How to backtest your portfolio (free, in 2 minutes)
A backtest answers one question: if I had held this portfolio from date X to date Y, what would have happened? Not what should happen, not what an average suggests — what actually did, using real prices and real dividend payments.
Step 1 — build the portfolio
On the Portfolio Calculate homepage, search any US stock or ETF and add it with a share count. The weights you set are the weights the backtest uses. A starter portfolio is preloaded so you can see how everything behaves before adding your own holdings.
Step 2 — set the scenario
In the "Backtest this portfolio" panel, pick the date range (data reaches back decades for older stocks), a starting amount, and optionally a monthly contribution. Two switches matter more than people expect:
- Reinvest dividends (DRIP) — on means every payout buys more shares; off means dividends pile up as cash.
- Rebalancing — "none" lets winners run; yearly or quarterly trims winners back to your target weights.
Step 3 — read the results honestly
The end value gets the attention, but the stats row is where the understanding lives: return/yr is your money-weighted annual growth; volatility is how rough the ride was; max drawdown is the worst peak-to-trough fall you would have had to sit through; Sharpe and Sortino tell you how much return you earned per unit of risk; and the benchmark line shows whether simply buying an index fund would have done better.
Step 4 — share or challenge it
The Share link button encodes the entire backtest into a URL. Send it to a friend, post it in a forum, or save it to revisit later — anyone who opens it sees exactly your scenario, live.
Run your first backtest →What a backtest can and can't tell you
It can tell you how strategies behaved: how much pain a 100% stock portfolio inflicted on the way to its returns, how dividends compounded, what rebalancing did and didn't do. It can't tell you what comes next — markets change regimes, and the best-looking backtest is often just the one that fit the last decade. Use backtests to understand behavior, not to chase the highest number.