How we calculate

What kind of number is it?

We label figures by what they are, because the difference decides how much weight they carry:

Measured. Official historical data from a statistical agency — for example the EIA's residential electricity price for a stated month.
Projected. A forecast published by that same kind of body. We never compute a cost from a projection without saying so.
Specification. A manufacturer's stated rating for a specific product.
Industry range. A spread compiled from published sources, presented as a range rather than a point estimate.
HomeEnergyX calculation. Arithmetic we did ourselves, from inputs named on the page and assumptions defined here.

Charging efficiency

Not all the energy your meter records reaches the battery. Losses occur in the vehicle's onboard AC-to-DC conversion, in the cabling, and in battery thermal management.

We assume 88% — that is, 12% losses — unless an article states otherwise.The realistic range is roughly 10 to 15%, and it is worse in cold weather when the pack has to warm itself before it will accept a charge. We treat 88% as a mid-range assumption, not a measurement of any particular car.

The consequence: we compute cost from metered energy, not from battery capacity. A 75 kWh battery takes about 85 kWh at the meter.

Cost per mile

cost per mile = electricity rate ÷ (vehicle efficiency × charging efficiency)

With the rate in cents per kWh and efficiency in miles per kWh, the result is cents per mile. We state the rate and the vehicle efficiency next to every result, because a cost-per-mile figure without both is not checkable.

Usable battery capacity

Where a manufacturer publishes both a gross and a usable capacity, we useusable, and say which we mean. Charging a "75 kWh battery" from empty to full means putting 75 kWh of usable energy in.

Circuit loading and the 80% rule

EV charging is treated as a continuous load in US installations — a load expected to run at maximum for three hours or more. A continuous load is limited to 80% of the circuit's rating. That is why a 40 A circuit supports 32 A of charging output, and a 60 A circuit supports 48 A.

We use this to explain why the two numbers differ. We do not use it to size your installation: conductor sizing depends on the equipment listing, the installation method, ambient temperature, distance and the local authority, and that is an electrician's calculation, not a website's.

Voltage assumptions

United States: 240 V nominal for Level 2, 120 V for Level 1, 60 Hz split-phase.
Europe: 230 V at 50 Hz, single- or three-phase.
Ecuador and much of Central America: 120/240 V at 60 Hz, the US arrangement.
Chile, Argentina, Uruguay: 220 V at 50 Hz, the European arrangement.

Real supply voltage sags under load. Actual delivered power is commonly a few percent below the nominal calculation, and we do not model that — treat our power figures as slightly optimistic.

Rounding

Cost per mile: one decimal place, in cents.
Money: to the nearest dollar above $100, to ten cents below it.
Energy: to the nearest kWh above 10 kWh.
Charging times: to the nearest minute, from unrounded inputs.

Because we round for display but compute from unrounded values, recomputing a result from the rounded figures we printed can land a fraction away from ours.

Currency conversion

Cross-country comparisons are converted to US cents at an indicative rate at the time of writing, and labelled as approximate. Rates move; the ranking of countries is the durable part of those comparisons, not the exact converted figure. Ecuador is dollarised, so its figures need no conversion.

What we do not do

We do not bench-test chargers, and we do not publish rankings dressed up as reviews. Where a figure is an illustration rather than a measurement, the text says so. If you find a number on this site you cannot trace back to one of the definitions above, that is a bug — tell us and we will fix it and log the change on the corrections page.