What stock, bond and international mix should I hold?
See what the common age rules, a target-date fund and 98 years of US returns say for your age and the biggest yearly loss you could sit through.
Biggest one-year loss you could sit through without selling
A 30% year takes $30 off every $100 invested.
What the part that is not stocks is held in
Bonds lose money in some years, including 2022.
MOST STOCKS WITHIN YOUR LIMIT
66% stocks
From 1928 to 2025, a mix of 66% US stocks and 34% 10-year Treasury bonds lost at most 29.8% in a year (1931), inside your 30% limit. On $420,000 that year would have cost $125,172. At 34 the “110 minus your age” rule gives 76% stocks and Vanguard’s target-date path 90%. This is history and published rules, not a recommendation.
Worst year
−29.8%
Average year
8.6%
Worst year in $
$125,172
Worst 10 years
1.4% a year
UNDERSTAND YOUR RESULT
Where the common answers land at 34
“100 minus your age”A long-standing rule of thumb
66%
“110 minus your age”A common update as people live longer
76%
“120 minus your age”The version Jack Bogle advocated
86%
Vanguard target-date pathRetiring at 65, the fund holds this at 34
90%
Most stocks within your loss limitWorst year −29.8% (1931)
66%
At 34 the “minus your age” rules give 66%, 76% and 86% in stocks, and Vanguard’s target-date path 90%. Your 30% loss limit allowed up to 66% in the record. They answer different questions: the rules and the fund follow your age, your limit follows how big a fall you can bear.
What each mix did, 1928 to 2025
| Stocks | Average year | After inflation | Worst year | Years with a loss | Worst 10 years |
|---|---|---|---|---|---|
| 0% | 4.5% | 1.5% | −17.8% (2022) | 20 of 98 | 0.1% a year |
| 10% | 5.3% | 2.2% | −17.8% (2022) | 16 of 98 | 1.4% a year |
| 20% | 6.0% | 2.8% | −17.9% (2022) | 15 of 98 | 2.7% a year |
| 30% | 6.6% | 3.5% | −17.9% (2022) | 15 of 98 | 3.0% a year |
| 40% | 7.2% | 4.1% | −19.1% (1931) | 20 of 98 | 2.9% a year |
| 50% | 7.8% | 4.6% | −23.2% (1931) | 20 of 98 | 2.4% a year |
| 60% | 8.3% | 5.1% | −27.3% (1931) | 21 of 98 | 1.8% a year |
| 66% (your limit) | 8.6% | 5.4% | −29.8% (1931) | 23 of 98 | 1.4% a year |
| 70% | 8.8% | 5.6% | −31.5% (1931) | 24 of 98 | 1.1% a year |
| 80% | 9.3% | 6.1% | −35.6% (1931) | 25 of 98 | 0.3% a year |
| 90% | 9.7% | 6.4% | −39.7% (1931) | 26 of 98 | −0.6% a year |
| 100% | 10.0% | 6.8% | −43.8% (1931) | 26 of 98 | −1.7% a year |
A 66% stock mix grew 8.6% a year on average (5.4% after inflation), lost money in 23 of 98 years, and its worst 10-year stretch (1929 to 1938) returned 1.4% a year. Its deepest fall from a year-end peak was −44.3% (1928 to 1932). The rest is held in 10-year Treasury bonds.
Two published paths through retirement
Vanguard target-date path“110 minus your age”Rising path (Pfau and Kitces)
Vanguard’s path holds 90% in stocks at 34, 50% at 65 and 30% from 72. Pfau and Kitces found in their tests that raising stocks after retirement did better than lowering them: start at 30% stocks at 65 and add a point a year to 60% at 95, so the portfolio holds the fewest stocks in the years a bad market does the most damage. It is one research finding and is still debated.
The five worst years for a 66% stock mix
| Year | Stocks | Bonds | Your mix | On $420,000 |
|---|---|---|---|---|
| 1931 | −43.8% | −2.6% | −29.8% | −$125,172 |
| 1937 | −35.3% | 1.4% | −22.9% | −$95,984 |
| 2022 | −18.0% | −17.8% | −18.0% | −$75,459 |
| 2008 | −36.6% | 20.1% | −17.3% | −$72,618 |
| 1974 | −25.9% | 2.0% | −16.4% | −$68,960 |
The worst year, 1931, took 29.8% off a 66% stock mix (stocks −43.8%, bonds −2.6%). 2 of the five worst years came before 1950. A record of 98 years holds only a handful of separate bad stretches, so a worse year than any of these is possible.
One published way to spread a 66% stock mix around the world
U.S. stocks
40%
International stocks
26%
U.S. bonds
24%
International bonds (hedged)
10%
Vanguard’s target-date funds hold stocks 60% US and 40% international and bonds 70% US and 30% hedged international; on 66% stocks that is 40% US and 26% international stocks, 24% US and 10% international bonds. Vanguard’s research found 30% to 40% international stocks gave more than 95% of the benefit of full market-weight diversification. The history on this page is US only, so it cannot say how international stocks did.
How it's computed
FORMULA
Mix return in a year = stock share × S&P 500 return + (1 − stock share) × 10-year Treasury bond (or bill) return, rebalanced every January
Most stocks within your limit = the largest stock share such that no calendar year from 1928 to 2025 lost more than your limit
“K minus your age” = K − age for K of 100, 110 or 120, kept between 0 and 100 percent stocks
Vanguard path = 90% stocks 25 or more years before retirement, a straight line to 50% at retirement and to 30% seven years after
Rising path (Pfau and Kitces) = 30% stocks at retirement, plus 1 point a year, 60% after 30 years
- The record is 98 calendar years of US large-company stocks (the S&P 500, dividends included), one 10-year Treasury bond and 3-month Treasury bills, from Damodaran’s annual returns (read September 29, 2026). It has only a few separate bad stretches, so the worst year in it is a floor on what happened, not on what can.
- Costs, taxes and the time between rebalances are not counted. Falls from a year-end peak use year-end values, so they understate falls inside a year. Bonds are one 10-year Treasury, not a bond fund or TIPS.
- Your loss limit is your own. The page finds the most stocks that kept every year of the record inside it, and it does not judge whether that limit is right for you.
- The target-date path is Vanguard’s published glide path read as years from retirement, with a straight line between its published points. The “minus your age” rules are rules of thumb, not calculations. The international split is one fund family’s, not a market-cap calculation.
- A portfolio with no stocks and only 10-year Treasury bonds still lost in some years, including 2022, so limits below about 18% have no bond answer. Cash in Treasury bills never lost in a calendar year.
WORKED EXAMPLE · SAMPLE NUMBERS
Limit 30%, the rest in 10-year Treasury bonds. At 66% stocks the worst year was 1931: 66% × −43.8% + 34% × −2.6% = −29.8%, inside 30%. At 67% stocks the worst year was −30.2%, past the limit. At 34, 110 − 34 = 76% stocks; Vanguard’s path at 31 years before retirement is 90%.
SOURCES
[1]Historical Returns on Stocks, Bonds and Bills: 1928-2025Aswath Damodaran, NYU Stern[2]Target-date fund glide pathVanguard[3]Target-date fund portfolio construction (equity and bond splits)Vanguard[4]Reducing Retirement Risk with a Rising Equity Glide Path (Journal of Financial Planning, January 2014)Wade D. Pfau and Michael E. Kitces, Financial Planning Association[5]Portfolio rules of thumb: do they stack up?Morningstar[6]Should you follow the “100 minus your age” rule?justETFHSBuilt by Hussain Sehorewala · checked against worked examples · Sep 29, 2026
Keep this number honest as your life changes.
Put it on your Money Map and it re-runs as you change the seven numbers. It stays in this browser, and the calculator stays free.
Questions about this result
No. It shows what US stock and bond mixes did from 1928 to 2025, what three published rules or paths hold at your age, and the most stocks that kept every year of the record within a loss you say you can bear. Your age, other savings, income, taxes and nerves decide the rest, and a fee-only planner can weigh them with you.
Rules of thumb that put your age in bonds and the rest in stocks: at 34, 100 minus 34 is 66% stocks, 110 minus 34 is 76% and 120 minus 34 is 86%. The older 100 dates from when bonds paid more; 110 and 120 followed as people lived longer, and Jack Bogle advocated 120. They ignore your other assets, income and how much loss you can bear.
Ten-year Treasury bonds lost 17.8% in 2022, and every mix with bonds shared some of that loss, while Treasury bills never lost in a calendar year. At a 30% loss limit the most stocks was 66% with bonds and 70% with cash. Cash also earned less over the record: 3.4% a year against 4.5% for the bond.
The record here is US stocks only, so it cannot say how international stocks did. As one published example, Vanguard’s target-date funds hold stocks 60% US and 40% international and bonds 70% US and 30% hedged international, and its research found 30% to 40% international stocks captured more than 95% of the benefit of full market-weight diversification.
Target-date funds such as Vanguard’s lower stocks as retirement nears and keep lowering them after it. Pfau and Kitces tested the opposite after retirement: start at 30% stocks and raise it a point a year to 60% over 30 years, on the idea that a bad market in the first years of retirement does the most harm. They found it did better than a steady 60% in their tests. Kitces’ “bond tent” is the version that also cuts stocks in the last years before retirement, so the bond share peaks around the retirement date and stocks are rebuilt afterwards. It is a research finding that is still debated.
It is the longest consistent US record, 98 calendar years, but it holds only a few separate bad stretches: 1929 to 1932, 1937, 1973 and 1974, 2000 to 2002 and 2008. The future does not have to repeat any of them, and a worse year than the worst here is possible. Use the record to see what a fall of a given size looks like, not to predict.
THE LEDGER · 12 minThe Advanced 2026 Tax Strategies That Create Generational Wealth
THE LEDGER · 8 minWhy Generic AI Falls Short for Financial Advice - And How Specialized Knowledge Changes Everything