Still working. The data server sleeps when idle, so the first load after a quiet spell can take up to about 40 seconds. This page will fill in on its own.
Still working. The data server sleeps when idle, so the first load after a quiet spell can take up to about 40 seconds. This page will fill in on its own.
Portfolio maintenance
About once a year, or not much more. Portfolio Lab’s backtest engine offers four rebalancing schedules — none, monthly, quarterly, annual — and its own regression tests confirm that switching a classic 60/40 mix from “never rebalance” to “rebalance every December” changes its long-run CAGR by less than 2 percentage points. Rebalancing exists to control risk, not to boost returns, and nothing in this site’s own numbers shows doing it more often reliably helps. Every extra rebalance outside a retirement account is also an extra taxable event — a cost the backtest itself doesn’t model.
Left alone, a mix drifts: the assets that ran the hardest end up the biggest share of the portfolio, quietly raising its risk. Rebalancing sells a slice of what rose and buys more of what lagged, resetting the mix back to the stock/bond split you actually chose. It trades a little of the winner’s further upside for keeping the portfolio’s risk where you meant it to be — a discipline tool, not a return booster.
The backtest engine resets a mix to target weights on whichever schedule you pick: every month, every quarter (March, June, September, December), every year (December only, so every allocation and benchmark rebalances in the same month), or never, letting it drift for the full backtest. The site’s own test suite specifically checks the “never vs. annual” case for a 60/40 US stock/bond mix over its full available history, and asserts the CAGR gap stays under 2 percentage points. For a simple two-asset mix, the frequency knob just doesn’t move long-run returns much.
Portfolio Lab’s backtests run pre-tax, with no trading costs and no advisor fee. A real taxable account doesn’t get that pass: rebalancing there means selling something that gained value, which can realize capital gains you now owe tax on. Rebalancing more often multiplies those taxable events for a return difference the site’s own tests can’t distinguish from zero. Annual, calendar-based rebalancing — the tool’s own default — gets almost all of the risk-control benefit at a fraction of the tax and trading friction.
Compare schedules yourself
Portfolio Lab’s backtest lets you switch between none, monthly, quarterly, and annual rebalancing on the same mix and see the growth chart change — no account needed.
Try quarterly rebalancing on a sample mix →Not reliably. Portfolio Lab's own regression tests check that a 60/40 mix rebalanced every December lands within 2 percentage points of CAGR of the same mix left to drift completely untouched — a gap easily swamped by which decade you happened to invest in. Nothing in the site's own numbers shows monthly or quarterly beating annual.
Yes, in the other direction. Portfolio Lab's backtests are pre-tax; in a real taxable account, every rebalance that sells a winner can trigger capital gains. Rebalancing more often adds more of those taxable events without a reliably better return, which is one more reason 'about once a year' is a sensible default rather than a compromise.
No. A target-date fund already rebalances itself internally on its own schedule, the same way Portfolio Lab's engine resets a mix to target weights. Build a rebalancing habit for anything you hold alongside it, not for the fund itself.
Educational, not advice. The schedules discussed here are the four Portfolio Lab’s backtest engine supports (none, monthly, quarterly, annual); real accounts can also use threshold- or band-based rebalancing, which this site doesn’t model separately.