How the math works, and where to check it.

Investment growth

Future value of a starting amount plus regular contributions, compounded over time.

Here r is your yearly return and m is how often growth is compounded per year (yearly, monthly, quarterly, and so on). We convert whatever annual rate and compounding you choose into one effective monthly rate, then step the balance forward month by month. If an expense ratio is set, it is subtracted from r before the conversion. Contributions are added at the end of each month (the conservative, honest choice), and weekly or yearly contributions are normalized to a monthly amount so every schedule is compared on the same basis.

Substituting into the formula: FV = 20,000(1.005654)³⁶⁰ + 500 × [ ((1.005654)³⁶⁰ − 1) ÷ 0.005654 ] = $736,971, the exact figure the tool shows.

Cap rate

A property's income yield, independent of how it is financed: the fastest gut-check on a deal.

Because it leaves out financing, cap rate lets you compare two properties on equal footing regardless of how each is bought. It does not account for your loan, appreciation, or taxes; those belong to deeper tools. Use it as a screen, not a verdict.

Solving for a goal

When you set a target and ask "what would it take?", we invert the same formula above.

The investment calculator can solve for the return, contribution, starting amount, or number of years needed to reach a target balance. There is no separate formula for this: we search for the input value that makes the projected future value equal your goal. The method is bisection: we bracket the answer between a low and high guess, repeatedly test the midpoint, and keep the half that still contains the target, narrowing until the projected balance lands within a dollar of your goal (or up to 80 refinements). It is slower than a closed-form solution but robust: it uses the exact same projection function as everything else, so a solved answer and a forward calculation can never disagree.

Every calculator, one shared library

Why this page doesn't need to duplicate all 18 formulas for you to trust them.

Every calculator on the site (not just the two above) is a thin page that imports one pure, DOM-free function from a shared calculation library and renders whatever it returns. There is no second copy of the math anywhere, and no AI model computing or adjusting a figure; the function that runs when you use a tool is the only place its formula lives. That is a narrower, more checkable claim than "every formula is written out on this page" (it isn't, yet, for all 18 tools), but it's the reason a bug in one calculator's math can't silently diverge from what any documentation might say about it, worked-example or not.

Market-data methodology

Two separate scoring models, each published in full: pick the one you want.

Every market ranking comes from a documented model, not a black box, and every weight, transform, gate, grade band, and source is on the methodology page. There are two distinct models on two geographies: a City Market Score for 1,863 ranked cities (of 2,303 tracked) and a Rental Investor ZIP Score for ~20,000 ZIP Codes, and their numbers are not interchangeable. Both are built entirely from free, keyless public sources (Zillow Research, Census ACS/PEP/Building Permits, Redfin Data Center, BLS LAUS, the FBI Crime Data Explorer) and leave a market unranked rather than scored around a missing core input.

Assumptions & honesty

Every projection here assumes a steady return and steady contributions. Real markets move up and down, and a bad first decade lands very differently from a bad last one even at the same average; these tools do not model that sequence risk. Figures are nominal: they are not adjusted for inflation, so a dollar in year 30 buys less than a dollar today. Contributions are assumed to arrive at the end of each period. Reference rates shown in the tools (for stocks, bonds, savings, and so on) are long-run historical averages offered as editable starting points, not promises.

None of this is financial advice. It is math laid out in the open so you can make your own call. If a number here ever disagrees with what a tool displays, treat it as a bug and tell us; the two are supposed to be identical.