Portfolio backtest — what a mix actually returned

Each fund's calendar-year returns are multiplied together in sequence from a common starting index of 100, with each year plotted at its year-end. The annualised figure is the geometric mean — the total growth factor raised to one over the number of years — not the average of the yearly returns, which would overstate the outcome whenever the returns varied. A reference line deflates the same path by your country's consumer price index, so the distance between the two lines is the part of the gain inflation took.

The observations are whole calendar years, which rules out any statistic that needs to see inside one. The period is whatever the selected funds have in common, so adding a young fund can shorten the window for everything else — check the years actually plotted rather than the period button you pressed. Nothing is deducted for tax, broker commission or the spread, and past compounding is a record of what happened rather than a projection: the simulation tools exist for the forward-looking question.

Why is there no maximum drawdown?

Because it cannot be computed honestly from yearly data, and a wrong number would be worse than none. The trough almost always happens inside a year: global developed equities ended calendar 2020 up about sixteen percent while falling roughly a third from peak to trough that February and March. A drawdown measured from year-end points would report close to nothing for 2020, which is the most misleading possible answer. The worst single calendar year is shown instead, because that is what this data genuinely supports.

Why is the annualised return lower than the average of the yearly returns?

Because losses and gains are not symmetric. A year of minus fifty percent followed by a year of plus fifty percent averages zero but leaves you down twenty-five, so the arithmetic mean describes a path nobody travelled. The geometric mean answers the question you actually care about — the constant rate that would have produced the same ending balance — and it is always the lower of the two whenever returns vary at all.

Why does adding a fund sometimes shorten the chart?

Because the comparison is restricted to years every selected fund has published, and a fund that launched recently has fewer of them. That is the price of a like-for-like comparison: the alternative is letting one line carry years the others never saw, which makes the older fund look better for no reason other than age. If the window collapses further than you want, deselect the youngest fund.

• Annual returns calendar — which fund led which year
• Rolling returns — how much the answer depends on when you started
• Monte Carlo simulation — the range of outcomes, not a single line