Start with a clearly defined price-return experiment
Choose a basket, allocate 100% across positive weights and select a common date range. The illustrative presets are convenient inputs, not recommendations, index constituents or historical investment records. Dollar values assume verified USD-quoted securities and fractional shares. No deposits or withdrawals are modeled.
Know what the price series includes
The underlying API route uses FMP historical-price-eod/full. This tool consumes its close field as a split-adjusted price series; it does not silently substitute a dividend-adjusted field. Cash distributions, reinvestment and other corporate actions such as spin-offs are outside the model. A price-only bond or dividend-stock comparison can therefore differ substantially from total return.
Provider reference: FMP price adjustment definitions. Match an issuer or broker record when corporate actions matter.
Allocation, observations and rebalancing
Initial dollars divided by the first common close determine fractional units. Without rebalancing, units remain constant. With rebalancing, the first common close of each new calendar period values the old holdings and resets units to target weights for subsequent observations. This same-close model omits spreads, execution delay, transaction costs and taxes.
Only dates available for every holding and SPY are used. Results show the actual interval and the number of dates omitted from the union of returned observations. The tool cannot prove that the provider returned every market session; missing data and delisted securities can distort a historical comparison.
Interpret the summary numbers
Price return is ending value divided by starting value minus one. Annualization uses elapsed days divided by 365.25. Drawdown is the largest observed percentage decline from a running peak, not a limit on future losses. Calendar-year rows include the move from the preceding observed year-end close; edge years can be partial.
Volatility uses sample standard deviation of simple close-to-close returns times the square root of 252. The illustrative zero-risk-free Sharpe uses the arithmetic mean of those returns, not CAGR. Annual scaling assumes regular observations and does not correct serial correlation. Method reference: William Sharpe’s explanation of the ratio and time dependence.
For a synthetic 50/50 allocation of 1,000 USD, prices moving from 100/100 to 110/90 leave value at 1,000 USD before costs. Returning to equal weights at that close changes the quantities held, not the portfolio’s immediate value. Subsequent outcomes depend on subsequent prices.