Investment backtest

Total and annualized returns answer different questions

Total return measures the gain or loss relative to invested capital. For a lump sum, CAGR expresses the starting and ending values as an equivalent annual compound growth rate. An 8% CAGR does not mean the investment rose by 8% in each year.

Later monthly contributions have less time invested. This page reports an annualized cash-flow return (XIRR) that accounts for contribution dates and amounts. Added deposits are not investment gains, and dividing total return by the number of years does not produce this annualized return.

Being above the initial capital does not mean a previous peak has recovered

This hypothetical lump-sum example is not historical performance for any asset. Both paths start at 100 and compare only four valuation points over two years. Dividends, taxes, costs and currency changes are excluded.

Path B falls from 150 to 75. Its ending value of 120 exceeds the initial 100 but remains below the earlier peak of 150. Maximum drawdown measures a decline from a prior peak, rather than a loss relative to initial capital. Recovery time refers to regaining the pre-drawdown level. The actual backtest uses stored daily valuations; this four-point example does not describe movements between those points.

For monthly investing, drawdown and recovery use a return series with the effect of contributions removed. The displayed loss amount is scaled from the first valuation; it is not the actual amount lost from an account.

Match currency and dividend settings before comparing

A dollar asset valued in won reflects both asset performance and historical exchange rates. If its dollar value rises 10% while KRW per USD falls 10%, the combined change is 1.10 × 0.90 = 0.99, a 1% loss in won. This example compares lump-sum starting and ending values and excludes monthly conversions and actual costs.

Excluding dividends omits dividend cash. Holding dividends keeps it as cash, while reinvestment buys more shares. The same asset and period can produce different results under these settings. Check currency, dates and dividend treatment before comparing an asset with its benchmark.

Before turning one start-date result into an investment rule

Starting just before a rise differs from starting just before a fall. Moving only the start date changes both entry timing and holding length. To examine different market windows with a consistent holding length, compare several periods of equal duration.

Missing observations or different listing dates can change the effective period. Read the dates shown in the results rather than assuming the requested dates were used unchanged. Historical outperformance does not establish future outperformance, and these results exclude taxes and trading costs.

Compare when to invest the same existing budget · Explore CAGR separately