Most professionals trail the index over any long stretch, most individual stocks trail Treasury bills over their lives, and simple arithmetic says the average active dollar must lose after costs. Where that leaves a desk that trades for a living — this one included.
This page disagrees with the desk's own business, and says so up front: for most people, with most of their money, most of the time, a broad low-cost index fund beats any desk, including this one. The evidence for that is not a matter of opinion. It is three findings that have held for decades, and a piece of arithmetic that cannot be argued with.
The professionals lose to the index
S&P Dow Jones Indices publishes the SPIVA scorecard, which compares actively managed funds with the index they are trying to beat. The year-end 2025 edition, with data through December 31, 2025, gives the share of actively managed U.S. large-cap funds that underperformed the S&P 500:
| Horizon | Large-cap active funds that trailed the S&P 500 |
|---|---|
| 1 year | 78.8% |
| 3 years | 66.8% |
| 5 years | 89.0% |
| 10 years | 85.6% |
| 15 years | 89.9% |
| 20 years | 92.9% |
These are professionals, with analysts, data and full-time attention, measured after their fees. Over twenty years, nine in ten of them would have done better for their investors by buying the index and going home. The one-year figure moves around — some years a third of managers beat the index, some years a tenth — but the long-horizon figures barely do, and they are the ones that matter to someone saving for decades.
Why it has to be that way
In 1991 William Sharpe wrote a two-page paper that settled the question in principle. Every dollar invested in the U.S. stock market is held either passively (in the whole market, at the market's weights) or actively (in something else). The passive dollars earn the market return, by definition. So the active dollars, taken together, must also earn the market return before costs — they are the rest of the same market. After costs, which are higher for active management, the average active dollar must earn less. Sharpe's own words: "these assertions will hold for any time period." Some active managers beat the market every year; the arithmetic only says they do it at the expense of other active managers, and that the group as a whole pays the costs.
Costs are not a detail. Morningstar's 2016 study of its own fund data found the expense ratio "the most proven predictor of future fund returns": among U.S. equity funds, the cheapest fifth succeeded — survived and beat their category — 62% of the time over 2010–2015, against 20% for the priciest fifth. Three times as likely, for doing nothing but charging less.
Most stocks lose to Treasury bills
The second finding is about the stocks themselves, and it surprises almost everyone. Hendrik Bessembinder examined every U.S. common stock in the CRSP database from 1926 to 2016 — 25,332 of them — and asked how many beat one-month Treasury bills over their own lifetime. The answer was 42.6%. More than half of all stocks delivered a negative lifetime return; the median stock's lifetime buy-and-hold return was −2.29%; and the single most common outcome for a stock was a loss of 100%. The market as a whole made money anyway, because the gains were concentrated in a tiny group: the best-performing 4.31% of companies — 1,092 of them — account for the entire net wealth creation of the U.S. stock market since 1926. Fifty percent of it came from just 90 firms.
Bessembinder and co-authors repeated the exercise on 64,000 stocks worldwide from 1990 to 2020 and found it sharper, not softer: 55.2% of U.S. stocks and 57.4% of non-U.S. stocks underperformed Treasury bills, and the top 2.4% of firms accounted for all of the world's net stock-market wealth creation. The reason a stock picker trails the index is not that stock pickers are foolish. It is that stock returns are a lottery with a few enormous winners, and the index holds all the tickets.
Individuals do worse than the professionals
The third finding is about people like us. Barber and Odean obtained the records of 66,465 households at a large discount broker from 1991 to 1996. The average household earned a gross return of 18.7% a year, and after trading costs 16.4%, against a market return of 17.9%. The fifth of households that traded most, turning over more than three quarters of their portfolio a year, earned 11.4% net — more than six percentage points a year behind the market, for years. The paper's title is its conclusion: "Trading is hazardous to your wealth." The households were not picking badly, on the whole; they were picking often, and paying for it every time.
Where that leaves a trading desk
Honestly: in a small, specific corner. The desk does not run a portfolio and is not an alternative to one. What it runs is a handful of tested futures systems with a disclosed stop and a disclosed worst case, sized from that stop, with the full record — losing months included — published for anyone to read. That is a tool for a particular kind of risk, taken with money whose loss would not change your life, by someone who has read the pages on this shelf first. It is not where savings go, and nothing on this site will tell you otherwise.
The burden of proof the evidence above puts on any active approach is the one the desk accepts for itself: beat a plain benchmark after every cost, on data it was not built on, for longer than luck can explain, and publish the months where it did not. What a backtest can and cannot tell you is the desk's own version of that test, and the demo desk is where the record lives.
Reading anyone's record, including ours
Five questions do most of the work. Is the return after fees and costs? Is it compared with the right benchmark — a large-cap stock fund against the S&P 500, not against cash? Does it cover ten years or more, so that the SPIVA pattern has had time to show? Does the record include the funds or systems that died along the way, or only the survivors? And if it is a stock picker: how many of the 4% did they hold, and did they hold them through the years those stocks looked finished?
A plain index fund answers all five without trying. That is why it is hard to beat, and why the honest thing for a desk to say is that beating it is not what the desk is for.
- SPIVA U.S. Scorecard, year-end 2025 — S&P Dow Jones Indices, March 2026
- SPIVA U.S. Scorecard, year-end 2025 (full report) — the multi-horizon table
- The arithmetic of active management — William F. Sharpe, Financial Analysts Journal 47(1), 1991 (reprint)
- Do stocks outperform Treasury bills? — Hendrik Bessembinder, Journal of Financial Economics 129(3), 2018 (author's draft)
- Long-term shareholder returns: evidence from 64,000 global stocks — Bessembinder, Chen, Choi and Wei, Financial Analysts Journal 79(3), 2023
- Trading is hazardous to your wealth — Barber and Odean, Journal of Finance 55(2), 2000
- Predictive power of fees — Russel Kinnel, Morningstar, May 2016
Education, not advice. Nothing on this page recommends buying or selling anything, and nothing is a promise of results — every figure is cited above or hypothetical. TOTH does not provide investment advice. New to the vocabulary? The glossary has plain definitions.