How we check every number
Every figure that can change carries the date we last checked it and a link to its source. Here’s where the data comes from, who checks it, and how the quiz and the calculator use it.
| Data | Source | Last checked |
|---|---|---|
| EPA range, city range & efficiency | Source: fueleconomy.gov | Last checked: Sep 27, 2026 |
| Highway range at 75 mph | Source: Car and Driver’s 75-mph tests; our estimate where none exists (method below) | Last checked: Sep 27, 2026 |
| Winter range | Source: Our estimate from Recurrent fleet data and the NAF winter test — method below | Last checked: Sep 27, 2026 |
| Prices (MSRP + destination) | Source: Automaker sites | Last checked: Sep 27, 2026 |
| Charge port & adapters | Source: Automaker sites & accessory stores | Last checked: Sep 27, 2026 |
| Calculator defaults (electricity, gas, public charging) | Source: EIA (Jul 2026 data) · AAA | Last checked: Sep 24, 2026 |
Who checks the data, and when
The car data is researched and checked by CAM Software LLC, which publishes this site. Each value is recorded with the page it came from, what kind of source that is (the automaker, the EPA, or an independent test) and the date it was checked.
We re-check a car’s figures whenever we update that car, and whenever someone reports a correction. We don’t run a fixed re-check schedule yet, so we don’t promise one: the “Checked” date on each page is the oldest check date among the figures on it, and it tells you how fresh they are. Confirmed corrections are dated on the corrections log.
What we show, and what we leave out
We show a value only when a primary source confirms it: the automaker’s own site or press release, fueleconomy.gov, or the independent test itself. If only a secondary source reports it (a review, a dealer listing, a forum), or it was inferred from another trim, we leave it off the page until we can confirm it. The row is simply missing, or says “Not confirmed”; we never fill the gap with a guess.
An unknown value never counts in a car’s favour: in the quiz it fails the filter it would be needed for and scores zero, and in Compare it can’t win a row.
The range numbers
We show up to four: EPA combined (the sticker), a highway number at 75 mph, city range from the EPA city cycle, and a winter estimate. Anything that is our estimate rather than a published figure carries “~”, and winter is always shown as a range.
City range. The EPA’s city-cycle range for that trim, from the EPA data file on fueleconomy.gov (after the EPA’s own real-world adjustment). It is a mild-weather figure and isn’t printed on the window sticker. We don’t make a “winter city” number, because no source measures one.
Highway range. Where Car and Driver has run its steady 75-mph test on the same model with the same EPA rating (possibly another trim or model year), we show that result as a test, and the car page’s range panel names the exact version tested. Where it tested another version of the same generation, we scale its result by the two versions’ EPA combined ranges and label it as our estimate, with “~”. Some versions that share an EPA rating with the tested car are also shown this way, as estimates, which is the cautious choice. Where there is no usable test, our estimate is the EPA highway range × 0.91, the median ratio across 34 Car and Driver tests. Estimates are rounded to 10 miles and are never called a test.
How we estimate winter range. The winter number is our estimate, not a test of that car. It is a band for mixed everyday driving on a cold day with the heater on: the high end is a freezing day (32°F), the low end a 20°F day. We apply a factor to the car’s EPA combined range, based on published real-world winter data (Recurrent, Nov 2025) and whether the trim has a heat pump. When the automaker doesn’t confirm the heat pump for that trim, we use the fleet-average factor instead. Where an independent winter test (NAF) ranked the model against other cars on the same day, we adjust for that. Both ends are rounded outward to the nearest 5 miles, so the band never looks more precise than the evidence. Colder days, short trips and highway speeds can go lower.
How the quiz picks your cars
1. Filters. A car is dropped if it doesn’t fit your answers: its price with the destination fee is over your budget (a monthly budget is converted to a price at 72 months and 6.97% APR with $0 down: the commercial-bank rate for 72-month new-car loans in 2026 Q2, Federal Reserve G.19, Sep 8, 2026); it has too few seats; it can’t tow 3,500 lb if you tow; it doesn’t match your brand choice or a must-have; or its winter estimate is less than 1.5× your longest typical day. For mild winters we use the 32°F end of the winter band, otherwise the 20°F end. If an automaker doesn’t publish a destination fee, we assume $2,000, and the card says so.
2. Scores. Each remaining car is scored on winter range per dollar, range to spare over your day, efficiency, fast-charging time, Tesla Supercharger access, heat pump, highway range and cargo space. Each is scaled across the cars that qualified for you, and the weights depend on your charging situation and answers: more weight on fast charging and Supercharger access if you take long trips, on the heat pump if your winters are cold, on cargo if you haul. If you chose “Prefer brands I know”, established makes get a small bonus (0.03 on a 0–1 score) rather than a filter. Ties are broken the same way every time, so the same answers always give the same picks.
3. Three picks, one per model. Best overall (the top score), plus the fastest-charging car and the most range per dollar — or, if charging is your hurdle, the lowest energy cost and the fastest charging. A label only goes to a car that really wins it among the picks shown; otherwise the card says “Runner-up”.
4. When fewer than three fit. We loosen one thing at a time, in a fixed order — up to 10% over budget, then must-haves one by one, then up to 25% over budget, then a winter estimate down to 1.25× your day, never lower — and each car found this way says what it misses. Nobody pays for a place in the results.
How we decide “Great fit”
Your parking and charging answers come first. It’s a great fit when you can charge at home and it covers your day, long trips are monthly or less, and every pick’s winter estimate is at least 2× your longest typical day. It becomes “Works with planning” if any of these is true: you’d charge at work; you park in an apartment or condo spot (charging usually needs the landlord’s or HOA’s permission); you park on the street or in a shared lot, or aren’t sure where you’ll park, so home charging isn’t confirmed; you’d use a standard outlet and it adds fewer miles overnight than your longest day; you take long trips every week; or a pick’s winter estimate is under 2× your longest day. If you can’t charge where you sleep, or a standard outlet can’t keep up, we say charging is your hurdle, show what public charging would cost next to gas and a hybrid, and still show the EVs that make it easiest.
How the cost calculator works
Energy a year = miles × the car’s EPA combined efficiency (kWh per 100 miles) × your price, plus 10% charging losses on every kWh, split between home and public charging by the share you choose. The gas and hybrid rows are miles ÷ mpg × your gas price. Until you enter your own, prices are US averages: home electricity 18.31¢/kWh (EIA, Jul 2026), public EV charging stations 42¢/kWh and gas $4.48 (AAA, Sep 24, 2026). Your own rate can be quite different, which is why the calculator shows how much it matters. Insurance, maintenance and depreciation aren’t included. Nothing you type is stored or sent.
What we won’t do
- Take money from carmakers or dealers
- Sell your details as a lead
- Publish a number we haven’t checked
- Use an unconfirmed figure as fact