Investing in the stock market requires a near superhuman ability to withstand pain. That’s the conclusion of a recent report by investment researcher Michael Mauboussin.

Mauboussin surveyed all of the stocks trading on U.S. exchanges over a 40-year period, between 1985 and 2024. He found that the median stock experienced a decline of 85% at one point or another. That would have tried any investor’s patience. But worse yet, more than half of these stocks never fully recouped their losses. The median stock recovered to just 90% of its prior highwater mark. Even among those that were able to reclaim their prior highs, it was a long process—about five years, on average.

Those numbers apply just to the median stock, but suppose you’d had above-average stock-picking skills. Then how would things have turned out? Even if you’d had the foresight to pick the 20 best performing stocks over that 40-year period, they still would have delivered an agonizing drawdown of 72%, on average.

It’s hard to remember, but Apple dropped 83% at one point. Nike once lost 66%. Even Nvidia, which was the best performing stock over the past 20 years, through 2024, gave up more than 90% at one point. And most notably, Amazon was, at one point, down 95% from its prior high.

It’s well known that the stock market is unpredictable, but these numbers provide additional insight into the market’s dynamics.

In general, the stock market is rational. Over the long term, stock prices do move in tandem with corporate profits. When a company’s earnings increase, often its share price does too. The problem, though, is that prices are only rational sometimes. Very often, stock prices disconnect from corporate earnings, and that gap can be significant. This was first proven empirically in the 1970s by Daniel Kahneman and his longtime collaborator, Amos Tversky.

In 1974, they published a paper titled “Judgment under Uncertainty: Heuristics and Biases.” It was one of the first papers in the nascent field of behavioral finance, and what they found was that investors exhibit an “availability heuristic.” That is, they tend to rely on the information that is most available. That’s a problem because the information that happens to be most available isn’t necessarily the information that is the most accurate or even relevant. Often, in fact, the information that happens to come to mind is, as Kahneman and Tversky put it, the information that’s most vivid. Extreme information or news, in other words, ends up being most memorable, and that’s what tends to drive decision-making.

Later research built on this idea. In the mid-1980s, economists Werner De Bondt and Richard Thaler published a paper titled “Does the Stock Market Overreact?” What they found is that stock prices definitely do overreact. A casual observer might find that conclusion to be intuitive, but De Bondt and Thaler were able to prove and to quantify it. They looked at stocks that had either outperformed or underperformed by a significant margin in recent years, then examined their performance over subsequent years.

What they found was that stocks exhibiting extreme performance tended to reverse course. Strong performers ended up lagging, and weaker stocks often went on to outperform. The implication: Investors systematically overreact to news, driving stocks too far in one direction or the other.

It’s for these reasons that Warren Buffett has scolded investors for their short-term thinking. During the market slump earlier this year, Buffett commented: “If it makes a difference to you whether your stocks are down 15% or not, you need to get a somewhat different investment philosophy,” adding that, “People have emotions, but you got to check them at the door when you invest.”

On this point, Buffett’s late partner, Charlie Munger, was more blunt: “If you’re not willing to react with equanimity to a market price decline of 50% two or three times a century,” Munger said, “you’re not fit to be a common shareholder…”

In making these comments, Buffett and Munger weren’t being preachy; they knew from decades of experience what De Bondt and Thaler had found in the lab—that the market is often irrational, and that it can irrationally punish even the best of companies’ stocks. About 10 years ago, in fact, Buffett noted that Berkshire Hathaway—his own company—had seen its stock drop 50% on three separate occasions over the years. “Someday, something close to this kind of drop will happen again, and no one knows when,” he added.

What complicates the equation is that the stock market sometimes makes an extreme move that is justified and not just the product of emotional overreaction. In January, for example, when Nvidia shares dropped 17% in a single day, it was in response to a potentially serious competitive threat. Similarly, Apple shares are down about 14% this year in response to some real concerns, including slowing iPhone sales, a weak position in AI and the impact of tariffs. Similarly, Alphabet—parent company of Google—has seen its stock underperform this year as AI tools like ChatGPT chip away at its market share.

Of course, no one knows where any of these stocks will go next. Whether it’s in response to tangible news, emotional overreaction, or some combination of the two, stock price movements will always be unpredictable. That’s a reality, but there is a way to mitigate it.

As noted earlier, Mauboussin found that the median stock suffered a drawdown of 85%, and even the best stocks saw a drawdown of 72%. But the S&P 500, a broad market index, never experienced a drop of that magnitude during the time period studied. The worst drop experienced by the S&P was 58%—terrible but far less bad than the experience of those individual stocks. This, in my view, is a key reason to avoid picking stocks and instead to invest using index funds.

In addition to helping manage risk, index funds offer another powerful benefit. As noted earlier, the median stock in Mauboussin’s study never fully recovered after experiencing a decline. But on average, stocks definitely do recover and significantly surpass their prior highwater marks. On average, stocks gained nearly 340% when they bounced back.

Why the distinction between the median and the average? It’s a technical point but an important one. The results of the median stock are unaffected by the performance of outliers. Outliers, however, are the main driver of the market’s overall return, and the average return benefits from the impact of outliers. When a stock like Nvidia gains 88,000%, as it has over the past 20 years, that pulls up the average. And while an investor who was exceptionally forward-looking, bordering on clairvoyant, might have purchased Nvidia shares—and held onto them—over the past 20 years, most wouldn’t have been that fortunate. The S&P 500, however, has owned Nvidia—along with Apple, Amazon and all of the market’s other biggest gainers—over the years.

The bottom line: Picking stocks can be entertaining, but according to the data, that turns out not to be the most reliable way to build wealth. Index funds, because they cast a wide net and are unaffected by emotion, allow investors to benefit from the growth of exceptional stocks while also limiting the impact of drawdowns.​