Did covered calls work on AMD? Not against a +55.7% stock
From Aug 2016 to Oct 2026 I sold a 0.30-delta call every 30 days against 100 AMD shares. The calls collected 27.3% of the share price a year in modeled premium and gave back 58.6% a year when they were closed or settled. The net from the options was -31.4% a year.
The covered-call position still compounded at +27.7% a year. Holding did far better at +55.7% a year, so the calls cost 28.0 pts a year relative to doing nothing.
Volatility explains the premium: typical volatility is 48% against a 30% median, so the average 0.30-delta strike sat 10.8% above the price. The drift explains the rest. A stock compounding at +55.7% a year keeps running through its strikes: the calls finished in the money in 32% of months, a little more than a 0.30-delta call would on a flat stock, and the months when they did were the expensive ones.
April 2026, the call that cost 78.8% of the share price
AMD's worst month for a call seller began in Apr 2026. The stock rose 92% over those 30 days, and closing the call that had been sold at the start cost 78.8% of the share price. That is more than the 27.3% a year of premium the calls brought in on average.
The same run is still sitting in the partial year. So far in 2026 AMD is up +162.5% and the covered-call version is up +55.7%, a difference of -106.8 pts, already the widest gap in the table.
| Year | Stock | Covered call | Difference (pts) |
|---|---|---|---|
| 2016 partial | +70.0% | +38.1% | -31.9 pts |
| 2017 | -9.3% | -3.3% | +6.0 pts |
| 2018 | +79.6% | +7.0% | -72.6 pts |
| 2019 | +148.4% | +142.9% | -5.5 pts |
| 2020 | +100.0% | +59.6% | -40.4 pts |
| 2021 | +56.9% | +29.7% | -27.2 pts |
| 2022 | -55.0% | -49.8% | +5.2 pts |
| 2023 | +127.6% | +88.0% | -39.5 pts |
| 2024 | -18.1% | -14.7% | +3.4 pts |
| 2025 | +77.3% | +28.5% | -48.8 pts |
| 2026 partial | +162.5% | +55.7% | -106.8 pts |
Read 2020 with care: the model estimates option prices from recent price swings, which ran far above what options actually cost in the weeks after the March crash, so it overstates the premium collected then. Real buy-write funds trailed the S&P 500 in 2020.
Calendar-year total returns; the covered call is averaged over every start date, as in the chart above.
Among full years the worst was 2018. The stock returned +79.6% and the covered call +7.0%, a difference of -72.6 pts. 2025 was next at -48.8 pts (+77.3% against +28.5%), then 2020 at -40.4 pts (+100.0% against +59.6%; the model likely flatters the calls that year, so the true gap was probably wider) and 2023 at -39.5 pts (+127.6% against +88.0%).
The odd one out is 2019. The stock returned +148.4%, the biggest full year in the table, and the covered call still made +142.9%, only 5.5 pts behind. A big rally did not guarantee a big gap, so I would not reduce AMD to "calls hurt in strong years".
The three years the calls came out ahead
They won in 2017, 2022 and 2024, and those are the only three years AMD lost money. In 2017 the stock fell 9.3% and the covered call only 3.3%, a difference of +6.0 pts. In 2022 it was -55.0% against -49.8% (+5.2 pts), and in 2024 -18.1% against -14.7% (+3.4 pts).
So the pattern is tidy: ahead when the stock fell, behind in every year it rose. But the lead was never more than +6.0 pts, while the deficit reached -106.8 pts. The worst fall, from Nov 2021 to Oct 2022, was -65% for the stock and -63% for the covered call.
AMD among the semiconductors
Over Oct 2021 – Oct 2026, the window every ticker I track shares, AMD's calls netted -32.3% a year, which ranks 6 of 8 semiconductors; only two did worse. The closest comparison is NVDA, another chip designer, whose calls netted -20.8% a year over the same window. QCOM lost the least in the sector, at -11.3%.
AMD today, and which strikes held up
On 2026-10-15 AMD closed at $562 and the model put its volatility at 64%. That prices a 30-day call at a $631 strike, +12.3% above the share price, for $13.96 a share, or $1396 a contract (2.48% of the price). Those are modeled figures, not quotes, and 100 shares cost $56,223.
At that volatility a strike 10% above the price would be a 0.34-delta call, so on AMD "far out of the money" at +10% is really a standard call. In the backtest it netted -30.5% against -31.4% for the 0.30-delta setup, a gap of under a point. The 0.15-delta call lost the least at -20.2%, finishing in the money in 19% of months. The 0.40-delta call lost the most, at -34.4%. The calculator lets you try your own strike, and the covered-call ETF guide covers funds that sell calls for you.
AMD wheel strategy backtest — cash-secured puts plus covered calls on the same stock.
Covered calls on similar tickers: