Search for the “best ETF for 2026” and the answers usually sort last year's winners. That is a ranking of rear-view mirrors, not a portfolio.
The better question is: what job is still empty? A U.S. stock core, international diversification and investment-grade bonds solve different problems. VTI cannot be “better” than BND any more than an engine is better than a brake. VOO and VTI can compete for the same core role; VXUS and BND cannot.
This guide uses four broad, low-cost Vanguard ETFs as examples because their jobs are easy to audit: VTI for the total U.S. market, VOO for the S&P 500, VXUS for stocks outside the United States and BND for the broad U.S. investment-grade bond market. It is a framework, not a forecast that any ticker will lead returns from today.
Educational information for a U.S. audience, not personalized investment, tax or legal advice. Funds can lose value. CalculatorAI is not affiliated with Vanguard.
The shortlistFour ETFs, three portfolio jobs
VTI and VOO are alternatives for a U.S. equity core. VXUS adds a region the first two mostly exclude. BND changes the asset class and usually the reason for holding it.
VTI
- Portfolio job
- Total U.S. stocks
- Holdings
- 3,515
- Expense ratio
- 0.03%
- What to watch
- Still market-cap weighted; mega-caps dominate
VOO
- Portfolio job
- Large U.S. companies
- Holdings
- 506
- Expense ratio
- 0.03%
- What to watch
- Top ten were 37.9% of assets at June 30
VXUS
- Portfolio job
- Stocks outside the U.S.
- Holdings
- 8,755
- Expense ratio
- 0.05%
- What to watch
- Currency, country and emerging-market risk
BND
- Portfolio job
- Broad U.S. investment-grade bonds
- Holdings
- 11,451
- Expense ratio
- 0.03%
- What to watch
- Interest-rate and credit risk; 5.7-year duration
| ETF | Portfolio job | Holdings | Expense ratio | What to watch |
|---|---|---|---|---|
| VTI | Total U.S. stocks | 3,515 | 0.03% | Still market-cap weighted; mega-caps dominate |
| VOO | Large U.S. companies | 506 | 0.03% | Top ten were 37.9% of assets at June 30 |
| VXUS | Stocks outside the U.S. | 8,755 | 0.05% | Currency, country and emerging-market risk |
| BND | Broad U.S. investment-grade bonds | 11,451 | 0.03% | Interest-rate and credit risk; 5.7-year duration |
Source: CalculatorAI · calculatorai.app · Vanguard product pages and fact sheets
The holdings count is not a score. Eleven thousand bonds do not make BND “safer” than cash, and 3,515 U.S. stocks do not remove equity drawdown risk. The count only tells you how broad the wrapper is inside the job it performs.
VTI vs VOOThe choice most people overthink
VTI tracks a broad U.S. market index spanning large, mid, small and micro-cap companies. VOO tracks the S&P 500 and therefore stops at leading large companies. Both charge 0.03%.
At June 30, Vanguard reported 506 stocks in VOO and a 37.9% top-ten weight. NVIDIA alone was 7.5%. VTI held 3,515 stocks as of July 31, but market-cap weighting means its largest positions still resemble VOO's. Adding thousands of small holdings changes the edges of the portfolio more than the center.
Own nearly the whole investable U.S. market
Large, mid, small and micro-cap stocks in one fund. Use it when total-market coverage is the job and you do not separately own smaller companies.
Own the S&P 500 large-cap core
A narrower but still diversified large-company portfolio. Use it when the S&P 500 is the intended benchmark or smaller-company exposure is handled elsewhere.
Owning both is not automatically extra diversification. It mostly increases the weight of the companies they share. Our VOO vs VTI comparison works through that overlap; the SPY vs VOO vs IVV guide explains why three S&P 500 wrappers are still one underlying allocation.
VXUSDiversification that may feel uncomfortable
VXUS tracks developed and emerging markets outside the United States. Vanguard's June 2026 fact sheet listed 8,755 stocks, a 0.05% expense ratio and only 15.0% in its ten largest positions—far below VOO's 37.9%.
That does not guarantee better returns. It means the sources of return, valuation, currency exposure and political risk are different. International diversification often looks least attractive immediately after a period of U.S. outperformance; that is exactly when a performance-ranking method is most likely to delete it.
The useful question is not “Will VXUS beat VTI next year?” Nobody knows. Ask whether a portfolio intended to represent global businesses should be 100% U.S., and what international target you can hold through both underperformance and recovery.
BNDThe ETF that should not be judged like a stock fund
BND tracks a broad market-weighted U.S. bond index. Vanguard reported 11,451 holdings, a 0.03% expense ratio, 5.7-year average duration and a 4.75% 30-day SEC yield around the time we checked. About half the portfolio was Treasury and agency debt in the June fact sheet, with government mortgage-backed and investment-grade corporate exposure making up much of the rest.
Yield is not a promised return. Bond prices move when rates and credit conditions change. Duration is the more useful first risk measure: roughly, a one-percentage-point move in yields can move price in the opposite direction by about the duration percentage, before convexity and income. That is an approximation, not a forecast.
BND's job is usually ballast, income and a source for rebalancing—not beating VTI in a bull market. If the money is needed next year, even an intermediate bond fund may have more price risk than the goal allows.
Build the allocation firstThen assign an ETF to each box
Investor.gov frames asset allocation around time horizon and risk tolerance. It also warns that several ETFs do not guarantee diversification if their top holdings overlap or if all funds target the same sector.
Name the goal and withdrawal date
Retirement in 25 years and a house deposit in two years cannot share one risk budget.
Set the stock-and-bond mix
Choose the loss range you can fund and actually hold through, not the mix with the best recent backtest.
Split U.S. and international stocks
Make the regional weight explicit. Do not let whichever market won recently make the decision by default.
Choose one fund for each role
VTI or VOO for the U.S. core, VXUS for ex-U.S. stocks, BND for broad investment-grade bonds in this example.
Set a rebalancing rule
A calendar date or percentage band turns the allocation into a process rather than a mood.
These examples show structure, not recommended portfolios:
Growth illustration
- VTI
- 70%
- VXUS
- 20%
- BND
- 10%
- Weighted fund fee
- 0.034%
Balanced illustration
- VTI
- 45%
- VXUS
- 25%
- BND
- 30%
- Weighted fund fee
- 0.035%
Lower-volatility illustration
- VTI
- 25%
- VXUS
- 15%
- BND
- 60%
- Weighted fund fee
- 0.033%
| Example | VTI | VXUS | BND | Weighted fund fee |
|---|---|---|---|---|
| Growth illustration | 70% | 20% | 10% | 0.034% |
| Balanced illustration | 45% | 25% | 30% | 0.035% |
| Lower-volatility illustration | 25% | 15% | 60% | 0.033% |
Source: CalculatorAI · calculatorai.app · CalculatorAI examples
The Asset Allocation Calculator can turn horizon and risk profile into a starting mix. Then the Portfolio Tracker shows whether market moves have pushed actual weights away from it.
FeesCheap does not mean free
The expense ratio is deducted inside the fund, so it does not arrive as a separate invoice. The SEC stresses that even small ongoing costs compound because the missing money also stops earning returns.
For a deliberately simple illustration, start with $100,000, assume 7% gross annual growth for 20 years and subtract one annual fee. A 0.03% fee leaves about $384,804; 0.50% leaves $352,365; 1.00% leaves $320,714. Taxes, trading spreads and real returns are excluded.
The difference between 0.03% and 0.50% removes about $32,440 more from this hypothetical ending balance.
Show these figures as a table
| Value ($ lost versus no fund fee) | |
|---|---|
| 0.03% fee — $2,164 less than a no-fee result | 2,164 |
| 0.05% fee — $3,601 less | 3,601 |
| 0.50% fee — $34,604 less | 34,604 |
| 1.00% fee — $66,255 less | 66,255 |
Source: CalculatorAI · calculatorai.app · CalculatorAI arithmetic · drafts/best-etfs-2026-numbers.mjs
Expense ratio is only one cost. ETFs trade at bid and ask prices, can trade above or below NAV and may involve commissions or tax consequences. Compare the spread and liquidity when two funds perform the same job; do not use a one-basis-point fee difference to justify a worse portfolio design.
Five trapsHow “best ETF” lists create bad portfolios
Ranking by one-year return
It selects what already became expensive or popular and silently changes the risk profile each year.
Buying several wrappers for the same index
VOO, SPY and IVV do not create three independent sources of diversification.
Counting tickers instead of holdings
Five ETFs can still be one mega-cap technology bet when their largest positions overlap.
Comparing bonds with stocks by return
A stabilizer is not failing because it trails the growth engine during a stock rally.
Ignoring the account
Income distributions, rebalancing taxes and available retirement-plan funds can change the best implementation.
Where these numbers come from
Vanguard's official pages and June 30, 2026 fact sheets supply fund objectives, expenses, holdings and concentration. The BND page supplied the more recent duration, holdings and SEC-yield snapshot. Investor.gov supplied the ETF, diversification and fee framework.
The future-value and weighted-fee calculations are reproduced in drafts/best-etfs-2026-numbers.mjs. We intentionally did not rank funds by 2026 return because past performance is not a forecast and because each role carries a different risk budget.
FAQFrequently asked questions
What is the best ETF to buy in 2026? There is no universal winner. For a broad U.S. stock core, VTI or VOO are reasonable examples; VXUS fills an international role; BND fills a bond role. The best choice is the lowest-cost, sufficiently liquid fund that accurately fills the missing allocation.
Is VTI better than VOO? VTI is broader because it includes smaller U.S. companies; VOO focuses on S&P 500 large caps. Both charged 0.03% when checked. Choose based on desired coverage, not a small recent return gap.
Should I own VTI and VOO together? You can, but most of VOO is already inside VTI. The combination tilts the total portfolio further toward S&P 500 companies rather than adding a new asset class.
How many ETFs do I need? Potentially only a few. What matters is coverage and overlap, not ticker count. One U.S. stock fund, one international stock fund and one broad bond fund can cover thousands of securities.
Are ETFs safe? An ETF is a wrapper, not a risk level. A broad bond ETF, a leveraged sector ETF and a total-market stock ETF can behave very differently. Read the objective, holdings, concentration, duration or leverage, fees and prospectus risks.
Next stepTurn a shortlist into a portfolio rule
Open the Portfolio Tracker, enter every fund you already own and look at weights rather than names. Decide which job is missing, set its target in the Asset Allocation Calculator, and buy only after you can state the rebalancing rule in one sentence.
The best ETF is not the ticker that wins 2026. It is the one you can explain, size and keep doing its assigned job when the ranking changes.
