writing did historically deliver is a smoother ride: more frequent small wins, shallower swings, and better behavior in flat or choppy markets.

So the honest framing is a trade, not a free lunch. You are exchanging some long-run upside for income and lower volatility. Whether that trade suited a particular ticker in a particular decade is an empirical question, which is what the backtest answers. Run it and read two numbers side by side: total return against worst drawdown. If a covered call rule trailed plain ownership by two points a year while cutting the worst drawdown by a third, that may be a trade you want. If it trailed by six points and barely moved the drawdown, it is not.

How far out of the money should covered calls be?

There is no universally correct distance, which is precisely why this parameter is worth testing instead of copying from a forum. Closer strikes collect more premium and get called away far more often; further strikes keep more upside and collect less. The complication is volatility: a strike 10% above the price is a long way out in a quiet market and nearly at the money in a violent one, which is why delta-based rules behave more consistently across regimes than fixed percentages.

The practical method is to test two or three variants on the same window and read the spread rather than hunting for a single optimum. If 20-delta, 30-delta, and 40-delta all land in the same neighborhood, the strategy is robust and you can pick on comfort. If the result swings wildly between them, you have found a curve-fit, not an edge, and the same logic applies as in walk-forward analysis: a parameter that only works at one exact setting will not survive contact with the next decade.

Should you sell covered calls through earnings?

Test it both ways, because it is one of the few covered call parameters with a genuinely lopsided profile. Implied volatility, and therefore premium, is highest going into an earnings report, which is exactly why the calls are expensive: the stock can gap well past your strike overnight. Skipping the earnings cycle collects less premium across the year but avoids the handful of events most likely to cap a large move. Including it collects the fat premium and accepts occasional assignment right when the stock jumps.

Published options research has generally found that avoiding earnings improved call-writing results on broad baskets, but that is a tendency across many names, not a rule about the one stock you hold. Add an "exclude the week of earnings" variant to your backtest and let your own ticker answer.

How do you backtest the wheel strategy?

State the whole cycle as a single rule, then test it end to end: sell cash-secured puts on the ticker at a set delta, take assignment if the put finishes in the money, sell covered calls on the assigned shares until they are called away, then start over. Three details decide whether the result is honest. First, the cash reserved against the puts has to earn something realistic while it sits, or the test flatters the strategy on capital efficiency. Second, the strike distance on both legs matters as much as it does for a standalone covered call. Third, and most important, the test has to include a sustained decline.

The wheel looks superb in flat and rising markets and reveals its real risk in a long slide, when you are assigned shares that keep falling and then write calls below your cost basis. Any wheel backtest that starts in 2010 or 2021 has skipped the only condition that matters. A poor man's covered call has the same requirement for a different reason: because the long leg is a dated option that bleeds time value rather than shares you can hold forever, its result is far more sensitive to the length of the flat stretch it has to sit through.

Can you backtest covered calls on Robinhood or Fidelity?

Not properly on either, though for different reasons. Robinhood has no historical strategy backtesting at all, as covered in does Robinhood have backtesting. Fidelity does have a real free Strategy Testing tool, but it is built around technical templates on equities rather than options structures, which is the distinction drawn in does Fidelity have backtesting and on the Fidelity alternative comparison. Broker-side options tools mostly compute what a trade would pay from today forward, which is a calculator, not a backtest.

The genuine options backtesting market splits in two. Chain-level data vendors sell historical quotes for every strike and expiry, and they are the right choice if you need per-strike bid and ask reconstruction. Everything else is either a scripting environment where you code the structure yourself or a plain-English bench like this one, where the structure is described in a sentence and the assumptions are printed instead of hidden.

The traps that make covered call backtests lie

• No benchmark. A chart showing steady premium income with no buy-and-hold line beside it is marketing. The cap is invisible unless the thing you capped is drawn.
• A window with no melt-up. Covered call rules look best in flat markets. A test that skips 2013, 2019, 2021, or 2023 never measures the cost of the cap.
• Price-only data on a dividend payer. Call writing is usually run on shares that pay. Drop the dividends and both the strategy and the benchmark are understated, which distorts the gap between them.
• Ignoring the spread. Options spreads are wider than stock spreads, and a weekly program pays them 52 times a year. Costs charged at zero can flip a losing rule into a winning one on paper.
• Ignoring taxes. In a taxable account, premium is generally short-term income and assignment realizes gains on the shares, so a program that looks efficient before tax can be considerably less so after. None of that appears in a backtest, and all of it appears on the forms you reconcile at filing time.
• Too few cycles. Eleven assignments across a decade is an anecdote. Count the cycles the way you would count trades in any other strategy test.

Backtest a covered call rule without the options spreadsheet

Done by hand, this is an options data subscription, a strike-selection formula sheet, and a weekend of debugging roll logic, with an error hiding in every step. That is the work a covered call backtest on Agenttrading compresses. Describe the structure in plain English, such as "sell monthly 30-delta covered calls on SPY and roll at expiration" or "run the wheel on QQQ at 25 delta", and the bench restates the rule as an explicit card before anything runs, so you can see exactly what it understood.

It then tests the structure on 20+ years of split- and dividend-adjusted daily history with a 0.1% cost per trade charged by default, prints the assumptions strip covering strike selection, roll schedule, and assignment handling, shades the worst drawdown along with its recovery time, plots the result against holding the shares, and stamps an honest verdict: HELD UP, MIXED, or UNDERPERFORMED. That last verdict stays as prominent as the first, because a bench that cannot tell you your income strategy trailed plain ownership is not analysis.

Other structures, including cash-secured puts and vertical spreads, are covered on options backtesting. The adjusted record every run uses is described under historical stock data, the drawdown a call-writing program still leaves you with is explained in investment risk analysis, and if the reason you are writing calls is income, compare the approach against dividend backtesting on the same tickers.

Agenttrading executes no trades and connects to no brokerage. It shows you what the record says about your covered call rule. What you do with that is your call.

Past performance does not guarantee future results. For educational and informational purposes only. Not financial advice. Consult a licensed advisor.