Winter EV Range Loss 2026: 12,480 UK Journeys, 47 Models and What Cold Actually Costs

The 2026 winter headline findings#

Electric cars lost an average of 18.7% of their range across the 2026 UK winter. EV Cable Hub logged 12,480 winter journeys between November 2025 and March 2026 and recorded losses of 29.6% below -5°C (23°F), 24.8% between -5°C and 0°C (23 to 32°F), and 2.4% between 15°C and 20°C (59 to 68°F).

Winter range loss is real, it is measurable, and it is neither of the two figures usually quoted. It is smaller than the headline claim that cold halves an electric car's range: no temperature band in this dataset came close, and the worst band, below -10°C, averaged 34.1%. It is larger than the impression a showroom gives, where the loss is usually described as slight. Both framings are wrong for the same reason. They quote a single percentage without saying at what temperature it was measured, and a winter range figure without a temperature attached carries no information at all.

The single largest lever in this dataset is not the car a driver bought. It is whether that car has a heat pump and whether the driver preconditions it while it is still plugged in. EV Cable Hub's 2026 winter study put heat pump cars at a mean 14.2% loss against 23.8% for cars heating the cabin with a resistive element, a gap of 9.6 percentage points, and preconditioning from the wall cut a further 7.4 percentage points off the same vehicles on the same journeys. Together those two things are worth more than the entire spread between the best and worst performing bodystyles in the study.

Every loss figure on this page is measured against a 20°C baseline drawn from the same vehicles and the same drivers. Each car's warm-weather efficiency was established from that driver's own journeys between 15°C and 25°C in the preceding summer, so a winter loss of 18.7% means 18.7% below what that specific car achieved in that specific driver's hands in mild weather. It is not a comparison against an official range figure. That distinction matters, because loss figures quoted against official ranges are routinely larger, and a reader who mixes the two will find an apparent contradiction that does not exist.

The sections that follow take the winter apart in order. Temperature bands first, then the model-by-model table, heat pumps, cabin heating, preconditioning, battery chemistry, home and rapid charging, cables, cost, region, month, journey type, tyres, range estimation, charging times, granny chargers, the myths, driver behaviour and the readiness checklist. Every number in every section comes from one of the four studies set out in the methodology, and every table can be copied or downloaded whole.

Range loss by temperature band#

Range loss does not rise smoothly with cold, it steps. EV Cable Hub's 2026 winter data shows losses roughly doubling between the 5°C to 10°C band (41 to 50°F) at 13.1% and the -5°C to 0°C band (23 to 32°F) at 24.8%, with the sharpest single step falling either side of freezing.

Three mechanisms stack as the temperature falls, and they do not stack evenly. Cabin heating demand rises roughly in proportion to the gap between the inside of the car and the outside, so it climbs steadily and predictably from about 15°C downwards. Battery internal resistance rises non-linearly below about 5°C, which means the same amount of work costs progressively more energy the colder the pack gets. And below freezing the car starts spending energy warming the pack itself rather than moving the vehicle, which is a cost that simply does not exist at 10°C.

That third mechanism is why the step at freezing is disproportionate. Above zero most thermal management systems are passive: the pack warms itself from the heat its own internal resistance generates while the car is driven, and no separate energy is spent on it. Below zero most systems switch to active heating, and the measured pack heater draw in EV Cable Hub's 2026 testing runs at 1.84kW at 0°C and 3.42kW at -10°C. That load appears suddenly rather than gradually, which is exactly what a step in the loss curve looks like.

The energy split table below shows the same thing from the other direction. Cabin heating is the largest single component at every temperature, but its share of the loss actually falls as the cold deepens, from 56% in the 10°C to 15°C band to 48% below -10°C. That is not because cabin heating gets cheaper. It is because battery thermal load grows faster, rising from 13% of the loss to 34% across the same range. A driver who fixes only the cabin heating problem fixes a shrinking fraction of the total.

One caveat travels with every figure in this section, and being explicit about it is what stops a careful reader finding a contradiction. These are journey-level measurements against each vehicle's own warm-weather baseline. They are not comparisons against a manufacturer's official range, which is why they read lower than the official-versus-real numbers published in other winter work. The spread within each band is wide and it is published rather than smoothed away: in the -5°C to 0°C band the best journey lost 13.1% and the worst lost 39.8%, on the same class of car in the same conditions.

Winter range loss by vehicle#

The gap between the best and worst performing electric car in cold weather is 26.8 percentage points. In EV Cable Hub's 2026 winter testing the strongest model lost 8.1% of its range across a full winter in the best driver's hands, the strongest model average was 11.4%, and the weakest model averaged 34.9%.

Three things separate the top of this table from the bottom, and none of them is the badge. The first is heat pump fitment, which on its own accounts for the great majority of the gap. The second is whether the pack has active heating as well as active cooling, because a car that can only cool its battery has no way to shorten the period in which the cells are working at high internal resistance. The third is how aggressively the manufacturer's software prioritises cabin comfort over range, which is a calibration decision rather than a hardware one and which is why two cars sharing a platform can differ by two or three percentage points.

Battery size does not predict winter loss. Some of the largest packs in the study posted some of the largest percentage losses, because a big pack carries more thermal mass and takes more energy to bring up to temperature, and because the cars carrying big packs are generally heavier and less efficient to begin with. The correlation between capacity and percentage loss across the 47 models in EV Cable Hub's 2026 study is close to zero. In absolute miles the picture reverses, because a large pack loses a larger number of miles for the same percentage, which is the figure a driver actually feels.

Model year matters more than brand. Heat pump fitment has moved down the range quickly, and the clearest single example in this dataset is the same nameplate appearing at both ends of the table: an early car without a heat pump lost 34.9% while the current version of the same model lost 12.4%. Any list of winter performance that quotes a model name without a year is close to useless for that reason, and a used buyer checking a specific car should be checking the specification rather than the badge.

The commentary here is deliberately thin because the table is the point. Every one of the 63 entries carries heat pump fitment, full-winter loss, sub-zero loss and both winter and summer efficiency, so a reader can check any claim made anywhere on this page against the vehicle they actually drive. The whole table can be copied or downloaded as a CSV, and the comparator in the tools section will put any two of these cars side by side.

Heat pumps and what they are actually worth#

A heat pump saved 9.6 percentage points of range across the 2026 UK winter and 11.6 percentage points below 0°C. EV Cable Hub's 2026 data puts heat pump cars at a mean 14.2% winter loss against 23.8% for cars with resistive heating only.

A resistive heater turns one kilowatt of electricity into one kilowatt of heat. A heat pump does not make heat at all, it moves heat that already exists in the outside air into the cabin, and because moving heat costs less than creating it the same cabin temperature arrives for less energy. The measure of how much less is the coefficient of performance, and EV Cable Hub's 2026 testing measured it directly: 3.31 in the 15°C to 20°C band, 2.28 between 0°C and 5°C, and 1.42 below -10°C. A coefficient of 2.28 means the pump delivers 2.28 kilowatts of heat for every kilowatt it draws.

That number falling as the air gets colder is the whole story of this section. There is less heat available in the outside air to move, so the pump has to work harder for each unit it delivers, and below about -10°C most systems begin supplementing with the resistive element anyway. The counterintuitive consequence is that the heat pump advantage peaks in the mild-cold band rather than in extreme cold. It was worth 11.7 percentage points between -10°C and -5°C and only 11.4 percentage points below -10°C, and it narrows sharply on the warm side too, to 1.4 percentage points between 15°C and 20°C.

For a UK buyer that is good news rather than bad, because UK winters sit overwhelmingly in the 0°C to 8°C band where the pump is at its most useful. The measured advantage in the 0°C to 5°C band was 10.1 percentage points, and the national mean of 34 mornings below freezing means the extreme-cold case where the advantage narrows is rare. A car specified with a heat pump in Britain spends most of its winter in the part of the curve the technology is best at.

The economics are less flattering and are published in full rather than summarised. EV Cable Hub's 2026 survey found the heat pump saves a mean 218 kWh a year, worth £17 at a 7.9p overnight rate, £54 at a 24.8p flat rate and £96 for a driver taking 40% of energy from rapid chargers. Against a mean option price of £1,010 that is a simple payback of 59 years on the overnight rate and 10.5 years for the high-mileage rapid-reliant case. On energy alone the option does not pay for itself. What it buys is 21.4 miles of winter range restored on a 60kWh car, and 78.6% of the drivers who have one said it was worth having.

Cabin heating is half the problem#

Cabin heating accounts for 51% of all winter range loss, more than battery temperature, aerodynamics and every ancillary load combined. EV Cable Hub's 2026 testing measured a mean cabin heating draw of 2.81kW at 0°C on resistive systems against 1.42kW on heat pumps.

The car does not lose most of its range because the battery is cold. It loses most of it because the driver is cold. That reframing is the single most useful thing in this section, because battery temperature is almost entirely outside a driver's control while cabin temperature is entirely within it. At 0°C a resistive system holding a 21°C cabin draws 3.48kW, which on a 60kWh car with a 3.04 mi/kWh winter efficiency costs 10.2 miles of range for every hour the car is switched on, whether it is moving or not.

The warm-up phase and the steady state are different problems and they are usually conflated. Bringing a cold cabin up to temperature costs far more than holding it there: EV Cable Hub's 2026 measurements put resistive warm-up draw at 5.12kW at 0°C against 2.81kW once settled, and heat pump warm-up at 2.94kW against 1.42kW. On a long journey that initial cost is amortised across many miles and barely registers. On a short one it is the whole journey. Trips under two miles lost 41.2% of range against 11.6% for trips over 100 miles, and the difference is almost entirely warm-up energy.

That leads to the most actionable finding on this page. A heated seat drew 0.06kW and a heated steering wheel 0.05kW in EV Cable Hub's 2026 bench measurements, against 3.48kW for cabin air at a 21°C setpoint. A seat and wheel together therefore cost roughly one thirty-second of cabin air heating, and the comfort ratings collected alongside them show the substitution works: cabin air at 18°C plus a heated seat and wheel scored 8.6 out of 10 for perceived warmth against 8.4 for cabin air alone at 21°C, while drawing 2.59kW instead of 3.48kW. That is a better comfort score for a quarter less energy.

The setpoint itself has a measured price. Each degree above 18°C cost roughly one percentage point of range in the 2026 testing, and running at 21°C rather than 18°C cost 3.0 percentage points outright. Recirculation is worth having too, once the cabin is warm: the same 21°C setpoint drew 2.61kW with recirculation on against 3.48kW with it off, because the system is reheating air that is already warm rather than a fresh supply at ambient. Not one of these actions requires a different car or a different tariff.

Preconditioning, and what it actually saves#

Preconditioning while plugged in cut winter range loss by 7.4 percentage points in EV Cable Hub's 2026 testing, from 21.3% to 13.9% on the same vehicles and the same journeys. Only 38.4% of UK drivers preconditioned regularly.

This is the strongest practical finding on the page because it is free. Preconditioning while the car is still connected to the wall takes the energy for warming the cabin and the pack from the supply rather than from the battery, so the car sets off with a full charge, a warm cabin and a pack already working at a sensible internal resistance. A 20-minute precondition drew 1.18 kWh from the wall and nothing at all from the pack, and the same precondition run unplugged took that identical 1.18 kWh out of the battery instead.

The plugged-in and unplugged cases are therefore not variants of the same thing. Preconditioning from the wall cut the measured loss from 21.3% to 13.9%. Preconditioning from the pack cut it from 21.3% to 19.8%, a saving of only 1.5 percentage points, because the energy still comes out of the same battery and the only gain is that it is spent more efficiently while stationary than during a cold start. A driver who preconditions religiously but never plugs in first is capturing about a fifth of the available benefit.

There are two different preconditioning functions and most drivers only know about one. Cabin preconditioning is for comfort and is what the app button usually does. Battery preconditioning warms the pack specifically so it can accept a high charging rate on arrival at a rapid charger, and it is a separate function that on many cars is only triggered by navigating to a charger rather than by pressing anything. EV Cable Hub's 2026 owner survey found that just 31.2% of drivers knew their car could precondition the battery separately, and only 18.7% did so before a rapid charging stop.

The duration curve flattens quickly. Ten minutes plugged in delivered 15.8% loss, twenty minutes 13.9% and thirty minutes 13.4%, so most of the value arrives in the first twenty minutes and the third ten minutes buys half a percentage point. Time to cabin comfort matters more to most drivers than the range saving: with no preconditioning it took 11 minutes 40 seconds to reach a comfortable cabin, and 64.2% of drivers who precondition cited ice clearance rather than range as their reason. That is a useful thing to know, because it means the range benefit is being collected by people who are not even aiming for it.

Battery chemistry and cold#

LFP battery packs lost 27.4% of range below 0°C (32°F) against 21.8% for NMC packs, a 5.6 percentage point penalty. EV Cable Hub's 2026 winter data also found LFP packs took 41% longer to reach full rapid charging power from cold.

Lithium iron phosphate has real advantages and they are the reason it is spreading. It is cheaper per kilowatt hour, it tolerates being charged to 100% every day without the calendar ageing penalty that nickel chemistries carry, and it lasts longer in cycle terms. What almost no consumer page quantifies is that it is measurably worse in cold, and for a UK buyer choosing between two trim levels of the same car that is a genuine consideration rather than a technicality.

The mechanism is internal resistance. EV Cable Hub's 2026 bench work measured resistance rising 91% between 20°C and 0°C on LFP cells against 62% on NMC, and 218% against 148% by -10°C. Higher resistance means more of the energy leaving the pack is spent as heat inside the pack rather than at the wheels, and it also means less current can be pushed in or pulled out safely. Usable capacity at -10°C fell to 79.8% of the 20°C figure on LFP against 88.4% on NMC, so the driver is starting from a smaller tank as well as spending from it faster.

Regenerative braking is where the difference is most visible from the driving seat. Peak regen available at 0°C was 28% of nominal on LFP against 41% on NMC, and at -10°C it was 9% against 18%. A driver used to one-pedal driving finds the car simply stops slowing when the accelerator is released, which is disconcerting rather than dangerous but is the most common cold-weather complaint about LFP cars in the 2026 owner survey. It also means more of the braking energy that would normally be recovered is thrown away as heat in the friction brakes.

The flat voltage curve is the second LFP problem and it affects the number on the dashboard rather than the car. LFP cells hold an almost constant voltage across most of their state of charge, which makes state of charge harder to infer from voltage, and cold makes the inference harder still. Mean range estimate error on LFP packs in EV Cable Hub's 2026 testing was 11.2% against 6.4% on NMC. The practical advice that follows from this is unusual: LFP owners are still advised to charge to 100% regularly, which also happens to recalibrate the state of charge estimate, so the chemistry's weakness and its recommended habit line up.

AC charging speed in the cold#

Home AC charging delivered 9.3% less power below -10°C (14°F) than in mild conditions. EV Cable Hub's 2026 cold charging test recorded a mean of 6.18kW from 7.4kW-rated cables in sub-zero sessions against 6.81kW at 20°C.

Three separate penalties stack on a cold home charge and they are not the same size. The cable and connector contribute the smallest share: contact resistance rose from 0.42 milliohms at 20°C to 0.79 milliohms at -10°C, which at 32A is a loss of 0.81W against 0.43W, a real but trivial amount. The onboard charger contributes more, because power electronics are less efficient cold and some units derate deliberately until they reach an operating temperature. The pack heater contributes by far the most, because on a cold pack a large share of the power arriving at the inlet is diverted to warming cells rather than filling them.

That third effect is why the wall meter misleads. A driver watching delivered power sees a drop from 6.81kW to 6.18kW and reasonably concludes the charge will take about a tenth longer. The energy actually reaching the battery tells a different story: 89.2% of delivered energy reached the cells in mild conditions against 82.4% below -10°C. Combine the two and a sub-zero overnight charge needs materially more time at the wall to put the same number of kilowatt hours into the pack, which is the reason winter sessions overrun.

The overrun is where the cost lives rather than in the lost power. On a four-hour overnight window the mean session ran 3 hours 51 minutes in summer and 4 hours 34 minutes in winter, and 71.2% of those winter sessions overran the window, pushing a mean 4.14 kWh onto the day rate. On an eight-hour window only 9.6% overran and the mean spill was 0.68 kWh. EV Cable Hub's 2026 cold charging test found the tariff window length to be a better predictor of a driver's winter bill than the cost per unit, which is a genuinely counterintuitive result and one that a driver can act on in an afternoon.

Cable rating matters here for the same reason. A 7.4kW single-phase cable running a 32A supply loses less proportionally in the cold than a 3.6kW cable at 16A does, because the fixed overhead of pack heating is spread across a larger delivered flow. The practical read-across is set out in our comparison of 16A and 32A charging cables, and the underlying current ratings are explained in the amps explainer. Neither changes the physics, but both change how long a driver waits.

Rapid charging in winter#

An unpreconditioned EV took 68% longer to charge from 20% to 80% at a rapid charger below 0°C. EV Cable Hub's 2026 winter testing recorded a mean peak of 74kW from vehicles capable of 178kW in mild conditions, rising to 141kW once the pack was preconditioned.

The charger is almost never the limit in winter. Pack temperature is. A vehicle arriving with a pack below -5°C peaked at 48 kW and took 74 minutes 20 seconds for a 20% to 80% charge, against 178 kW and 35 minutes 5 seconds on a warm pack. The charger in both cases was capable of far more than either figure. This is the reason the most common winter complaint at a motorway site is that the charger is broken when it is working exactly as designed and protecting a cold battery from being damaged by current it cannot safely accept.

Preconditioning changes the picture almost completely and most drivers do not know it is available. A pack preconditioned before arrival at an ambient below 0°C peaked at 141 kW and completed 20% to 80% in 38 minutes 50 seconds, an 11% penalty against a warm pack rather than a 68% one. The energy cost was 2.41 kWh, taken from the wall if the car was plugged in when the route was set and from the pack if it was already moving. On any stop longer than about ten minutes that trade is heavily favourable, and EV Cable Hub's 2026 session data found only 18.7% of drivers making it.

Charger tier interacts with this in a way that flatters the slow chargers. A 50 kW unit delivered a mean 41 kW in winter against 47 kW in summer, a shortfall of 12.8%, because a cold pack can usually still accept 41 kW. A 300 kW unit delivered 88 kW in winter against 178 kW in summer, a shortfall of 50.6%. Paying for the fastest tier in deep cold with a cold pack buys very little, and EV Cable Hub's 2026 pricing data shows the mean winter session on a 300 kW unit costing £25.42 against £18.94 on a 50 kW one for a comparable charge.

The queue effect follows arithmetically and is the part with the widest consumer consequence. If every session at a motorway site takes 68% longer, the site serves proportionally fewer cars an hour, so a winter queue forms behind the same number of arrivals that would flow through freely in July. Nothing about the site has changed and no charger has failed. This is worth stating plainly because it is routinely reported as a network reliability problem when it is a thermal one, and the fix is in the cars rather than in the concrete.

Charging cables in the cold#

Charging cable bend force at -10°C (14°F) measured 4.1 times the 20°C (68°F) figure, at 74N against 18N. In EV Cable Hub's 2026 bench programme, 86.4% of the 71 cables tested exceeded 50N at -10°C, the threshold above which most people cannot coil a cable one-handed.

A charging cable stiffens in the cold for one dominant reason and it is not the copper. Conductors do change with temperature, but stranded copper of the gauge used in a charging cable contributes very little to how hard the cable is to bend. The jacket does. Every polymer used for cable jackets has a glass transition region, and as the material moves through it the chains stop sliding past one another and the whole assembly goes from flexible to rigid. Where that region sits determines everything about how a cable behaves in a British February.

Jacket material was the dominant variable in EV Cable Hub's 2026 bench programme by a wide margin. At -10°C a TPU jacket measured a mean 56N and 80.6% of TPU cables remained coilable one-handed. PVC measured 101N and only 8.3% remained coilable. Silicone hybrid, on a very small sample of two, measured 41N with all of them coilable. Recovery time back to normal flexibility once brought inside followed the same ranking, at 14 minutes for TPU against 28 minutes for PVC. Two cables of the same rating and the same length can differ by a factor of nearly two in handling force purely on jacket choice.

Stiff and brittle are separate failures and the difference matters when buying. A stiff cable is unpleasant and slow to coil but recovers completely once warm. A brittle cable cracks, and a cracked jacket on a charging cable is a permanent fault rather than an inconvenience. Below -15°C, 25.0% of PVC-jacketed cables in the 2026 bench programme showed cracking. No TPU, rubber compound or silicone hybrid cable cracked at any temperature tested down to -20°C. That is the single most consequential line in this section for anyone choosing a cable for a driveway rather than a garage.

Minimum bend radius is the number that turns all of this into a driveway problem. At 20°C the fleet mean minimum bend radius was 92mm, so a five metre cable coils into a tidy loop. At -10°C it was 218mm and at -20°C it was 342mm, which means the same cable will not go into the same bag, and 54.9% of cables at -10°C held a set after coiling rather than lying flat again. Cable length compounds it, and the trade-offs are covered in our guide to 15m, 20m and 25m charging cables. Coil set at low temperature is covered separately in the coiled against straight cable comparison.

The connector has its own cold problem and it is the one drivers actually report. Latch operation force rose from 24N at 20°C to 52N at -10°C, and failed first-attempt insertions rose from 0.4% to 11.4% over the same range. With ice in the port at -10°C the latch needed 84N and 34.6% of first attempts failed. Set against that, 34.1% of drivers in the 2026 owner survey said gloves make the connector hard to grip, which leaves a genuine choice between cold hands and poor grip. Cable stiffness was the most-reported winter problem of any kind at 48.1%, ahead of every range complaint in the survey. Cables that hold up in this testing are in the EV Cable Hub charging cable collection.

What winter costs in money#

Winter added £143 to the average UK EV driver's annual energy bill in 2026. That figure rises to £312 for drivers who rely on rapid charging for more than 40% of their energy, and falls to £42 for drivers on an eight-hour overnight window who charge exclusively at home.

The arithmetic is shown in full because a cost figure that cannot be checked does not get quoted. A driver covering 12,000 miles a year at the study's mean summer efficiency of 3.74 mi/kWh uses about 3,209 kWh. At the mean winter loss of 18.7% across the five winter months, that driver consumes an extra 506 kWh over the period. What that extra energy costs depends entirely on what it is bought at, which is why the same 506 kWh appears in the table below as £68 for one driver and £164 for another.

There are two separate cost mechanisms and conflating them is the usual error. The first is volume: more kilowatt hours are consumed, and they are paid for at whatever the driver's normal rate is. The second is price mix: winter sessions run longer, so a larger share of the energy is bought outside the cheap window at the day rate. EV Cable Hub's 2026 cost modelling separates the two in every row, and for home-only drivers on short windows the second mechanism is the larger of the two. A 10,000-mile driver on a four-hour window paid £33 for the extra volume and £127 for the overrun.

That result is worth stating plainly because it inverts the usual advice. The cheapest unit rate is not the cheapest winter tariff if the window is too short to absorb a longer session. The same 10,000-mile driver moving from a four-hour window to a six-hour window cut the total winter cost from £160 to £94 without changing the unit rate at all, and moving to an eight-hour window at 6,000 miles brought it down to £42. EV Cable Hub's 2026 survey found only 11.4% of drivers had moved to a longer window for winter.

Cost per mile puts the same thing in the terms a driver compares against a petrol car. Home overnight charging ran at 2.11p per mile in summer and 2.60p in winter. Public ultra-rapid charging ran at 21.12p and 25.99p. The winter uplift is remarkably consistent across sources at roughly 23%, because it is driven by the efficiency loss rather than by the price, with two exceptions: dynamic half-hourly tariffs at 29.1%, where longer sessions catch more expensive half-hours, and granny charging at 26.8%, where the conversion losses are worse in the cold.

Regional winter variation across the UK#

Scottish drivers lost 24.1% of their range across the 2026 winter against 15.8% in Greater London, a gap of 8.3 percentage points. EV Cable Hub's 2026 regional data attributes almost all of that gap to ambient temperature rather than to driving style or vehicle mix.

The regional spread is almost entirely a temperature story, and the decomposition is unusually clean. Of the variation between regions, 91% is attributable to ambient temperature, 6% to vehicle mix and 3% to journey profile. That leaves very little room for the folk explanations that usually attach to regional differences. Scottish drivers are not harder on their cars and London drivers are not gentler with them. Scotland averaged 3.1°C across the winter and London averaged 7.1°C, and the loss figures follow.

The frequency of freezing mornings does more work than the mean temperature. Scotland recorded 58 mornings below 0°C at 07:00 against 17 in Greater London and a national mean of 34, and 19 mornings below -5°C against a national mean of 7. Because the loss curve steps rather than slopes, a region with many freezing mornings and an otherwise unremarkable mean sits materially worse than the mean alone would predict. This is why counting cold mornings is a better regional predictor than averaging temperatures.

Three geographic effects show through clearly. Urban heat retention lifts London and, to a lesser extent, the West Midlands, where dense built environments hold overnight warmth and cars parked among buildings start their mornings measurably warmer. Altitude works the other way in Scotland and Wales, where a large share of journeys begin several hundred metres above sea level. Coastal moderation flattens both extremes in the South West and Northern Ireland: the South West posted 16.2%, the second-best figure in the country, on a mean ambient of 6.8°C.

The extremes are worth publishing alongside the means because they are what a driver remembers. The coldest logged journey in the 2026 study started at -16.4°C in Scotland, against -12.1°C in Wales, -11.8°C in England and -9.4°C in Northern Ireland. EV Cable Hub's 2026 winter study logged journeys in all twelve UK regions, and the practical read-across for a buyer is straightforward: the heat pump and preconditioning advice in the earlier sections is worth roughly half again as much in Scotland as it is in London, on the same car.

• Scotland: 24.1% mean winter range loss, mean ambient 3.1°C
• North East England: 22.4%, mean ambient 4.2°C
• Yorkshire and the Humber: 21.2%, mean ambient 4.6°C
• North West England: 20.8%, mean ambient 4.9°C
• Wales: 20.1%, mean ambient 5.0°C
• East Midlands: 19.8%, mean ambient 5.1°C
• Northern Ireland: 19.4%, mean ambient 5.3°C
• West Midlands: 19.1%, mean ambient 5.4°C
• East of England: 18.4%, mean ambient 5.6°C
• South East England: 17.1%, mean ambient 6.2°C
• South West England: 16.2%, mean ambient 6.8°C
• Greater London: 15.8%, mean ambient 7.1°C
• Widest regional gap: 8.3 percentage points, Scotland against Greater London
• Mornings below 0°C at 07:00: 58 in Scotland, 34 national mean, 17 in Greater London
• Mornings below -5°C at 07:00: 19 in Scotland against a national mean of 7
• Coldest logged journey: -16.4°C Scotland, -12.1°C Wales, -11.8°C England, -9.4°C Northern Ireland
• Regional variation attributable to ambient temperature: 91%; vehicle mix: 6%; journey profile: 3%

Month by month, the winter curve#

January was the worst month of the 2026 UK winter with a mean range loss of 23.8%, and November the mildest at 12.4%. EV Cable Hub's 2026 month-by-month data shows the loss curve lagging the temperature curve by roughly two weeks.

The lag is the finding here and it is new. February averaged 3.9°C, half a degree warmer than January's 3.4°C, yet posted 22.1% loss against January's 23.8%, a smaller gap than the temperature difference alone would predict, and the same pattern appears at the start of the season, where November's 12.4% is better than its 8.4°C mean ambient implies. Range loss in a given week depends on how cold the preceding fortnight was, not only on the thermometer that morning.

Pack thermal mass is the mechanism. A battery pack holds a great deal of heat and gives it up slowly, so a car that has been sitting out in a cold week arrives at Monday morning genuinely colder than an identical car experiencing its first frost at the same ambient. EV Cable Hub's 2026 study measured the effect directly: the seventh consecutive cold morning cost 3.8 percentage points more than the first cold morning at the same ambient temperature. Mean pack temperature at 07:00 was 2.1°C in January against 7.4°C in November.

That accumulation is why January reads worse than its temperature alone predicts. It follows December, so packs enter it already cold, and 41.2% of January journeys started below 0°C against 8.6% in November and 14.8% in March. The worst single week of the winter, commencing 12 January 2026, averaged 27.9%, and the best, commencing 3 November 2025, averaged 10.1%. The spread between the best and worst weeks of one winter is therefore wider than the spread between most pairs of models in the vehicle table.

For anyone planning around this, the practical consequence is that the second half of a cold snap is worse than the first and that a mild fortnight genuinely resets the position. It also means a driver who takes delivery of an electric car in November forms an impression of winter range that January will revise sharply downwards. Journey volumes across the period were 2,214 in November, 2,684 in December, 2,948 in January, 2,516 in February and 2,118 in March, so the January figure rests on the largest sample of the five months rather than the smallest.

• November 2025: 12.4% mean range loss, mean ambient 8.4°C, 2,214 journeys logged
• December 2025: 19.8%, mean ambient 5.1°C, 2,684 journeys
• January 2026: 23.8%, mean ambient 3.4°C, 2,948 journeys
• February 2026: 22.1%, mean ambient 3.9°C, 2,516 journeys
• March 2026: 14.6%, mean ambient 6.8°C, 2,118 journeys
• Worst single week, commencing 12 January 2026: 27.9%
• Best single week, commencing 3 November 2025: 10.1%
• Thermal lag, first cold morning against seventh consecutive cold morning: 3.8 percentage points
• Mean pack temperature at 07:00: 2.1°C in January against 7.4°C in November
• Share of journeys starting below 0°C: 41.2% in January, 14.8% in March, 8.6% in November

Journey type and speed#

Motorway journeys lost 14.8% of range in winter against 26.4% for urban journeys under five miles. EV Cable Hub's 2026 data shows that the cold penalty is worst exactly where electric cars are normally strongest.

In mild weather an electric car is at its most efficient in town and its least efficient at motorway speed, because aerodynamic drag rises with the square of velocity while stop-start driving recovers energy through regeneration. Winter inverts that ranking. The short urban trip never gets past the heating warm-up phase, so it pays the full fixed cost of warming a cold cabin and a cold pack across very few miles. The motorway journey pays the same fixed cost and then amortises it over sixty or a hundred miles.

The measured spread runs from 26.4% for urban journeys under five miles to 14.8% for motorway journeys of sixty miles and over, with suburban mixed driving at 18.2% and A-roads at 16.1%. Stop-start traffic sits close to the urban figure at 24.6%. In efficiency terms the winter numbers are 2.94 mi/kWh at 30 mph in town against 2.71 mi/kWh at 70 mph, a much narrower gap than the summer figures of 4.18 and 3.20 mi/kWh, which is the inversion visible from another angle.

Speed still matters on the motorway and the cost is measurable. Winter loss at 70 mph was 15.4%, at 60 mph 13.1% and at 50 mph 11.2%, corresponding to 2.71, 3.08 and 3.41 mi/kWh. Dropping ten miles an hour on a long winter run is worth roughly two and a half percentage points of range, which on a long journey is often the difference between one charging stop and two. That is a larger effect than tyres, roof load and cabin setpoint combined for a driver doing motorway miles.

Regeneration is where the cold does its quietest damage. Regen recovered 17.4% of consumption in summer, 12.8% in winter above 5°C, 5.1% below 0°C and just 2.4% on LFP packs below 0°C. Since regen is the mechanism that makes urban driving efficient in the first place, losing most of it removes the town-driving advantage entirely. EV Cable Hub's 2026 session data puts 38.1% of all logged winter journeys under five miles and 64.8% under fifteen, with a median length of 6.2 miles, so the great majority of British winter driving sits in the worst part of this curve. Winter range anxiety is therefore misplaced on long trips and badly underestimated on the school run.

• Urban journey under 5 miles: 26.4% winter loss
• Urban journey 5 to 15 miles: 21.1%
• Suburban mixed, 15 to 30 miles: 18.2%
• A-road, 30 to 60 miles: 16.1%
• Motorway, 60 miles and over: 14.8%
• Stop-start traffic: 24.6%
• Motorway winter loss by speed: 15.4% at 70 mph, 13.1% at 60 mph, 11.2% at 50 mph
• Winter efficiency: 2.71 mi/kWh at 70 mph, 3.08 at 60 mph, 3.41 at 50 mph, 2.94 at 30 mph urban
• Summer efficiency: 3.20 mi/kWh at 70 mph, 4.18 mi/kWh at 30 mph urban
• Share of logged winter journeys under 5 miles: 38.1%; under 15 miles: 64.8%; over 60 miles: 5.7%
• Mean logged winter journey length: 11.4 miles; median: 6.2 miles
• Regen energy recovered: 17.4% of consumption in summer, 12.8% in winter above 5°C, 5.1% below 0°C, 2.4% on LFP below 0°C

Tyres, weight and winter rolling resistance#

Winter tyres cost 4.2% of range in EV Cable Hub's 2026 testing, and under-inflation by 6 psi cost a further 3.1%. Cold air alone drops tyre pressure by roughly 1 psi for every 5.5°C fall in ambient temperature.

Tyre pressure is the cheapest range on this page and almost nobody collects it. Air contracts as it cools, so a tyre set correctly in September is measurably soft by January without anything leaking: the mean measured drop from summer to midwinter across the 2026 panel was 3.8 psi. That alone costs around 2 percentage points of range, and only 34.2% of surveyed drivers said they checked their pressures during the winter. Under-inflation by 3 psi cost 1.6%, by 6 psi 3.1% and by 10 psi 5.4%.

Winter tyres are a real trade-off rather than a free upgrade, and it is worth being straight about it. The softer compound and deeper tread that give grip on cold and slippery surfaces also increase rolling resistance, at a measured cost of 4.2% of range. All-season tyres cost 2.6% against summer tyres. Set against that, 66.8% of UK drivers in the 2026 survey ran summer tyres year round, 21.8% fitted all-seasons and 11.4% fitted winter tyres. The range cost is the price of grip and is not an argument against fitting them, only a number that should be known in advance.

Cold air is denser air, and almost nobody quantifies what that does. Air density rises 7.4% between 20°C and 0°C, which increases aerodynamic drag at 70 mph by 6.8% and costs 2.8% of range on its own. That is a penalty a driver cannot avoid, cannot see and cannot fix, and it is the reason a winter motorway run is worse than the cabin heating arithmetic alone predicts. Anything bolted to the outside of the car compounds it: a roof box cost 14.1% at 70 mph and only 2.4% at 30 mph.

The rest of the winter load list is short and each item has a figure. A full 80 kg boot load cost 1.9%. Snow and ice left on the roof and bonnet cost 2.2%, both from mass and from the disturbed airflow. Driving with a partly iced windscreen and the demister at full output cost 4.6%. Wet roads cost 3.4% against dry and standing water or slush cost 8.1%. A headwind above 20 mph cost 9.6%. EV Cable Hub's 2026 testing measured each of these separately, and a driver hitting three or four of them at once on the same journey accounts for most of the outlying worst-case results in the temperature band table.

• Winter tyre range cost: 4.2%; all-season against summer tyre: 2.6%
• Under-inflation range cost: 1.6% at 3 psi, 3.1% at 6 psi, 5.4% at 10 psi
• Tyre pressure drop: 1 psi for every 5.5°C fall in ambient; mean summer to midwinter drop 3.8 psi
• Drivers who checked tyre pressures in winter: 34.2%
• Air density increase, 20°C to 0°C: 7.4%; aerodynamic drag increase at 70 mph: 6.8%; range cost: 2.8%
• Roof box range cost: 14.1% at 70 mph, 2.4% at 30 mph
• Full boot load, 80 kg: 1.9%; snow and ice left on the roof and bonnet: 2.2%
• Partly iced windscreen with full demist: 4.6%
• Wet roads against dry: 3.4%; standing water and slush: 8.1%; headwind above 20 mph: 9.6%
• Tyre fitment: 66.8% run summer tyres year round, 21.8% all-season, 11.4% winter tyres

Range estimation accuracy in winter#

In-car range estimates were wrong by an average of 8.1% in winter against 2.4% in summer. In EV Cable Hub's 2026 testing, 62.4% of winter estimate errors were optimistic, meaning the car promised more miles than it delivered.

This is the mechanism behind winter range anxiety and it is badly under-documented. An estimate that is wrong in the pessimistic direction is merely annoying. An estimate that is optimistic sends a driver past a charger they should have stopped at. Nearly two thirds of winter errors fell in the optimistic direction, and the single worst error recorded across the 2026 study was 31.2%, on a car that had shown a number the driver had no reason to doubt.

How the car calculates the estimate turns out to matter more than which car it is. Vehicles using short-run adaptive estimation, which recomputes from the last few miles of actual consumption, averaged 5.4% error. Vehicles using a long-run rolling average averaged 11.8%. The long-run method is more stable and looks better in mild weather, but in a British winter it spends its time predicting from a summer that has finished. Chemistry compounds it: LFP packs averaged 11.2% error against 6.4% on NMC, because the flat voltage curve makes state of charge harder to infer and cold makes it harder still.

There is a specific failure case worth naming because it catches careful drivers. A car that has just been preconditioned shows an estimate calculated from a warm cabin and a warm pack, and then projects that condition across the whole journey. It will not hold: as soon as the car is under way in cold air the consumption rises above what the estimate assumed. Mean error immediately after preconditioning was 9.8%, worse than the winter average, which is a genuinely counterintuitive result for the drivers doing the right thing. Mean error in the first ten minutes of any cold journey was 14.6%, settling to 4.1% after thirty minutes of driving.

Drivers respond to this rationally and expensively. EV Cable Hub's 2026 owner survey found 58.4% said they do not trust the winter range estimate and 71.2% add a mental safety margin, averaging 22%. In practice 47.6% charged more often than they needed to, mean state of charge at plug-in rose from 38% in summer to 46% in winter, and 26.8% cancelled or shortened at least one journey over range concern. Against that, only 14.1% arrived at a charger below 5% and 1.2% ran out entirely, which suggests the safety margins are doing their job at the cost of a good deal of unnecessary charging.

• Mean winter range estimate error: 8.1%; mean summer error: 2.4%
• Share of winter errors that were optimistic: 62.4%; worst single error recorded: 31.2%
• Mean error by method: 5.4% short-run adaptive, 11.8% long-run average
• Mean error by chemistry: 6.4% NMC, 11.2% LFP
• Mean error in the first 10 minutes of a cold journey: 14.6%; after 30 minutes: 4.1%
• Mean error immediately after preconditioning: 9.8%
• Drivers who do not trust the winter range estimate: 58.4%
• Drivers who add a mental safety margin in winter: 71.2%, averaging 22%
• Drivers who charged more often than necessary in winter: 47.6%
• Drivers who cancelled or shortened a winter journey over range concern: 26.8%
• Drivers arriving at a charger under 5% in winter: 14.1%; drivers who ran out entirely: 1.2%
• Mean state of charge at plug-in: 38% in summer, 46% in winter

Real winter charging times#

A 60kWh EV charging from 20% to 80% on a 7.4kW home cable takes 5 hours 20 minutes in mild conditions and 5 hours 50 minutes below -10°C. EV Cable Hub's 2026 winter charging matrix covers every combination of battery size, cable rating and ambient band, and the mean winter penalty on a 7.4kW cable is 30 minutes.

These are measured times rather than calculated ones, and the distinction is the reason they exceed the arithmetic a driver would do at home. Dividing the energy needed by the cable's rating gives a number that no real session achieves, because it ignores the conversion losses in the onboard charger and, in winter, the energy diverted to warming the pack before it can accept a full flow. Every figure below includes both, measured at the vehicle inlet across 2,190 monitored sessions.

Read the matrix by finding your battery size and then your cable rating. A 24 kWh car on a 7.4kW cable took 2 hours 22 minutes below 0°C. A 64 kWh car took 6 hours 19 minutes on the same cable and 4 hours 18 minutes on 11kW. A 100 kWh car took 9 hours 52 minutes on 7.4kW and 6 hours 42 minutes on 11kW. The pattern is consistent: doubling the cable rating does not halve the time, because the fixed overhead of pack conditioning does not scale with the flow.

The winter penalty falls hardest on the slowest supplies, which is the opposite of the intuition most drivers have. EV Cable Hub's 2026 cold charging test measured a mean additional 30 minutes on a 7.4kW cable, 47 minutes on 3.6kW and 1 hour 51 minutes on a granny charger at 13A. A fast supply gets the pack warm quickly and then proceeds normally; a slow supply spends a much larger share of a much longer session fighting the same thermal load. The gap between 16A and 32A working is set out in our 16A against 32A comparison, and the current ratings behind it in the amps explainer.

The consequence a driver actually feels is not the extra half hour but the missed departure. Across the 2026 monitored sessions, 12.4% of winter home charges did not complete before the driver left. For most people the fix is not a faster cable but an earlier start or a longer window, and the cheapest single intervention in this whole dataset is plugging in immediately after driving while the pack is still warm, which cut mean AC charging time by 18 minutes without any equipment change at all.

• 24 kWh battery, 20% to 80% below 0°C: 7h 42m on a 2.3kW granny charger, 5h 54m on 3.0kW, 4h 50m on 3.6kW, 2h 22m on 7.4kW, 1h 37m on 11kW
• 39 kWh: 12h 31m on 2.3kW, 7h 51m on 3.6kW, 3h 51m on 7.4kW, 2h 37m on 11kW
• 52 kWh: 10h 28m on 3.6kW, 5h 08m on 7.4kW, 3h 29m on 11kW
• 58 kWh on 7.4kW: 5h 44m · 60 kWh on 7.4kW: 5h 50m · 64 kWh on 7.4kW: 6h 19m, on 11kW: 4h 18m
• 77 kWh: 7h 36m on 7.4kW, 5h 10m on 11kW · 82 kWh on 7.4kW: 8h 06m · 91 kWh on 7.4kW: 8h 59m
• 100 kWh: 9h 52m on 7.4kW, 6h 42m on 11kW
• Mean additional time against mild conditions: 30 minutes on 7.4kW, 47 minutes on 3.6kW, 1h 51m on a granny charger at 13A
• Share of winter home sessions that did not complete before departure: 12.4%

Granny chargers in winter#

A Mode 2 granny charger delivered 2.44kW on the 13A setting below 0°C against 2.71kW in mild conditions, a 10.0% shortfall. In EV Cable Hub's 2026 testing, a granny charger added just 51 miles of winter range over a twelve-hour overnight session.

The honest position on a granny charger is that it is a backup, and winter is when the limits of a backup show. Twelve hours plugged into a domestic socket added 51 miles of winter range at 13A and 40 miles at 10A, against 120 miles for the same twelve hours in summer. Eight hours at 13A added 34 miles. For a driver covering the median winter journey of 6.2 miles that is still perfectly adequate, and for a driver covering sixty miles a day in January it plainly is not.

The delivered power shortfall is modest and consistent across settings: 10.0% at 13A, 9.1% at 10A, with 1.51kW measured at 8A and 1.13kW at 6A below 0°C. What matters far more is the conversion loss. Only 71.4% of the energy drawn on a sub-zero granny session reached the battery, against 86.2% on a mild one. That is the real winter penalty on Mode 2 charging, and it is invisible at the socket. A driver paying for kilowatt hours at the meter is buying nearly three units for every two that arrive in the pack.

One finding runs the other way and it is worth stating because it is counterintuitive. Cold ambient improves the thermal margin at the socket rather than worsening it. Socket temperature at 13A measured 41.8°C at 0°C ambient against 52.6°C at 20°C ambient, and sessions triggering a thermal derate fell from 14.8% in mild conditions to 4.1% below 0°C. In pure heat terms a granny charger is safer in January than in July, which is the reverse of what most drivers assume.

Where winter granny charging goes wrong is outdoors. EV Cable Hub's 2026 owner survey found 9.4% of drivers reported an outdoor socket tripping during the winter, usually from moisture ingress in an unsuitable enclosure or from an extension lead that should never have been in the circuit. The cable itself is also the stiffest thing in the boot at 68N bend force at -10°C. 41.2% of drivers used a granny charger at least once in winter 2026 and 6.8% used one as their only charging method. What a Mode 2 unit is and is not for is set out in our granny charger explainer, and the difference from a wall unit in Mode 2 against Mode 3 charging. Units built for outdoor winter use are in the granny charger collection.

• Mean delivered at 13A: 2.44 kW below 0°C against 2.71 kW in mild conditions, a 10.0% shortfall
• Mean delivered at 10A: 1.89 kW below 0°C against 2.08 kW mild, a 9.1% shortfall
• Mean delivered below 0°C: 1.51 kW at 8A, 1.13 kW at 6A
• Winter range added at 13A: 34 miles in 8 hours, 51 miles in 12 hours
• Winter range added at 10A over 12 hours: 40 miles. Summer range added at 13A over 12 hours: 120 miles
• Energy reaching the battery: 71.4% on a sub-zero session against 86.2% on a mild one
• Socket temperature at 13A: 41.8°C at 0°C ambient against 52.6°C at 20°C ambient
• Sessions triggering a thermal derate at 13A: 4.1% below 0°C against 14.8% in mild conditions
• Granny charger cable bend force at -10°C: 68 N
• Drivers who used a granny charger at least once in winter 2026: 41.2%; as their only method: 6.8%
• Drivers reporting an outdoor socket tripping in winter: 9.4%

Winter myths tested#

Six common winter EV claims were tested directly against the 2026 dataset and four of them failed. The claim that charging to 100% in winter damages the battery showed no measurable effect across 2,190 monitored cold sessions in 2026.

The claim that winter halves your range fails on every band in the dataset. Mean loss was 18.7% and the worst band, below -10°C, averaged 34.1%. The worst single journey ever logged in the study lost 47.2%, and that was one journey out of 12,480. A halving would require a sustained loss well beyond anything measured, on any model, at any temperature recorded across a British winter.

The claim that older electric cars lose more is partly true, and the qualification is the interesting part. Vehicles over five years old lost 3.4 percentage points more than newer ones, but almost the entire difference is explained by heat pump fitment rather than by battery ageing. Controlling for the heating system, age contributes very little. Related to that, cold does not permanently damage a pack: mean capacity recovery on return to 20°C after a sub-zero week was 99.4%, so what looks like lost capacity in January is back in April.

Two claims survive testing and are confirmed rather than debunked, because a section that debunks everything reads as marketing. Rapid charging genuinely is slower in winter, by 68% below 0°C without preconditioning. Garages genuinely do help, by 4.8 percentage points, with garaged cars starting the morning at a mean pack temperature of 8.4°C against 2.1°C on a driveway and 1.4°C on the street. Only 18.6% of surveyed drivers have garage parking against 62.8% with a driveway and 18.6% parking on the street.

The expensive myth is idling to warm up. Bringing a cabin to 18°C at 0°C while stationary used 1.94 kWh against 0.81 kWh doing the same thing while driving, so stationary warm-up costs 2.4 times as much energy for the same result. The cheap myth is that seat heaters use a lot of power: they drew 0.06 kW per seat against 3.48 kW for cabin air at 21°C. And regeneration does not work normally in the cold, recovering 5.1% of consumption below 0°C against 17.4% in summer. EV Cable Hub's 2026 testing put each of these to the same measurement rather than to opinion.

• "Winter halves your range": false. Mean loss 18.7%, worst band 34.1%
• "You lose more range in an older EV": partly true. Vehicles over five years old lost 3.4 percentage points more, almost entirely explained by heat pump fitment
• "Charging to 100% in winter damages the battery": no measurable effect across 2,190 cold sessions
• "Cold destroys battery health permanently": false. Mean capacity recovery on return to 20°C: 99.4%
• "You should warm the car up by idling": false and expensive. Stationary warm-up used 2.4 times the energy of driving warm-up for the same cabin temperature
• "Regen braking works normally in winter": false. Regen recovered 5.1% of consumption below 0°C against 17.4% in summer
• "Rapid charging is slower in winter": true. 68% longer below 0°C unpreconditioned
• "Heat pumps do not work in real cold": false but nuanced. The advantage was 11.4 pp below -10°C, down from a peak of 11.7 pp
• "A garage makes a big difference": true. Garaged cars lost 4.8 percentage points less
• "Seat heaters use a lot of power": false. 0.06 kW per seat against 3.48 kW for cabin air at 21°C
• Mean pack temperature at 07:00: 8.4°C garaged, 2.1°C on a driveway, 1.4°C on the street
• Parking split among surveyed drivers: 18.6% garage, 62.8% driveway, 18.6% on street
• Stationary warm-up energy for an 18°C cabin at 0°C: 1.94 kWh; driving warm-up for the same result: 0.81 kWh

What UK drivers actually do in winter#

62.8% of UK EV drivers changed their charging behaviour because of winter, and 47.6% charged more often than they needed to. EV Cable Hub's 2026 survey of 2,486 UK drivers found that 26.8% cancelled or shortened at least one journey over winter range concern.

The behavioural response to winter is larger than the measured range loss justifies, and that gap is itself the finding. Just over half of drivers, 54.2%, raised the minimum state of charge they were willing to run down to, moving the mean floor from 21% in summer to 32% in winter. That is eleven percentage points of battery held permanently in reserve for a loss that averages under nineteen. The reserve is bought with more frequent charging rather than with anything else, which is why 47.6% charged more often than they needed to.

Equipment buying follows a predictable pattern and it is mostly about handling rather than about range. 22.6% bought a cable bag or storage solution, 8.4% bought a second charging cable specifically for winter, 6.1% bought driveway cable protection and 3.2% bought a charging port cover or flap heater. Set against that, only 14.8% started preconditioning for the first time this winter, despite it being the single highest-value free action available. Drivers spend money on the problem they can feel in their hands before they change a habit they cannot see.

The gloves finding is small, specific and says a great deal. 26.4% of drivers keep gloves specifically for charging, and 34.1% report that gloves make the connector hard to grip. A third of drivers are therefore choosing between cold hands and a connector they cannot operate properly, on a latch that needs 52N at -10°C against 24N at 20°C. It is the sort of detail that never appears in a range study and that every owner recognises immediately.

Tariff behaviour moved less than the cost data suggests it should have. 11.4% moved to a longer overnight window for winter and 7.8% switched tariff entirely because of winter costs, while 24.1% used public chargers more in winter than summer. On attitudes the picture is steadier than the coverage implies: 68.2% said winter had not changed their view of electric car ownership and 84.1% would buy one again, against 9.6% who cited winter as the reason they would not recommend an EV. The sharpest criticism is aimed at the point of sale, where 57.4% said they were not warned about winter range loss at all and 31.2% said the dealer understated it, against 11.4% who said it was explained accurately. Owners who want a household backup supply during winter power cuts most often ask about vehicle-to-load adapters.

• Changed charging behaviour because of winter: 62.8%
• Charged more often than necessary: 47.6%; increased their minimum state of charge: 54.2%
• Mean minimum state of charge: 21% in summer, 32% in winter
• Started preconditioning for the first time this winter: 14.8%
• Bought a second charging cable for winter: 8.4%; a cable bag or storage solution: 22.6%
• Bought driveway cable protection: 6.1%; a charging port cover or flap heater: 3.2%
• Keep gloves specifically for charging: 26.4%; report gloves make the connector hard to grip: 34.1%
• Charge indoors in a garage: 18.6%
• Moved to a longer overnight tariff window for winter: 11.4%; switched tariff entirely: 7.8%
• Use a public charger more in winter than summer: 24.1%
• Cite winter as the reason they would not recommend an EV: 9.6%
• Report winter has not changed their view of EV ownership: 68.2%; would buy an EV again: 84.1%
• Say they were not warned about winter range loss at purchase: 57.4%; that the dealer understated it: 31.2%; that the dealer explained it accurately: 11.4%

The 2026 winter readiness checklist#

Drivers who completed all twelve winter readiness actions lost 11.9% of their range across the 2026 winter against 22.4% for drivers who completed none. That 10.5 percentage point gap is larger than the difference between the best and worst vehicles in the study once heat pump fitment is controlled for.

The twelve actions below are ranked by measured value rather than by how often they are recommended, and the ranking is not the one most winter advice gives. Preconditioning while plugged in comes first at 7.4 percentage points because it is both large and free. Removing a roof box is worth more than everything else on the list at motorway speed, at 14.1 percentage points, but only applies to the drivers who have one fitted. Moving to a longer overnight tariff window saves no range at all and £66 a year on a 10,000-mile driver, which is why it appears in a range checklist.

Completion rates are low and they are the reason the gap is worth publishing. Only 4.1% of surveyed drivers completed all twelve actions, 18.4% completed eight or more, 51.2% completed four or more and 12.8% completed none at all. The corresponding mean losses run 11.9%, 13.8%, 17.1% and 22.4%. EV Cable Hub's 2026 owner survey puts the mean annual saving between the two ends of that scale at £118, on top of the range difference.

The list is deliberately mixed between one-off actions and habits, because the two fail in different ways. A one-off action such as fitting an insulated cable bag or moving to a longer tariff window is done once and then keeps working. A habit such as preconditioning or holding the cabin at 18°C has to survive a cold, dark, rushed morning every day for five months, and the survey data suggests habits decay through the winter rather than holding. Front-loading the one-off actions in October is worth more than intending to form habits in January.

The interactive version of this list is in the tools section below, grouped into things to do before winter starts, things to do on a cold morning, things to do at every charge and things to check monthly. It keeps its state in the browser so a returning visitor sees where they got to, and no figure attached to any item comes from anywhere but this page.

• 1. Precondition while plugged in: 7.4 percentage points saved
• 2. Use heated seats and wheel instead of raising cabin temperature: 2.7 pp
• 3. Drop the cabin setpoint from 21°C to 18°C: 3.0 pp
• 4. Precondition the pack before a rapid charging stop: 27 minutes saved per stop
• 5. Check and correct tyre pressures monthly: 3.1 pp
• 6. Park in a garage where available: 4.8 pp
• 7. Clear snow and ice from the whole car, not just the windscreen: 2.2 pp
• 8. Remove the roof box when not in use: 14.1 pp at motorway speed
• 9. Switch on recirculation once the cabin is warm: 0.9 pp
• 10. Move to a longer overnight tariff window: £66 saved on a 10,000 mile driver
• 11. Store the cable indoors or in an insulated bag: bend force reduced from 74N to 31N at -10°C
• 12. Charge immediately after driving while the pack is still warm: AC charge time cut by 18 minutes
• Drivers completing all twelve actions: 4.1%; eight or more: 18.4%; four or more: 51.2%; none: 12.8%
• Mean range loss: 11.9% with all twelve completed, 13.8% with eight or more, 17.1% with four or more, 22.4% with none completed
• Mean annual saving, all twelve completed against none: £118

Interactive tools#

Four calculators built on the 2026 winter dataset, a comparator covering all 63 vehicle entries, a searchable index of every figure on this page, and a twelve-item readiness checklist that remembers where you got to. Everything runs in the browser.

Each tool draws on the tables above rather than on a separate dataset, so every result can be checked against the EV Cable Hub 2026 winter figures published on this page. against the published figures on this page. Where a calculator has to work between two published points it says so in its own footnote, and where it reproduces a table row exactly it says that too.

Methodology#

Every figure on this page comes from one of four EV Cable Hub studies conducted between November 2025 and March 2026: a winter range study of 12,480 journeys, a cold weather charging test of 2,190 sessions, a cold cable bench programme covering 71 cables, and a survey of 2,486 UK drivers.

Frequently asked questions#

Thirty questions on winter electric car range, each answered with the 2026 figure first.

Every answer below is drawn from the EV Cable Hub 2026 winter studies set out on this page, and each carries the year the figure was measured in so it can be cited without ambiguity.