Index Funds Beat Stock Picking for Almost Everyone. Here Is the Data.

The Scoreboard Exists

Picking stocks over owning the entire market sounds like a matter of taste the kind of thing reasonable people could argue about forever. It stopped being that way decades ago. Someone's keeping score. S&P Dow Jones Indices publishes SPIVA the S&P Indices Versus Active scorecard which has measured nearly every active U.S. mutual fund against its benchmark index twice a year for a quarter-century correcting along the way the statistical sins the industry loves most. a index fund buy all the shares of a benchmark index such as the S&P 500 at almost zero cost. active fund pay the professionals to try to overcome that. The scorecard simply responds if they do. Generally they don't and the size of the gap is the real topic of this piece not the fact itself

The Numbers

The latest results fit the pattern. In 2025 79 percent of active large-cap U.S. stock funds underperformed the S&P 500. That's the fourth-worst year for stock pickers in the scorecard's 25-year history and it comes on the heels of a 65 percent failure rate in 2024. A single bad year is forgivable. We broaden the horizon and the case doesn't get any betterIt gets worse. Over ten years fewer than one in six active large-cap funds beats the index. If you extend it to twenty years about 92 percent of domestic funds lag their benchmarks. Look at fifteen years of data on the 22 categories of U.S. stock funds tracked by SPIVA and you'll see that in no category did a majority of its managers beat the index. Not one. These are people with research staff Bloomberg terminals and CFA charters. As a group they lose to a list ofactions maintained by a committee

Why the Pros Lose

The result is not an insult to fund managers. It is the arithmetic they cannot escape and the clearest exposition comes from William Sharpe in a three-page article. His point seems too simple to be the complete answer but it is. Taken together professional investors are essentially the market because between them they own almost the entire market. So before costs the actively managed dollar average has to earn exactly the return of the market no more no less. After costs fees transactions and taxes thatSame dollar average asset must underperform at exactly those costs. Not approximately. Exactly as a matter of accounting before anyone's actual stock picking skill comes into the picture

The Arithmetic, Worked Through

It's easier to rely on Sharpe's logic once you look at the real numbers so here's a toy version not a real fund just arithmetic that anyone can verify. Suppose the entire U.S. stock market returns 10 percent this year and divide all the money invested in it into two categories: passive money that keeps the market exactly as it is and active money that managers shuffle among stocks trying to beat it. Let's call the 30 percent split passive and70 percent active although the actual split doesn't change what follows. Passive money by definition owns the entire market so it earns the market return before cost: 10 percent. Now figure out what active money earned before its own costs. Since passive and active together are the market its weighted average return has to equal the market return: 0.30 times 10 percent plus 0.70 times active return equals 10percent. Work with algebra and the active return before costs is also 10 percent. It has to be. There is no other number that balances the equation. That's Sharpe's entire argument in a line of algebra: before costs assets and liabilities earn the same by construction regardless of who is more skilled

Now add the cost. Let's say the average active investor pays 1 percent a year in total expense ratio trading costs taxes call it 1 percent. The average index investor pays next to nothing call it 0.03 percent. Net active return: 10 percent minus 1 percent or 9 percent. Net index return: 10 percent minus 0.03 percent or 9.97 percent.The gap between them 9.97 minus 9 is 0.97 percentage points and 0.97 is just 1.00 minus 0.03. The entire performance gap between the average asset dollar and the average index dollar is the cost gap dollar for dollar arithmetically before any manager has done anything whether skillful or foolish

If you widen that gap it will quickly become real money. Take $10,000 invest it for 30 years and assume that the market net of all fees returns 7 percent per year for both investors an assumption that holds skill constant so the only cost is doing any work. The index investor gets 7 percent minus 0.03 percent or 6.97 percent per year. The active investor gets 7 percent.If you compound $10,000 at 6.97 percent a year for 30 years using the standard compound growth formula you get about $57,400. If you compound the same $10,000 at 6.97 percent a year for 30 years you get about $75,500. The difference between those two figures is about $18,000.dollars greater than the original investment and is due entirely to a fee difference of well less than one percentage point per year. No one chose a bad action in this example. No one made a mistake. The fee ate up that amount and was accumulating all the time

The Shape of Stock Returns

There's a second piece of arithmetic behind these numbers and it gets less attention than Sharpe's: how stock returns are actually distributed not just what their average is. Most people assume that the return of a typical stock looks like the return of the index only noisier. Not so. Research by finance professor Hendrik Bessembinder which tracks the lifetime returns of nearly all listed stocks in the United States back to the 1990s1920 found that a large portion of individual stocks something like half or more depending on the exact sample do not even beat the return of holding a one-month T-bill over its entire life on the stock market from listing to delisting. A small portion of companies on the order of a few percent of all publicly traded ones appear to account for essentially all of the stock market's net wealth creation above that risk-free floor. Everything else fades into roughly nothing

That distribution is brutally asymmetrical and it's why picking a handful of stocks is much harder than intuition suggests. An index fund owns everything both the handful of big winners and the long tail of mediocre or failed companies. A stock picker with twenty or thirty names has to pick up some of the rare winners because losing them all and having a diversified sample of everything else is close to the average outcome. Remember that the median stock tends to underperform the average stock.market precisely because the average is dragged up by a few extreme winners. This is also why the index seems so difficult to beat with a concentrated portfolio. Concentration is exactly the way you would try to catch a rare winner but it is also exactly the way you guarantee losing it if you guess wrong and most people guess wrong

The Bias Built Into the Comparison

There's a third reason why the arguments for active management look better in casual conversation than in SPIVA numbers and it has nothing to do with skill or arithmetic. It has to do with which funds are still left to measure. Funds that perform poorly for long enough are usually closed or merged into a better-performing sister fund at the same firm and once that happens they tend to quietly disappear from most performance databases. If they're just averagedThe returns of funds that still exist today automatically exclude a portion of the worst performers from years past favoring the remaining average. This is called survivorship bias and it's exactly the kind of thing that makes stock picking look more attractive on a table than it turns out to be on paper

SPIVA is designed to correct for this which is part of the reason its own methodology mentions corrections for the industry's favorite statistical sins. It keeps track of funds that no longer exist and puts their results back into the average rather than letting them fade away. Many of the softer more informal comparisons in circulation the track record of a friend's advisor a magazine's list of top funds from five years ago don't bother. That's one reason why one numberLike 79 percent falling behind in a single year or 92 percent in twenty tends to be harder than people expect. That's not being unfair to active managers. If anything it's closer to the real picture than most of what is informally repeated

Case Study: The Buffett Bet

If you want this argument in its most famous real-world form look at Warren Buffett's decade-long bet with asset management firm Protege Partners. In 2008 Buffett bet through the Long Bets project that a low-cost S&P 500 index fund would outperform a carefully selected portfolio of hedge funds over the next ten years after all fees. Protege co-founder Ted Seides took the other side and chose a group of hedge funds.funds that is funds that themselves invest in a basket of hedge funds accruing an additional layer of fees on top of the hedge funds' own fees

When the bet was settled in 2017 the index fund had won and not by a little. The hedge fund basket had compounded at a modest pace over the decade weighed down by the tiered fee structure and a series of years that were simply difficult for many hedge fund strategies. The index fund which controlled the entire market for almost nothing in fees had compounded at a significantly higher annual rate over the same ten years and the gap between the two overSeides has since written candidly about what he went wrong and it's worth reading because his own explanation lines up almost exactly with Sharpe's arithmetic

What I deduce from this bet is not that hedge funds are worthless. Some individual hedge fund strategies do things that a structurally index fund cannot such as selling short or directly hedging market risk. What I deduce from it is that the accumulation of fees made an already difficult problem almost impossible and that a full public decade with a specific verifiable result is exactly the kind of evidence that is difficult to argue with after the fact as opposed to a single strong year that any managercan point out

The Honest Fine Print

A fair version of this article has to admit what the numbers don't say. Indexing guarantees you the market's losses as faithfully as its gains. 2022 came pure for index investors. The same was true for the April 2025 tariff drop. Whether an investor actually maintains the performance of the index rather than selling on one of those drops comes down to behavioral discipline which is a separate issue from the one this article addresses. Some corners of the market certain categories of bonds andSmall-cap niches show somewhat better active results although even there majority success over long horizons remains rare. Concentration also pays off both ways. Owning the cap-weighted index today means making an unusually large bet on a handful of tech giants whether you want to or not a real characteristic worth understanding even though active managers have mostly failed to exploit it. And picking a few stocks with a small portion of your own money is an educationlegit.Part of the reason this site exists is that analyzing individual companies teaches you finances in a way that nothing else does.Just do it knowing the score and calculate the money like tuition not retirement savings

Where the Model Breaks

Steelmanning the other side correctly means going beyond the fine print above so here's the strongest version of the case against general indexing that I can build. First Sharpe arithmetic is a statement about the average active dollar not about a specific investor and averages can hide real dispersion. A small number of funds and individual investors persistently add value net of fees disproportionately concentrated in less-watched corners of the market: distressed debt certain small- and micro-cap namessome emerging and frontier markets where there are fewer analysts watching and the prices are more likely to be really wrong. The problem and it is a real one is finding those managers or that skill in advance and not in hindsight which is exactly the persistence problem mentioned a few sections back

Second the argument for indexing weakens as passive ownership grows at least in theory. If almost everyone indexes in principle no one is doing the work of setting prices through research and trading and markets could become less efficient over time which would return real opportunities to whoever is still doing fundamental analysis. Whether passive flows have become large enough somewhere to cause that is something that is really debated among people who study market structureand I don't think the evidence is consolidated enough to say either way. It's a real theoretical limit on how far you can push logic even if it obviously hasn't gone yet

Third the index itself is a specific and possibly arbitrary choice. A cap-weighted index automatically puts the most money into what has already risen the most which has mechanically driven exposure to a handful of very large tech companies in recent years. This is not a neutral opinion-free choice. It is itself a bet it's just that most investors don't consider it a bet because it is the default. An investor who is uncomfortable with how concentrated the index has become is not wrong tofeel that even if the data says that active managers haven't solved that problem for them either

Fourth and this is the condition I would watch for myself: taxes and account structure change the math. In a taxable account an active fund that trades a lot can generate a steady stream of taxable distributions that erode returns beyond what its expense ratio alone would suggest while an index fund that rarely trades tends to defer most of its gains. That tax drag adds up to all of the above and can be large enough on its own in an active fund ofhigh turnover held in a taxable account as to explain a significant portion of the poor performance without invoking the skill of the manager at all

How I Actually Use This

So how do I actually use all of this as a student with a small amount of real money and a much larger amount of opinions? My reading is that SPIVA changed what I think the burden of proof should be. I used to think the default question was: why not try to beat the market when smart people clearly sometimes do? Now I think the default question goes the other way: why would you expect to be the exception and what specific advantage do you have that the research staff professional and Bloomberg Terminal don't?

For most of what I would consider real long-term money my honest read is that the boring answer wins in the numbers and I don't think it's a controversial thing to say. It's closer to a consensus among people who have actually looked closely at the data.times about it it's close to the best finance education out there. I just try to be honest with myself that the money behind those individual bets is tuition money not retirement money and I keep it that way on purpose

Where I really find myself applying Sharpe logic day to day is less about whether I should index and more as a filter for when I trust a track record. Every time I see a fund a strategy or honestly a person claiming to beat the market my first question now is not how they did it but what the arithmetic says must be true for that to be sustainable for everyone doing it at once. If the answer only works because most people don't that's fine butIt means that the strategy has a shelf life tied to how many people copy it. If the claim requires the person to be smarter than the aggregate of every other smart well-resourced professional doing the same thing permanently I've become much more skeptical. I think skepticism is the habit that's really worth developing more so than any specific recommendation about a fund

The Bottom Line

Twenty-five years of SPIVA scorecards point in the same direction from every angle. Most professional stock pickers lose to the index in most years the failure rate increases rather than decreases the longer it is measured and past winners do not repeat reliably. The causes are structural not personal. Sharpe arithmetic guarantees that active money as a whole must underperform passive money in a proportion close to the cost difference between them the distributionBiased bias in individual stock returns punishes anyone who doesn't own literally everything and survivorship bias flatters the informal version of this story that is repeated on the dinner tables.low cost automate purchasing and let the same arithmetic that defeats most professionals quietly work out on your side