Can You Beat the Market? The Honest Numbers on the S&P 500

What “beating the market” actually means

In most American conversations, “the market” means the S&P 500. It is an index of roughly 500 large U.S. companies, chosen by a committee at S&P Dow Jones Indices and weighted by the market value of their freely tradable shares. It took its 500-stock form on March 4, 1957, and today covers around four-fifths of the value of all U.S.-listed companies. Because it is weighted by size, a move in Apple or Nvidia shifts the index far more than a move in its smallest member.

Beating it is a narrower claim than making money. A fair comparison has four conditions:

• The same period. Your 2015–2025 against the index's 2015–2025, not against whatever stretch makes you look good.
• Total return on both sides. The index level quoted on the news excludes dividends. From February 25, 2005 to October 1, 2026 the S&P 500 price index rose 6.3-fold, while the SPDR S&P 500 ETF (SPY), with dividends reinvested, grew 9.3-fold. That is a difference of about 2 percentage points a year.
• After costs and taxes. Fund fees, trading spreads and taxes reduce your return, while the index pays none of them. In the U.S., net short-term gains on positions held a year or less are taxed as ordinary income.
• Comparable risk. A leveraged bet on a handful of tech stocks can top the index in a boom and lose half its value in the next crash. A fair comparison accounts for the extra risk taken along the way.

The SPIVA report: what percent of mutual funds beat the S&P 500

S&P Dow Jones Indices has published its SPIVA (S&P Indices Versus Active) scorecard for about 25 years. It compares actively managed funds with their benchmarks after fees, and it keeps funds that were merged or liquidated in the count as failures, which matters more than it sounds.

Turn the 20-year number around and about 7% of large-cap funds beat the index over two decades, roughly one in fourteen. Only about 37% of the funds that existed in mid-2006 were still around in 2026. The rest were closed or folded into other funds, usually after poor results, so a list of today's funds with 20-year records makes the industry look far better than it was. Our article on survivorship bias covers that trap in detail.

Single years swing: 2025 was the fourth-worst year in the scorecard's history, while 2024 and the first half of 2026 were milder. The pattern across horizons is steadier. For a professional with analysts and data feeds, beating the S&P 500 in a given year is roughly a coin flip at best; doing it for 15 or 20 years is rare.

The DALBAR study and the investor behavior gap

SPIVA grades the funds. DALBAR's annual Quantitative Analysis of Investor Behavior (QAIB) tries to grade the people who own them, estimating investor returns from money flowing in and out of funds. The 2026 edition, released in April, found that the average equity fund investor earned 17.16% in 2025 against 17.88% for the S&P 500. That 0.72-point gap was the smallest since 2012. A year earlier it was 8.48 points: investors earned 16.54% while the index returned 25.02%. Bond fund investors fared worse in 2025, earning 2.41% while the Bloomberg U.S. Aggregate index returned 7.30%.

Morningstar's “Mind the Gap” study measures the same problem with dollar-weighted returns and reaches a similar conclusion. In the ten years through 2025, investors in U.S. mutual funds and ETFs earned 8.7% a year while the funds themselves returned 9.9%. The 1.2-point gap equals about 12% of the funds' total return, lost to the timing and size of investors' purchases and sales.

Nobody keeps a census of private brokerage accounts, so there is no exact answer to how many investors beat the S&P 500. Both studies point the same way, though: the typical investor earns less than the funds they hold, and most of those funds earn less than the index.

Why stock picking is so hard: the Bessembinder study

In 2018 Hendrik Bessembinder of Arizona State University published “Do Stocks Outperform Treasury Bills?” in the Journal of Financial Economics. He looked at every common stock in the CRSP database from 1926 to 2016, issued by 25,332 companies.

• Only 42.6% of those stocks beat one-month Treasury bills over their listed lifetimes, dividends included. Roughly four in seven did worse than cash.
• More than half had negative lifetime returns. The single most common outcome, with returns rounded to the nearest 5%, was a loss of 100%.
• The best 1,092 companies, 4.3% of the total, accounted for all of the net wealth the stock market created, close to $35 trillion by the end of 2016. The other 96% together roughly matched T-bills.
• Just 90 companies produced more than half of that wealth.

This is why an index is hard to beat. It owns everything, so it is guaranteed to hold the few giants that do most of the work. A portfolio of 10 or 20 stocks will more often than not miss most of them, which is the paper's own explanation for why poorly diversified active strategies usually trail market averages.

Our test: 33 famous stocks against the S&P 500, 2005–2026

Our price database stores daily total return series: adjusted for splits and spin-offs, with dividends reinvested. The export we used for this article contains 33 individual U.S. stocks with real trading data on February 25, 2005, the first day of our SPY series. We held each one from that day to October 1, 2026, about 21.6 years, with dividends reinvested, and compared it with SPY over exactly the same dates. SPY turned $1 into $9.35, or 10.9% a year.

Result: 16 of the 33 stocks beat the index, 17 did not. The closest call went the other way: Walmart returned 10.87% a year against SPY's 10.90% and missed by 0.04 points. Cisco, the weakest of the winners, was ahead by less than 0.3 points a year, while Berkshire Hathaway's class B shares trailed by about 0.6 points.

Three things stand out. The median stock in the group ended almost exactly where the index did (9.28x against 9.35x). The simple average of all 33 is about 70x, but only because Nvidia multiplied 1,050 times; Nvidia, Netflix and Apple together produced 76% of the combined gains of all 33 stocks. That is Bessembinder's skew in miniature. And size and reputation in 2005 protected no one: General Electric, which Bessembinder ranks among the five biggest wealth creators in U.S. market history through 2016, returned only about 3x in our data, even counting the GE HealthCare shares spun off in 2023, and Citigroup lost more than half of its value even with dividends.

Why the time horizon matters

Even the benchmark can disappoint over a decade. Using month-end closes of the S&P 500 price index since January 1962, we computed every rolling 10-, 15- and 20-year window and its annualized return.

Ten years guarantees nothing. About one in eleven 10-year windows ended with the index lower on price, the worst being February 1999 to February 2009, the stretch covered in our article on the lost decade. The best 10-year window ended in August 2000, near the top of the dot-com bubble. Longer windows narrow the range but do not cancel inflation: from July 1967 to July 1982 the price index rose just 13% while the consumer price index rose 2.9-fold, a loss of about 61% in purchasing power before dividends, the era of 1970s stagflation.

The horizon matters for judging skill too. A manager who beat the index over one decade may simply have been positioned for that decade, which fits SPIVA's finding that the share of winners shrinks as the measuring period lengthens.

Is the stock market a zero-sum game?

In absolute terms, no. Companies earn profits, pay dividends and grow. The S&P 500 price index stood at 70.96 on January 2, 1962 and at 7,666 on October 1, 2026, about 108 times higher, or 7.5% a year before dividends. Owners as a group became much richer.

Beating the market, however, is a relative contest, and there William Sharpe's short 1991 paper, “The Arithmetic of Active Management,” applies. Every dollar in the market is held either passively, in proportion to the index, or actively. Passive dollars earn the market return before costs. Since the market is the sum of both groups, active dollars as a whole must also earn the market return before costs. After costs, which are higher for active management, the average active dollar must earn less. Against the index, one investor's outperformance is another's underperformance, and fees push the total below zero.

Lasse Heje Pedersen argued in 2018 that the equality is not exact, because index funds must trade when the index changes and when companies issue or buy back shares, which leaves some room for active managers. The point stands in its practical form: active investors as a group cannot all beat the index, and after costs most of them don't.

What this means for an individual investor

None of this proves that nobody beats the market. It says the winners are hard to identify in advance and short track records are mostly noise. A few consequences follow from the data:

• A low-cost index fund is the baseline. Whatever approach you use, the honest test is what an S&P 500 fund did over the same dates, dividends included.
• Costs compound. A 1-point annual fee on a 10% gross return leaves about 17% less money after 20 years.
• One year proves little. In 2025 about one large-cap fund in five beat the index; over 20 years, about one in fourteen.
• Behavior costs real money. The DALBAR and Morningstar gaps measure what investors lose by moving money in and out at the wrong times.

This is general information, not personal financial advice. Your own goals, taxes and time horizon matter more than any average.

Testing yourself against the S&P 500, blind

Most people who feel sure they could have beaten the market are remembering the winners, a bias we look at in the blind investing game article. Blindfolio takes the hindsight away. A round covers a 10-, 15- or 20-year stretch of real market history somewhere between 1962 and 2026, and companies always appear under codenames. You can choose a start decade, in which case you see the dates, or play a random round, the fully blind version, where the years stay hidden and the era is revealed only at the end. Time runs week by week and you can pause at any moment. Real headlines (years masked in random mode) bring news on inflation, jobs and rate decisions, and the current Fed funds rate is shown on the bank deposit. At the end the real names are revealed and your portfolio is compared with the S&P 500 over the same period, dividends reinvested, as if your starting cash and every paycheck had gone into an index fund.

The rolling table above shows why that comparison matters: depending on the window, the index's price alone returned anything from about −5% to +17% a year over ten years. To compare results with friends or a class, everyone can pick the same start decade and duration (each player gets a different window inside that decade), or everyone plays random rounds and you compare which eras you drew. The game overview explains how a round works.