Does the Wheel Strategy Make Money? The S&P 500 Since 1990, Tested

The wheel is one of the most popular options strategies on Reddit and YouTube. You sell a put, and if you're made to buy the shares, you sell calls on them until they're sold. Then you start again.

It's sold as steady income, and the pitch is easy to like. You get paid every month, and the worst case is owning a stock you wanted anyway. So we ran it on the S&P 500 every month from January 1990 to October 2026, 441 months in all, and checked what it actually made.

How the wheel works

The wheel has three steps, and it loops.

1. 1Sell a cash-secured put. You promise to buy the shares at a set price, the strike, if they fall below it by expiry. You get paid a premium up front, and you keep cash aside to cover the purchase.
2. 2If the shares stay above the strike, the put expires and you keep the premium. Sell another put. If they fall below it, you're assigned: you buy the shares at the strike.
3. 3Now you own shares, so sell a covered call. You promise to sell them at a strike above today's price. If they rise past it, they're called away and you go back to step 1. If not, sell another call.

Every month brings in a premium, whichever step you're on. In our test the typical month paid 0.89% of the account. That works out to roughly 13% a year, which is why the wheel looks like income.

How we tested it

We ran the wheel on the S&P 500 with one-month options, trading on each month's standard expiry day, the third Friday. Puts were sold 2% below the index and calls 2% above it, a common middle-of-the-road choice. Cash waiting to buy shares earned the Treasury bill rate, and shares earned their dividends.

There's no free record of every option price back to 1990, so we priced the options with a standard model, using the VIX that day as the market's own guess of how much the S&P 500 would move. Then we checked the model against two indices Cboe builds from real option trades: one that sells 2% out-of-the-money puts every month, and one that holds the index and sells 2% out-of-the-money calls.

We tuned the model so each one matched Cboe's index over the whole period. Year by year, the typical gap was 1.5 points. That's close enough to trust the big picture, not each single month.

What $10,000 became

The wheel made 7.8% a year. The S&P 500 made 11.1%. That gap of 3.3 points a year doesn't sound like much, but over almost 37 years it's the difference between $157,000 and $472,000.

The wheel did fall less. Its worst drop was 38%, in 2008-09, against 55% for the index. So it's a smoother ride. But it isn't a safe one: a 38% fall is a bear market by any measure.

Why it wins most months and still loses

The wheel made money in 370 of 441 months. The S&P 500 rose in only 289. So by win rate, the wheel wins easily.

The problem is the size of the wins and losses. The wheel's average winning month made 1.6%. Its average losing month lost 4.0%, more than twice as much.

That's built into the trade. When you sell an option, the most you can make is the premium plus a small move. In the 57 months the S&P 500 rose more than 5%, it gained 7.8% on average and the wheel made 2.7%. In the 35 months it fell more than 5%, it lost 9.5% and the wheel still lost 6.9%.

The chart shows the shape. On the left, when the market falls, the dots sit close to the dashed line: the wheel loses almost as much. On the right, when the market rises, they flatten out at a few percent. You keep the falls and give away the rallies.

March 2020: the wheel's worst month

On 21 February 2020 the S&P 500 stood at 3,338, and the wheel sold a put at 3,271. Four weeks later, on 20 March 2020, the index was at 2,305. The put was assigned: the wheel bought at 3,271 something worth 2,305. It lost 28.5% that month.

Then the rule said to sell a call 2% above the price, at 2,351. The market bounced hard. By 17 April 2020 the index was at 2,875, so the shares were called away at 2,351. The wheel made 6.5% in a month when the S&P 500 made 25%.

So it bought near the top of the fall and sold near the bottom, and the loss was locked in. The wheel finished 2020 −7.1% while the index made +17.3%.

The wheel is just selling puts

Here's the part most wheel videos skip. Owning shares and selling a call on them pays off exactly like selling a put at the same strike and keeping cash. It's a rule of option pricing called put-call parity.

Both trades have the same shape. If the price ends above the strike, you make a small, fixed amount. If it ends below, you lose about as much as the shares fall. So when the wheel "switches" from puts to calls, the bet hardly changes. You're short a put the whole time.

The numbers agree. Cboe's index that only sells 2% out-of-the-money puts, on real prices, made 7.0% a year over the same dates. The wheel made 7.8%. The small extra came from the call months, when the wheel held shares and so more of the market's rise. Neither got close to the index.

Why do option sellers earn anything at all? Options usually cost a bit more than the moves they end up paying for. Researchers call it the volatility risk premium. It's real, but it's pay for taking crash risk, and in a rising market it's smaller than what you give up.

Does "never sell below your cost" fix it?

Many wheel traders follow one extra rule: once you're assigned, never sell a call below the price you paid. That way you can't lock in a loss the way the plain wheel did in 2020.

It helped. With the rule, $10,000 became $290,945, or 9.6% a year. In 2020 it held the shares through the rebound and finished the year +15.5%.

But look at why. The rule keeps you holding shares after every fall, so it spent 51% of all months holding shares, against 26% for the plain wheel. It made more because it became more like buying the index, and it fell more too: 50% at the worst, close to the index's 55%.

After the 2000 crash it got stuck. It held shares for 78 months in a row, from October 2000 to April 2007, waiting for the index to get back to 1,436. At times its call strike sat 70% above the price, and calls that far away pay almost nothing. For six and a half years it was a buy-and-hold fund that called itself an income strategy.

Even then it trailed the index by 1.5 points a year. The rule changes when you lose, not whether you give up the big months.

When the wheel does win

The plain wheel beat the S&P 500 in 13 of the 36 years from 1991 to 2026. Those were flat or falling years, when the premiums were worth more than the index's gain.

The pattern is clear. In 1994, 2000, 2002 and 2022 the index went nowhere or fell, and the premiums coming in kept the wheel ahead. In 1995, 2013, 2019 and 2020 the index rose 17% or more, and the wheel kept only part of it.

So the wheel isn't a way to beat the market. It's a bet that the market goes sideways. And over long periods, the S&P 500 has mostly gone up.

The last ten years

Maybe 1990 is too far back. From 21 October 2016, $10,000 in the wheel became $18,607. With the cost rule it became $29,063. In the S&P 500 it became $42,186.

This was a strong decade for stocks, which is the wheel's worst case. Cboe's put-selling index, on real prices, turned $10,000 into $19,252 over the same dates.

Running it yourself

Our test used the index. If you wheel in a brokerage account, a few things change.

• You need a lot of cash. One option contract covers 100 shares. At SPY's close of $735.29 on 6 November 2026, one cash-secured put needs about $73,529 set aside. That's why many people wheel cheaper single stocks instead.
• Single stocks pay more and risk more. A stock's options cost more than the index's because one company can halve overnight. If it does, you're assigned on the way down, and you're stuck selling calls far below what you paid.
• You need shares to be delivered. SPX and XSP options settle in cash, so you never get the shares and there's nothing to sell calls on. The wheel needs options that deliver shares, like SPY's.
• You can be assigned early. SPY options can be exercised any day before expiry, most often just before SPY pays a dividend. Our index test didn't include that.
• Costs add up. Each trade pays a commission at most brokers and the gap between the bid and ask price. Selling an option every month adds up to twelve or more trades a year.
• Taxes vary by country. In many places, including the US, premiums are taxed as short-term gains, often at a higher rate than long-term gains on shares you simply hold. Check your own country's rules.

What this test can't tell you

• The option prices are modelled, not traded. They're checked against Cboe's real-price indices, which the model matched over the whole period and within about 1.5 points in a typical year.
• We used one version: one-month options, 2% from the price. Strikes closer to the price collect more premium. The at-the-money version made 8.3% a year, still well behind the index.
• Options settle at the expiry-day close. Real index options settle on that morning's opening prices, which made March 2020 a little different in Cboe's own index.
• It's one market. Wheeling single stocks gives very different results, mostly depending on which stocks you pick.

So does the wheel strategy make money?

It does. Over almost 37 years it turned $10,000 into $156,536, and it won in most months. But it made less than simply owning the index, by 3.3 points a year, and it still fell 38% in 2008.

The premium feels like free money because it arrives every month. It isn't. It's what you're paid for giving up the market's best months and keeping most of its worst ones. If you want the S&P 500's return, an index fund gets you more of it, for less work.