Winter EV range: why three cold-weather tests report different losses

A winter range number only means something once you know what it was measured against. Norway's El Prix test put cars an average 38 percent below their WLTP figures in January. A Canadian road test found losses as small as 14 percent. Both can be true for the same car, and a buyer who mixes the baselines will either panic or plan badly.

This is a reading guide built from published tests and datasets, not our own testing. We did not drive any of these cars.

The Norwegian El Prix, run by the national motoring club NAF with Motor magazine, is the harshest. The FIA reports 24 of the newest EVs on the Norwegian market were driven in Folldal over two days in January 2026, in temperatures as low as minus 32C. Range fell by an average of 38 percent against WLTP, the European lab benchmark. Every car fell short. The cars start fully charged and follow the same route through urban traffic, high-speed roads, mountain roads and snow. The last stretch is repeated until a car can no longer hold the required speed, according to Drive Tesla Canada's account of the event.

In that field the Lucid Air Grand Touring managed 520 km, the Mercedes-Benz CLA 421 km and the Tesla Model Y 359 km.

The Canadian Automobile Association (CAA) test was milder and used a different benchmark. In early February 2025, it drove 13 EVs on public highways from Ottawa toward Mont Tremblant at minus 7C to minus 15C, running each to empty and comparing the result with the Natural Resources Canada rating. By subtracting the published loss from 100 percent, the Chevrolet Silverado EV and Polestar 2 retained about 86 percent. The Hyundai Ioniq 5 retained about 64 percent, the Toyota bZ4X 63 percent and the Volvo XC40 Recharge 61 percent. The Silverado started at 73 percent charge and the F-150 Lightning at 89 percent, so CAA adjusted those two results.

The third yardstick is Recurrent's owner data, which measures each car against its own best real-world range, not a lab or government number. In Recurrent's February 2025 analysis of more than 18,000 cars across 20 models, as reported by CleanTechnica, heat-pump cars lost roughly 10 percent in winter. A later Recurrent release, relayed by ThinkEV, put heat-pump models at 83 percent of ideal-weather range at freezing, against 75 percent for resistive heating. Selectra's summary of the same body of work gives an average of 78 percent retained at 0C, with the best model at 88 percent and the worst at 69 percent.

WLTP, NRCan ratings and a car's own best week of driving are different baselines. A car that beats its optimistic lab rating in summer and trails it in winter looks worse against WLTP than against itself. Temperature also differs by tens of degrees between the studies, and the El Prix route is built to push cars to empty in snow.

Treat the three as bracketing the problem: cold costs a typical car somewhere between a tenth and well over a third of its rated range, depending on how cold, how fast and compared with what.

Geotab's analysis of 4,200 cars and 5.2 million trips, summarised by Selectra, estimates cabin heater draw at 3,000 to 5,000 watts and finds that heating the cabin matters much more than the battery's own chemistry until temperatures fall to around minus 18C. Argonne National Laboratory's figure for blown-air heaters is 2,000 to 4,000 watts. Selectra puts a heat pump's output at three to four units of heat per unit of electricity, which fits Recurrent's observation that a Tesla Model 3 with a resistance heater lost about twice as much range as the heat-pump version. Heat pumps lose efficiency below roughly 14F (minus 10C). Heat-pump hardware helps on average, but averages hide model-to-model spread.

- Pick the baseline. If a dealer quotes WLTP or EPA range, expect a bigger gap at minus 15C than at 0C. Ask which rating a winter figure uses.
- Look at the model's own test. CAA's spread, from 61 to 86 percent retained in one afternoon's conditions, shows how much the model matters.
- Precondition while plugged in. If the car supports it, warm the cabin and battery while it is connected to a home charger, so the initial heat comes from the grid rather than the pack.
- Allow charging time. Cold batteries accept power more slowly. In El Prix, FIA says most cars came close to their stated 10 to 80 percent times when the battery was preconditioned. In CAA's test, 15 minutes at a 350 kW charger added about 100 km on average.
- Keep a margin. Plan trips on the cold figure, not the sticker.

CAA's Ian Jack has said the association wants a standardised Canadian EV label showing winter performance, not one average. Norway's NAF president, Tor Arne Litlere, told the FIA that "situation based knowledge about the deviations from WLTP range is a true value." Until such labels exist, the buyer has to do that reconciliation.

These tests cover earlier model years and specific trims, and some vehicles have since had software or hardware updates. The El Prix and CAA results are from different winters and countries. Recurrent's sample sizes vary between its releases, and we could not retrieve its original report pages, so those figures are as relayed by the outlets named above. Check the current test for your exact model and market before relying on any single number.