About Payback Math

Solar panels, electric cars, heat pumps and home batteries are all sold on savings. The savings claim is usually the centre of the pitch — and it is usually made by someone who profits from you believing it.

Checking that claim is harder than it should be. The answer depends on your electricity price, your fuel price, your climate and your local export rules, and all four vary enormously between states. A payback figure calculated for a national average is close to meaningless for any individual household.

Payback Math does that calculation with local data, shows the workings, and publishes the result whichever way it lands.

4 calculators, each covering all 51 states and DC, built on the same energy-price dataset:

• Heat pump vs furnace — Is a heat pump cheaper than gas heating?
• EV vs gas — Is an EV cheaper to run than gas?
• Solar panels — Do solar panels pay for themselves?
• Home battery — Does a home battery pay for itself?

Each page opens with the answer, then lets you replace our assumptions with your own numbers and watch the result move.

Most published payback figures are optimistic for predictable reasons: they divide cost by first-year savings instead of simulating the years, apply one equipment efficiency to every climate, price solar production before real-world losses, ignore charging and round-trip losses, or add up two battery savings that the same battery cannot both deliver.

We model each of those honestly. The result is a set of less flattering answers — a home battery pays for itself inside its warranty in 0 of 51 states, and a heat pump is cheaper to run than gas in 17 — which is the point rather than a problem.

None of that is an argument against buying. It is an argument for knowing the real number first, and for buying the thing for the reason it actually delivers.

Every model, formula and constant is published on the methodology page. Every dataset is listed with its vintage, retrieval date and a link to the primary file on the data sources page, including an explicit list of the fields that are still internal estimates. Changes to data and to the models are recorded in the changelog, and the rules we work to are set out in our editorial policy.