ResourcesGuide

Should You Choose ETFs by Their 10‑Year Returns?

A 10-year average return is the first number most people check. What the evidence says about picking funds by past performance, what that average hides, and what to look at instead.

Published Updated 7 min readHow we research

Key takeaways

  • Not on their own. Past returns are a weak guide to future ones: only 4.5% of top-half US large-cap funds stayed in the top half five years running, below the 6.25% chance alone would give (S&P DJI, 2025).
  • Recent winners often reversed. Five of the 10 best ETFs of 2017–2020 ranked among the 10 worst of 281 in 2021–2025 in Exent’s data, including ARKK and the solar fund TAN.
  • The average hides the path. ARKK’s 15.9% a year over 10 years was close to SPY’s 15.2%, but ARKK fell 81.1% at its worst against 34.1% for SPY.
  • Chasing has a measured cost: the average fund dollar earned 8.7% a year against 9.9% for the funds themselves over 2016–2025, with the widest gaps in the most volatile funds (Morningstar).
  • Check what the fund holds, what it costs, how closely it tracks its index and how far it has fallen, then use past returns only to confirm it did its job.

What is your own portfolio exposed to? Check it with Exent →

Exent Gems

200 stocks Wall Street’s best‑ranked analysts agree on

We rank 1,191 analysts on 7 years of calls and keep the ones with proven records. Gems are the stocks where several of them agree.

106 of 20020%+ below their highList of Nov 19, 2026
59.0%target hit rate, 5★ analystsvs 53.0% for the average

Free sample from the Gems list, prices at the Nov 18, 2026 close: the top 3 stay locked, 3 names from rank 101 down change every trading day. Targets are analysts’ opinions, not promises.

See this week’s GemsLimited offer$1.95 for full access· full access

Exent is our product. Analyst ratings are opinions; educational content, not investment advice.

Can you choose ETFs based on historical performance and 10‑year average returns?

You can use them as a check, but not as the reason to buy. A 10-year average return tells you what a fund earned in one particular decade. It says little about the next one, because the main driver of an ETF’s return is the market, sector or style it holds, and leadership between those rotates.

The figure fund documents show is an average annual total return: the steady yearly rate that would turn the starting value into the ending value, with dividends reinvested. It is a fair summary of the past. It becomes a problem when it is used to rank funds and buy the top of the list, which in practice means buying whatever did best in the decade just ended. For a broad index fund, a good 10-year record mainly confirms that it tracked its index cheaply. For a sector or thematic fund, it mostly tells you that its theme had a good decade.

Does past ETF performance predict future returns?

Only weakly. S&P Dow Jones Indices’ U.S. Persistence Scorecard Year-End 2025 found that just 4.5% of US large-cap funds in the top half stayed in the top half every year over five years, when random chance would have produced 6.25%. Apart from small caps, none of the domestic equity funds in the top quartile in 2021 were still there through 2025.

The same pattern shows up in ETFs. We ranked 281 non-leveraged ETFs with a full record by their return in 2017–2020, then looked at where the 10 best landed in 2021–2025 (Exent market data):

ETF2017–2020 return2021–2025 returnRank in 2021–2025, of 281
ARKK (innovation stocks)+561.8%−37.3%276
TAN (solar)+537.6%−51.9%280
ARKG (genomics)+491.3%−68.7%281
ICLN (clean energy)+286.1%−37.7%277
PALL (palladium)+252.1%−36.7%275
SPY (S&P 500), for reference+80.2%+95.3%60

Five of the 10 best funds of 2017–2020 finished among the 10 worst of 2021–2025. Across the whole top quartile (70 funds) the picture was mixed: 32 stayed in the top quartile, helped by a technology rally that ran through both periods, 14 fell to the bottom quartile, and only 29 beat SPY in the second period. Funds that closed are missing from the data, so the real record of past winners was likely worse. See separating skill from luck.

What does a 10‑year average return hide?

The path, the risk and the source of the return. Two funds with the same average can have lived through very different decades. Common ETFs, 10 years to Sep 30, 2026 (Exent market data):

ETFReturn a year, 10 yearsWorst fall in the 10 years2022 return
SPY (S&P 500)15.2%−34.1%−18.2%
QQQ (Nasdaq‑100)20.9%−35.6%−32.6%
VTI (total US market)14.7%−35.0%−19.5%
VUG (US growth)17.9%−36.0%−33.1%
VTV (US value)12.3%−36.8%−2.1%
SCHD (US dividend)12.4%−33.4%−3.3%
IWM (US small caps)9.8%−42.3%−20.5%
VEA (developed markets)9.7%−39.9%−15.3%
VWO (emerging markets)7.6%−40.3%−18.0%
SMH (semiconductors)34.4%−45.3%−33.5%
ARKK (innovation stocks)15.9%−81.1%−67.0%
AGG (US bonds)1.1%−23.4%−13.0%
GLD (gold)11.7%−22.0%−0.8%
  • Drawdowns. ARKK and SPY averaged about the same over the decade, but ARKK lost 81.1% at its worst and returned −3.9% a year over the last five years. Few investors hold through a fall like that, which is why the average is not what most of them earned. See maximum drawdown.
  • Simple vs compound averages. ARKK’s calendar-year returns from 2017 to 2025 averaged 33.5% a year when added up and divided by nine, but compounded at 17.1%. Big losses pull compound returns down far more than a simple average shows.
  • Concentration. QQQ’s higher average came with 47.2% of the fund in its 10 largest holdings, against 38.6% for SPY (fund holdings, Oct 20, 2026). Much of the gap is a bet on a handful of companies. See is the S&P 500 too concentrated.
  • Fees and tracking. Funds on the same index differ by their costs and how closely they follow it; that gap is small but permanent. See SPY vs the S&P 500.

Is a high 10‑year return skill, or the decade’s style?

For index ETFs it is almost always the decade. The 10 years to 2026 favored large US growth and technology companies, so growth, Nasdaq and semiconductor funds sit at the top of the table and value, small-cap and international funds near the bottom. No fund manager chose that; the index did what its market did.

Decades can look very different. The S&P 500 price index ended 2009 at 1,115.10, below its 1,469.25 at the end of 1999, according to the index history in SPY’s January 2026 prospectus. Someone choosing funds in 2010 by their 10-year record would have ranked the S&P 500 near the bottom just before a decade in which its price index nearly tripled, to 3,230.78 at the end of 2019. See has the S&P 500 ever lost money over 10 years.

Why do investors who chase past returns earn less than their funds?

Because they tend to buy after a fund has risen and sell after it has fallen. Morningstar’s Mind the Gap 2026 study estimated that the average dollar in US mutual funds and ETFs earned 8.7% a year over the 10 years to Dec 31, 2025, while the funds themselves returned 9.9%. The 1.2-point gap came from the timing and size of investors’ purchases and sales.

The gap grew with volatility. It was about 0.4 points a year for the least volatile fifth of funds and more than 2.0 points for the most volatile fifth. Investors in US stock funds kept almost all of their funds’ return: 12.8% against 13.3%. The funds that look best on a 10-year ranking are often the volatile ones where this gap is widest. See how to stop chasing hot stocks.

What should you check instead of past returns?

Start from the job the ETF will do in your portfolio, then check whether it does that job cheaply and predictably. Past returns come last, as a sanity check.

  • Role: core stock holding, bonds, international, income or a deliberate tilt. Compare funds within a role, not across roles.
  • Holdings: what index it tracks, how many stocks, and how much sits in the top 10.
  • Cost: the expense ratio, plus how far the fund’s return trailed its index over 5 and 10 years.
  • Risk: its worst fall and how long it took to recover, alongside the return.
  • Risk-adjusted return: a Sharpe ratio compares return with the swings it took to earn it.
  • Overlap: whether it adds anything to the funds you already own.
  • Taxes and size: distributions in a taxable account, and enough assets and trading volume to stay open.

How to check your ETFs beyond past returns with Exent

  1. Upload a screenshot of your holdings (or a CSV, Excel, PDF or Google Sheet); no brokerage login.
  2. Exent compares the portfolio with a risk-matched benchmark and reports Sharpe, Sortino, Calmar and maximum drawdown.
  3. Crash replays (2008, 2018, 2020, 2022, 2025) and correlation clusters show how your ETFs behaved together in each selloff.
  4. Swap or reweight funds in the what‑if simulator to see the effect before you trade.

Exent is our product. Educational content, not investment advice.

Frequently asked questions

Use them as a check, not a selection rule. Past winners often lagged later, and a 10-year average hides drawdowns, concentration and the luck of the decade.

For a broad index fund it mainly shows the index did well and the fund tracked it cheaply. For a sector or thematic fund it mostly shows its theme had a good decade.

There is no fixed number. Compare a fund with others that have the same role and index; in Exent’s data SPY returned 15.2% a year over the 10 years to Sep 30, 2026, an unusually strong decade.

Rarely. S&P Dow Jones Indices found that only 4.5% of top-half US large-cap funds stayed in the top half five years running, below the 6.25% that chance would give.

The fund’s role in your portfolio, what it holds, its fees and tracking, its worst historical fall and how much it overlaps with what you already own.

Sources

  1. U.S. Persistence Scorecard Year-End 2025 — S&P Dow Jones Indices, Dec 31, 2025
  2. Mind the Gap 2026 — Morningstar, Aug 6, 2026
  3. SPDR S&P 500 ETF Trust prospectus (Form 497) — SEC EDGAR, Jan 26, 2026

See what your portfolio is really made of.

Upload a screenshot of your holdings and get concentration, overlap, crash exposure and a risk-matched benchmark in about a minute.

Check my portfolioLimited offer$1.95 for unlimited checks· unlimited checks

Educational content, not investment advice. Examples are illustrative; past performance does not guarantee future results.

DISCLAIMER: Exent AI is an informational research platform and does not provide investment advice, financial advice, or act as an investment adviser, broker-dealer, or fiduciary. All content, analytics, and model outputs are provided for educational and informational purposes only and should not be relied upon as a recommendation to buy, sell, or hold any financial instrument. Investing involves risk, including the possible loss of principal. Exent AI does not execute trades, hold customer funds, or provide brokerage services. Past performance and hypothetical results are not indicative of future results. Third-party data and AI-generated outputs may be incomplete or inaccurate.