In brief

An investor can beat the S&P 500. Some do. The difficult part is identifying, in advance, who will outperform—and then continuing to outperform after fees, taxes, and mistakes.

The basic obstacle is arithmetic. Investors who hold the market collectively earn the market’s return before costs. Every active position above the market weight must be matched by someone else holding less. The winners’ gains relative to the market are offset by the losers’ shortfalls. Once the higher research, trading, management, and tax costs of active investing are deducted, the average actively managed dollar must trail a comparable low-cost passive dollar.

Historical evidence shows the practical result. In the SPIVA U.S. Year-End 2025 Scorecard, 78.78% of active large-cap U.S. equity funds underperformed the S&P 500 over one year. The underperformance rates were 85.59% over 10 years, 89.93% over 15 years, and 92.89% over 20 years, all ending December 31, 2025.

Those figures do not prove that index funds will win in every period. They show that beating an appropriate benchmark has historically been a minority outcome—and that the challenge became more severe over the longer periods in this dataset.

First, what does “beat the S&P 500” mean?

The comparison needs a definition before it needs a result.

The S&P 500 measures the large-cap segment of the U.S. stock market. It contains 500 leading companies and covers approximately 80% of available U.S. market capitalization, according to S&P Dow Jones Indices. It is float-adjusted and market-capitalization weighted, so companies with larger publicly tradable market values have larger index weights.

To “beat” it fairly, I would compare:

  1. the same start and end dates;
  2. total returns, including reinvested dividends;
  3. results after the investment costs actually paid;
  4. the same currency;
  5. a similar risk profile; and
  6. taxes when the decision concerns a taxable investor’s spendable outcome.

Several apparent victories disappear when these terms are aligned. A technology portfolio may exceed the S&P 500 while taking much more sector and valuation risk. A leveraged portfolio may post a higher return while accepting a much larger drawdown. A chart may compare a stock’s price return with the index’s total return, or vice versa. A manager may show performance before an advisory fee that the client could not avoid.

Raw return still matters, but it does not answer every question. A meaningful comparison asks whether the investor received more return for the risks and costs taken.

The arithmetic that active investors cannot escape

William Sharpe explained the core problem in his 1991 essay, “The Arithmetic of Active Management”.

Choose a market. A passive investor owns every security in that market at its market weight. Active investors, by definition, hold something different. Because all shares must be owned by someone, the active investors collectively hold what remains after the passive holdings are accounted for.

Before costs:

Average passive dollar return = market return

Average active dollar return = market return

After costs:

Average active dollar return = market return − higher active costs

This does not depend on a belief that markets are perfectly efficient. It does not say security analysis has no value. It is an accounting identity across the selected market.

A simplified $100 market

Imagine a market containing $100 of stock:

  • $40 is held passively in exact market weights.
  • $60 is held by active investors making different selections.
  • The entire market earns 8% before costs.

The passive $40 earns 8% before its small implementation costs. Because the return on the complete $100 must also be 8%, the active $60 collectively earns 8% before costs too.

Within that active group, one investor may earn 15% and another may earn 1%. Skill, luck, and risk exposures determine how the result is divided. But the group cannot collectively create an excess return merely by trading shares among itself.

If passive implementation costs 0.05% and active implementation costs 1.00% in this hypothetical teaching example, the net averages become:

Approach Gross market return Assumed annual cost Net return
Passive market-weight investor 8.00% 0.05% 7.95%
Average active dollar 8.00% 1.00% 7.00%

The numbers are assumptions, not current fund fees or forecasts. The purpose is to show where the structural disadvantage appears: after costs.

An important limitation

Sharpe’s arithmetic applies to a properly defined market and all investors active within it. The S&P 500 is not the entire global investment market. An active U.S. large-cap fund may hold cash, smaller companies, or foreign shares. Individual investors and institutions outside the measured mutual-fund universe also take active positions.

That is why an empirical fund comparison is not identical to the accounting proof. A careful scorecard matches funds to an appropriate benchmark, includes funds that later disappear, and distinguishes equal-weighted fund averages from asset-weighted investor outcomes.

Costs create a hurdle that compounds

Active investing can add several layers of cost:

  • fund expense ratios or advisory fees;
  • bid-ask spreads;
  • market impact from trading;
  • commissions or contract fees where applicable;
  • research, data, and subscription costs;
  • taxes triggered by realized gains and distributions; and
  • the value of the investor’s time.

Some of these costs are now lower than they were historically. Zero-commission stock trades removed one visible charge at many brokers, but they did not remove spreads, market impact, taxes, fund expenses, or behavioral costs.

Small annual differences become large dollar differences because the money paid in fees no longer compounds. The SEC’s July 2025 fees and expenses bulletin illustrates this with a $100,000 portfolio assumed to grow 4% annually for 20 years. Its approximate ending values are $208,000 with a 0.25% annual fee, $198,000 with a 0.50% fee, and $179,000 with a 1.00% fee.

The example is deliberately simplified, but the lesson is durable: a manager does not merely need to find better investments. The manager must produce enough gross excess return to overcome every incremental cost, year after year.

Taxes can raise the hurdle again

Taxes do not affect every investor equally. Trading inside a tax-advantaged retirement account differs from trading in a taxable brokerage account. Tax rates, holding periods, losses, distributions, state rules, and personal circumstances all matter.

Still, active turnover can create a disadvantage in taxable accounts:

  • selling an appreciated position can realize a gain earlier;
  • a short holding period can produce short-term rather than long-term federal capital-gain treatment;
  • an active mutual fund may distribute gains even when the shareholder did not personally sell shares; and
  • paying tax today leaves less capital invested for future compounding.

S&P DJI’s After-Tax Scorecard for year-end 2024 reported that the median active large-cap core fund trailed the S&P 500 after tax over every measured horizon, by as much as 4.4 percentage points annually. The study uses standardized assumptions rather than any one investor’s tax return, so it should be read as comparative evidence—not a personalized tax estimate.

Tax efficiency alone does not make an investment good, and an index fund can distribute taxable income. The point is narrower: an active strategy must overcome taxes as well as expenses when taxes are applicable.

The index automatically keeps successful companies

The S&P 500’s float-adjusted market-cap weighting has a powerful mechanical feature: as a constituent’s market value rises relative to the others, its index weight rises. A passive investor does not have to predict which company will become the next extraordinary winner before the move begins.

The reverse is also true. A company whose value declines becomes a smaller part of the index. If it is later removed, the index methodology handles the change. An index fund is not perfectly costless and does not trade without friction, but it does not require the investor to repeatedly identify winners and losers in advance.

This matters because stock returns have historically been highly uneven.

Hendrik Bessembinder’s 2018 study, “Do Stocks Outperform Treasury Bills?”, examined U.S. common stocks over 1926–2016. When lifetime wealth creation was measured in dollars, the best-performing 4% of listed companies accounted for the net gain of the entire U.S. stock market above one-month Treasury bills; the other stocks collectively matched bills.

That finding does not mean a randomly selected stock has a 4% probability of making money, and it does not say every broad index will always outperform. It shows that aggregate market wealth creation was extremely concentrated in a small minority of extraordinary stocks.

Why concentrated winners make stock picking hard

Suppose a market has ten stocks. Nine return 0%; one returns 100%. An equal initial investment in all ten returns 10% before costs. An investor who owns only three stocks has a wide range of possible outcomes:

  • choose the winner and two flat stocks: the portfolio returns 33.3%;
  • miss the winner: the portfolio returns 0%.

These are invented teaching numbers, not a historical dataset. They illustrate why a market can perform well even though most individual securities do not match it. Missing a small number of major winners can be more damaging than avoiding several ordinary losers is helpful.

Market-cap weighting sets a surprisingly high bar

The S&P 500 is sometimes described as a list of 500 stocks, but its return is not the return of the average constituent. Larger companies have more influence.

This creates a particular challenge when a narrow group of large stocks leads the market. An active manager who limits concentration or views the leaders as expensive may hold less of them than the index. That can be prudent risk management, but it can also cause substantial relative underperformance while the leaders keep rising.

The 2025 SPIVA report provides a concrete example. S&P DJI found that only 30% of S&P 500 constituents outperformed the index during 2025. An investor could therefore select a stock that rose and still trail the benchmark. The same report notes that the S&P 500 Equal Weight Index underperformed its capitalization-weighted counterpart through most of that year as mega-cap-heavy information technology performed strongly.

This pattern is not permanent. Equal weighting and smaller stocks can lead in other periods. The lesson is that “many stocks look attractive” does not mean the average stock picker has favorable odds against a capitalization-weighted benchmark in a specific regime.

Good information is not necessarily an edge

Public companies, analysts, institutions, algorithms, and millions of investors constantly process information. A company can report excellent results and still fall because the market expected even more. A struggling company can rise because the news was less bad than its price implied.

To outperform through selection, being correct about the company is not enough. The investor’s view must be more accurate than the view already embedded in the price.

That usually requires getting several decisions right:

  1. identify a security the market has mispriced;
  2. estimate the direction and size of the error;
  3. buy before the error closes;
  4. size the position appropriately;
  5. distinguish a temporary decline from a broken thesis;
  6. decide when to sell; and
  7. repeat the process often enough to separate skill from luck.

A single great choice can beat the index for years. It can also create a dangerous lesson if the investor concludes that one outcome proves a repeatable process.

Trading behavior often subtracts return

Brad Barber and Terrance Odean studied 66,465 households with accounts at a large discount broker from 1991 through 1996. Their 2000 paper, “Trading Is Hazardous to Your Wealth”, reported that the households that traded most earned an 11.4% annual return while the market returned 17.9% during the sample period.

The study came from an earlier brokerage era with higher explicit trading costs, so its exact return gap should not be projected onto today’s accounts. Its central evidence remains relevant: more activity did not translate into better results for those households.

The SEC-requested Behavioral Patterns of U.S. Investors review identifies several recurring behaviors associated with poor decisions:

  • active trading;
  • holding losing investments too long and selling winners too soon;
  • focusing on past fund performance while ignoring fees;
  • familiarity bias;
  • mania and panic;
  • momentum chasing without a disciplined process;
  • inadequate diversification; and
  • noise trading.

These are not signs of low intelligence. They are human responses to uncertainty, social proof, regret, and loss. Knowing their names does not make an investor immune.

A common behavioral cycle

A familiar cycle looks like this:

  1. An asset produces strong recent returns.
  2. Attention and confidence grow.
  3. The investor buys after much of the advance.
  4. The asset falls or stops leading.
  5. The investor loses conviction and sells.
  6. A new recent winner becomes attractive.

The investor experiences the return of their entry and exit decisions—not the clean long-term return shown on the asset’s factsheet.

Past winners are difficult to select in advance

Looking backward, successful managers seem obvious. Looking forward, the investor must distinguish persistent skill from favorable style exposure, concentration, and luck.

The S&P U.S. Persistence Scorecard for year-end 2025 tracked funds without ignoring those that disappeared. Among 164 large-cap funds that began in the top quartile in 2021, none remained in the top quartile through 2025. Among 334 above-median large-cap funds from 2021, only 4.49% remained above median in every subsequent year through 2025; a uniform random process would imply 6.25%.

That does not prove every winning manager was lucky. It shows that a simple rule—buy the recent winners—did not reliably identify persistent leadership in that sample.

The same report illustrates why survivorship matters. Poor funds can merge, liquidate, or change style. A database that looks only at funds still operating today erases some choices that a real investor could have made at the beginning of the period.

What the long-run fund evidence says

SPIVA has compared active funds with category-appropriate indices since 2002. Its year-end 2025 large-cap results are especially useful because the report includes multiple horizons and corrects for survivorship bias.

Period ending Dec. 31, 2025 Active large-cap funds underperforming the S&P 500
1 year 78.78%
3 years 66.84%
5 years 88.96%
10 years 85.59%
15 years 89.93%
20 years 92.89%

Source: SPIVA U.S. Year-End 2025, Report 1a. The report compares active large-cap funds with the S&P 500 on absolute return. Past performance does not guarantee future results.

Three cautions matter:

  1. The one-year result changes from year to year. In SPIVA’s annual history for 2001–2025, large-cap underperformance ranged from 45% to 87%. Active managers sometimes have much better years.
  2. Fund results are not every individual investor’s results. Individuals may hold different portfolios, trade at different times, and face different costs.
  3. A benchmark must fit the strategy. Comparing a small-cap, international, value, or balanced portfolio directly with the S&P 500 can confuse an intentional allocation difference with manager skill.

The evidence supports “most did not beat,” not “nobody can beat.”

Why a few success stories feel like the norm

Winning investors receive attention because exceptional outcomes are interesting. Thousands of quiet failures do not produce famous books, interviews, or social posts.

This creates several distortions:

Survivorship bias

We see the funds, firms, and investors that remain. Failed funds are merged or closed; abandoned newsletters disappear; unsuccessful personal portfolios are rarely published in full.

Selection bias

A platform can feature the best-performing account without showing how many accounts were eligible. The result may be real but unrepresentative.

Outcome bias

A profitable decision can have been poorly reasoned, and a sound decision can lose money. Judging the process only by one outcome rewards luck.

Hindsight bias

After a company becomes dominant, its success can look inevitable. Beforehand, investors faced competitors, changing technology, valuation risk, regulation, management uncertainty, and incomplete information.

Hidden risk

Two portfolios can end with the same return while taking very different paths. Concentration, leverage, illiquidity, options, or exposure to one economic outcome can make a result difficult to survive or repeat.

Does market efficiency explain everything?

The Efficient Market Hypothesis helps explain why widely known information is difficult to exploit, but the case for indexing does not require believing that every price is always correct.

Markets can be wrong. Securities can be mispriced. Skilled investors can discover an error and profit. The problem is competitive:

  • mispricing attracts research and capital;
  • exploiting it can reduce it;
  • identifying it costs money;
  • a visible pattern may weaken after discovery; and
  • the investor must distinguish an enduring edge from a historical coincidence.

Sharpe’s arithmetic remains true even in an inefficient market. Active winners can earn excess returns, but other active holders must be on the other side before costs. The average active dollar does not become above average simply because opportunities exist.

When beating the S&P 500 is the wrong goal

The S&P 500 is a useful benchmark for U.S. large-cap stocks, not a complete financial plan.

An investor may reasonably hold:

  • international stocks for geographic diversification;
  • small- and mid-cap stocks for broader market coverage;
  • bonds or cash to reduce volatility or fund near-term spending;
  • inflation-linked securities for a specific liability;
  • a lower-risk allocation near retirement; or
  • a portfolio designed around taxes, liquidity, or income needs.

Any of these choices can trail the S&P 500 during a strong U.S. large-cap market. That does not automatically make the plan a failure.

The correct benchmark should reflect the portfolio’s stated opportunity set and risk. A 60/40 portfolio should not be criticized merely because it trails an all-stock index in a bull market; it was built to hold less equity risk.

Can an individual investor still outperform?

Yes—but the word “can” carries less information than it seems.

An investor might outperform through:

  • genuine analytical skill;
  • a rewarded factor exposure;
  • concentration in a major winner;
  • taking more systematic risk;
  • leverage;
  • an unusual information or execution advantage;
  • a favorable market regime; or
  • luck.

These paths are not equivalent. A repeatable edge should survive reasonable changes in time period, benchmark, costs, and risk adjustment. A lucky concentrated outcome may not.

An active investor should be able to answer:

  1. What is my benchmark, chosen before seeing the result?
  2. What is my reason for believing I have an edge?
  3. Why might other market participants not already reflect this information in price?
  4. What evidence would prove my thesis wrong?
  5. How much can I lose if I am wrong?
  6. What are my complete costs and tax consequences?
  7. Is the strategy scalable and repeatable?
  8. How many years of evidence would distinguish skill from luck?

If those questions do not have clear answers, the strategy may be speculation rather than a measurable investment process.

A core-and-explore example

Indexing does not require abandoning curiosity. One practical structure is to separate the portfolio into two roles:

  • a diversified, low-cost core intended to capture broad market returns; and
  • a limited “explore” allocation for individual securities or active ideas.

For example, an investor might choose a hypothetical 90% core and 10% active sleeve. If the active sleeve lost 50% while the core was unchanged, the portfolio-level loss from the sleeve would be approximately 5% before interactions, rebalancing, taxes, and costs. If the entire portfolio were committed to the same idea, the loss would be 50%.

The 90/10 split is only an arithmetic illustration, not a recommended allocation. Its purpose is to show how position sizing can preserve room for learning without making one idea responsible for the entire plan.

What I take from the evidence

I do not read this history as proof that analysis is pointless. I read it as a warning about the standard an active decision must clear.

The S&P 500 is difficult to beat because it is:

  • a diversified basket of leading U.S. companies;
  • weighted so that large winners become more important;
  • maintained as companies and the economy change;
  • available through low-cost index products;
  • free from the need to forecast each next winner; and
  • a benchmark that active investors collectively resemble before costs.

The greatest advantage available to an ordinary investor may not be superior prediction. It may be the ability to keep costs low, diversify, avoid forced selling, contribute consistently, use appropriate accounts, and stay with a reasonable plan through uncomfortable markets.

That approach will not beat the S&P 500 every year. A fund tracking the S&P 500 will itself trail the published index by its expenses and implementation differences. A diversified portfolio that includes assets beyond U.S. large-cap stocks will also behave differently.

But for money assigned to U.S. large-cap equities, accepting the market return can remove the requirement to repeatedly find the small minority of winners, select tomorrow’s skilled manager, and time every change correctly. That is not settling for mediocrity. It is recognizing how high the hurdle really is.

Bottom line

Beating the S&P 500 is possible, but the average active dollar cannot outperform the market before costs, and costs, taxes, trading, behavior, and concentrated winners raise the hurdle further. A low-cost index approach does not guarantee success; it removes several decisions that historical evidence shows are difficult to make correctly and repeatedly.

Evidence map and further reading

Historical results describe specific samples and periods. They are not forecasts, and they do not guarantee that an index strategy will outperform an active strategy in any future period.