Founding Member pricing: 40% off for life. Limited to 100 founding members. See plans →

Backtest a portfolio: any mix of stocks, bonds, gold, cash and more

Choose a mix of nine building blocks and see what it returned, how far it fell and how it did in 2008, 2020 and 2022.

Weights (%)
Presets

Total return (dividends reinvested), rebalanced monthly, no costs or taxes. Data to 2026-06.

Return a year–
Worst fall from a high–
Volatility · Sharpe–
2008 · 2020 · 2022–

Year by year: your mix vs US stocks alone
YearYour mixUS stocksDifferenceMix, worst fall in the year
Want to test a rule, not just a mix?This page holds every mix through everything. The Backtest Templates (Excel + Python, €79) let you test a trend-following rule on any market yourself: strategy vs buy & hold, worst fall, time invested, a robustness check across settings. Any Yahoo Finance symbol, your own data, no coding needed for the Excel version.
See the templates →

What this backtest shows

A portfolio backtest answers one question honestly: if you had held this mix, rebalanced every month, what would have happened? Not what will happen. The result is four numbers that matter (return a year, volatility, the worst fall from a high and how long it took to recover) and the behaviour in the three stress years every investor should know: 2008, when stocks halved and bonds rose; 2020, the fastest crash on record; and 2022, when stocks and bonds fell together.

The building blocks are total-return series of the main ETFs (dividends and interest reinvested), the same data behind our Research Terminal. Cash is the US Treasury bill. The test starts in the first month for which every chosen block has data, so a mix with gold starts in late 2004 and a mix with Bitcoin in 2017; the start date is shown with every result.

How to read the numbers

Return a year (CAGR) is the compound growth rate over the whole period. Volatility is how much the monthly returns swung, annualised; roughly, a 10% volatility means a typical year lands within ±10% of the average. The Sharpe ratio is the return above cash divided by the volatility: the return you got per unit of risk, and the fairest way to compare two mixes. The worst fall is the largest drop from a high to the following low, the number that decides whether you would have held on; our max drawdown guide explains why it matters more than volatility for most people.

What a backtest cannot tell you

Three things. First, the period: 1993–2026 contains two 50% stock crashes, a 40-year bond bull market that ended in 2022 and the best decade gold ever had. A different window would give different numbers; that is why the 60/40 guide goes back to 1872. Second, costs and taxes: none are deducted here; a fixed mix with monthly rebalancing trades little, but every sale can trigger tax. Third, and most important, the past is not a forecast. Use the backtest to understand how a mix behaves in crises, not to pick the one with the highest return in hindsight. The Portfolio Crash Test does the crisis part for twelve falls since 1955 with a longer bond and gold record.

From a fixed mix to a process

Every mix here is held through everything: 2008, 2022, the lot. That is the simplest process and a good one. The Research Terminal shows the alternative we research: the same building blocks, but each one held only while its own trend is up, with the recorded signals for 30 markets. Free account, no card.

Frequently asked questions

How do I backtest a portfolio for free?

Enter the percentage for each building block above (they must add up to 100), and the page shows return a year, volatility, Sharpe ratio, worst fall, the 2008, 2020 and 2022 results, the growth and drawdown charts and every calendar year. The data are total returns of the main ETFs, rebalanced monthly, from 1993 (or from the first month all chosen blocks exist).

What is a good backtest result?

There is no single good number, because return and risk trade off. Over 1993–2026 a 100% US stock portfolio returned about 10% a year with a worst fall of about 55%; a 60/40 mix gave up about two points of return for a worst fall of roughly half that. Judge a mix by whether you could have held it through its worst fall, then by its Sharpe ratio.

Why does the backtest start in 2004 or 2017 and not in 1993?

Because a mix can only be tested from the first month for which every block has data: the gold ETF starts in November 2004, the long Treasury ETF in 2002, the emerging-markets ETF in 2003, Bitcoin in our data in August 2017. The start date is shown with every result.

Does the backtest include dividends, costs and taxes?

Dividends and interest are included (total return). Costs and taxes are not deducted: ETF fees are already inside the ETF prices, but trading costs and the tax on dividends and sales are yours to add. Monthly rebalancing of a fixed mix trades little.

What is monthly rebalancing?

Once a month the portfolio is set back to the target weights: after a strong month for stocks you sell a little and buy the blocks that lagged. It keeps the risk where you decided it should be and is what makes a fixed mix a strategy rather than a drift.

How is this different from Portfolio Visualizer?

It is simpler: nine building blocks, monthly rebalancing, the numbers that matter, no account needed. For the stress-test view (what the mix did in twelve crashes since 1955) use our Portfolio Crash Test; for 30 markets with trend signals and a recorded live record, the free Research Terminal.

Free Research Terminal

From assumptions to real numbers, free

  • Strategy vs buy & hold for the S&P 500, Nasdaq-100, gold, oil, Bitcoin and Nvidia: return, Sharpe ratio, worst fall
  • Hedge Health: are bonds protecting stocks right now?
  • Drawdowns, correlations, momentum and the glossary for every term
Open your free account →Free forever · no credit card · 30 secondsOr run the free Portfolio Crash Test →

Go deeper: in-depth guides

More calculators

Illustration only — not investment advice and not a forecast. Real returns vary from year to year and can be negative. Disclaimer.