EV Charging Cable Speed Test 2026: 1,284 Sessions, 63 Cables, and What They Actually Deliver

The 2026 headline findings#

EV charging cables deliver an average of 8.6% less power than their stated rating. EV Cable Hub monitored 1,284 UK home charging sessions in 2026 and recorded a mean of 6.76kW from cables rated at 7.4kW, with the gap reaching 10.5% on 22kW three-phase cables.

A rating is a design maximum, not a delivery promise. It describes the current a cable can carry continuously without exceeding its thermal limit, under the conditions the standard specifies: a defined ambient temperature, a defined supply voltage, and nothing else on the circuit. A UK driveway in February, on a house whose main fuse is also feeding an oven and an immersion heater, is not those conditions. The finding on this page is that the gap between the two is consistent, measurable and larger than most drivers assume. It is not that anyone is being mis-sold.

That distinction matters because it changes what a driver should do about it. Four causes account for the whole 8.6%, and they are very unequal. Household supply constraint is the largest at 61% of the total. The cable itself accounts for 19%. The vehicle's onboard charger accounts for 14%, and ambient temperature 6%. The single most expensive assumption in this category is that a slow charge means a poor cable, when in three cases out of five it means a busy consumer unit.

The distribution is as informative as the mean. EV Cable Hub's 2026 session panel found 41.7% of sessions held the cable's rated current for more than 90% of their duration. Those cables were doing exactly what they were sold to do. At the other end, 23.4% of sessions never reached the rated current at any point. The same cable can appear in both groups on different nights, in different homes, at different temperatures, which is why a single average is a poor way to describe cable performance and why the full distribution is published below.

The sections that follow take the finding apart in the order the causes matter. Supply constraint and where the power actually goes come first, then length and voltage drop, then temperature, then the current ratings, then the vehicle side. Cost, durability, connectors and the regional picture follow. Every figure is drawn from one of four EV Cable Hub studies described in full in the methodology, and the limitations of each are published alongside them.

Rated power versus delivered power, by cable rating#

No cable rating in EV Cable Hub's 2026 testing reached its stated figure. The shortfall ranged from 8.1% on 3.6kW cables to 10.5% on 22kW cables, and above 3.6kW it widens steadily as the rating rises.

The pattern is not a straight line from bottom to top. The 3.6kW single-phase cables were the closest to their rating of anything tested, at 8.1% short. The two Mode 2 ratings sat worse than that, at 9.6% and 9.7%, and every rating above 3.6kW got progressively worse: 8.6% at 7.4kW, 9.6% at 11kW and 10.5% at 22kW. The mid-range is where a cable is most likely to deliver close to what it says.

Two mechanisms explain the upper half of the range. Higher-rated cables simply have more headroom to lose: a 5% proportional loss on a 22kW cable is 1.1kW, where the same proportion on a 3.6kW cable is 180W, and the household supply is far more likely to bind on the larger draw. The second is specific to three phase: phase imbalance introduces a loss path that single-phase installations do not have at all, and it is the reason the 11kW and 22kW ratings sit worst in the table.

The 22kW figure deserves its own sentence, because it is the rating most often bought on hope. A 22kW cable needs a three-phase supply to mean anything, and EV Cable Hub's 2026 owner survey found three-phase supply in 3.7% of UK homes. On a single-phase supply a 22kW cable delivers exactly what a 7.4kW cable delivers, at roughly twice the weight and roughly twice the price. The 146 sessions behind the 22kW row are drawn from the small number of homes that genuinely have the supply for it.

Sustained current tells the same story in the unit an electrician would use. Against a 32A rating the panel recorded a mean sustained draw of 29.13A, and against 16A a mean of 14.42A. Peak figures came far closer to the rating than sustained figures did (31.39A peak against 32A rated), which is the crux of the whole study. Cables very often touch their rating. They much less often hold it, and it is the holding that determines how long a charge takes.

The outcome distribution below is the honest version of the headline. Just under half of all sessions were textbook, holding rated current for more than 90% of their duration. About a third held it for part of the session and then derated, usually as household load rose. Around a quarter never got there at all. A further 4.2% of sessions were interrupted before completion and 2.1% needed a manual restart; those two are counted separately because they overlap with the first four categories rather than sitting alongside them.

Where the power actually goes#

Household supply constraint accounts for 61% of all power lost between a cable's rating and its delivery. The cable itself accounts for 19%, the vehicle's onboard charger 14%, and ambient temperature 6%.

This is the section that decides whether the rest of the page is worth trusting, so it comes early and it is unflattering to the obvious conclusion. On a 7.4kW cable the mean total loss is 0.64kW, and the cable contributes 0.12kW of it. The other 0.52kW is the house, the car and the weather. Anybody selling a cable on the promise of recovering the full 8.6% is selling something that does not exist.

Each cause was isolated differently. Supply constraint was established from the property's main fuse rating and the concurrent household load logged at the consumer unit, so a session was classed as supply constrained only where measured spare capacity fell below the cable's demand. Cable loss was measured on the bench rather than inferred, by putting each of the 63 cables through conductor resistance and voltage drop testing at four currents. Vehicle derating was identified where delivered power fell while both supply headroom and cable temperature were unchanged. Temperature was logged at the charge point throughout.

The cable is the only cause that applies to every session. Supply constraint affected 74.2% of them, vehicle derating 38.6% and temperature 51.3%, but conductor resistance is present in 100% of sessions because a cable always has resistance. That is why the cable contribution is small and universal while the supply contribution is large and intermittent, and it is a distinction worth holding on to when reading the length section below.

Property type predicts the supply picture better than anything else EV Cable Hub's 2026 survey collected. A detached house built after 2000 had a mean of 41.2A spare and delivered 7.04kW on a 7.4kW cable. A pre-2000 terrace had 16.8A spare and delivered 6.42kW, and 88.4% of its sessions were supply constrained. The six flats with allocated parking in the sample were the most constrained of all at 91.7%, though six homes is too small a base to publish as a finding and is reported here only for completeness.

Main fuse rating is the single number a driver can check. EV Cable Hub's 2026 survey found 57.2% of homes on a 100A main fuse, 31.6% on 80A and 8.6% still on 60A. On a 100A fuse 41.3% of sessions were constrained; on 60A that rose to 96.2%. The 2.6% of homes running load management, where the charger sheds current when the rest of the house draws it, recorded constrained sessions in only 12.1% of cases, which is the clearest evidence in this dataset that the constraint is manageable rather than fixed.

Cable length and voltage drop#

A 30m charging cable delivers 8.3 percentage points less power than a 3m cable on the same supply. Cable-attributable loss rises from 2.4% at 3m to 10.7% at 30m, and voltage drop at 32A reaches 11.1V over 30m, which is 4.8% of a 230V nominal supply.

Voltage drop is the whole mechanism and it is simple enough to state in one sentence: current flowing through copper meets resistance, resistance turns some of the voltage into heat, and the vehicle sees whatever voltage is left. Double the length and you double the resistance. The power that arrives is the remaining voltage multiplied by the current, so a cable that drops 11.1V on a 230V supply hands the car 4.8% less to work with before anything else in the system has had a say.

The figures in this section isolate cable-attributable loss on unconstrained supplies, which is why they differ from the 8.6% all-causes average quoted elsewhere on this page. A 10m cable on a house with plenty of headroom delivered 7.07kW, against the 6.76kW mean across all 604 sessions on 7.4kW cables. Stating that openly matters, because the two sets of numbers describe different things and reading one as the other produces an apparent contradiction that is not there.

The practical trade-off runs in the opposite direction to the physics. Length costs power, but a cable that will not reach is worth nothing at all, and the cost of buying short is far higher than the cost of the voltage drop. Going from 5m to 10m costs 1.3 percentage points of delivered power, which on a typical overnight charge is a few minutes. Going from 5m to 10m because the first cable would not reach the socket costs the price of a second cable. EV Cable Hub's 2026 order data shows 18.7% of 5m buyers went on to buy something longer, against 6.4% of 10m buyers.

Conductor cross-section is the variable that actually decides how much length costs, and it is the one least often printed on a product page. At 32A over 25m, 6mm² conductor drops 9.3V, or 4.0% of nominal. The same length in 4mm² drops 14.2V, or 6.2%, which puts it outside the 5% voltage drop guidance in the UK wiring regulations. In 2.5mm² a 25m run drops 23.0V. Length and cross-section have to be read together: a long cable in heavy conductor can outperform a short one in thin conductor, and frequently does.

What UK drivers actually buy sits slightly long of what the physics would choose. EV Cable Hub's 2026 order data puts 10m at 34.2% of orders and 5m at 22.8%, with 15m at 18.6% and everything above 20m accounting for under 11% between them. The regret data runs the other way from the loss data: 38.4% of 3m buyers reported their cable was too short against 0.6% of 30m buyers. Our guide to choosing a 10m cable and the conductor and gauge explainer cover how to weigh the two, and an extension is the option where a longer cable is needed occasionally rather than every night.

One more thing follows from the resistance figures and it is about durability as much as speed. A cable that runs hot is a cable losing power, and the two are the same measurement expressed differently. At 32A a 30m run in 2.5mm² dissipates 227W as heat and rose 34.2°C above ambient over four hours, where the same run in 10mm² dissipates 56W and rose 12.4°C. That is not dangerous on its own, but repeated cycling at the top of that range shortens jacket life. Heavier conductor buys durability as well as delivered power, which is not how it is usually sold.

Cold weather, heat, and charging speed#

Below -10°C charging cables deliver 9.3% less power than in mild conditions. EV Cable Hub's 2026 testing recorded a mean of 6.18kW from 7.4kW-rated cables in the coldest sessions against 6.81kW in the 15°C to 20°C baseline band, and cable bend force at -10°C measured 3.8 times the 20°C figure.

Three separate things happen to a charging cable as the temperature falls, and only one of them is about electricity. Conductor resistance itself actually falls slightly in cold copper, which would help. What outweighs it is connector contact resistance, which rose from 0.42 mΩ at 20°C to 0.79 mΩ at -10°C in the bench programme as materials contract and contact pressure changes. The third and largest effect is not in the cable at all: a cold battery accepts less, and the vehicle's own thermal management draws power to warm it while the charge is running.

The temperature curve is not symmetrical and the top of it is worth noting. Delivered power peaks in the 15°C to 20°C band at 6.81kW and falls away in both directions: to 6.18kW below -10°C, and to 6.58kW above 30°C. Heat costs less than cold, at 3.4% against 9.3%, but it is not free, and the mechanism is different. In hot weather the cable and the connector are closer to their thermal derate thresholds before the session starts. Only 24 sessions fell above 30°C, so that row should be read as indicative.

Cold stiffness is the finding drivers actually complain about, and it is barely documented anywhere. Mean bend force across the 63 cables measured 18N at 20°C and 68N at -10°C. At -15°C it reached 91N, and 96.8% of cables exceeded the 50N mark at which the bench programme found one-handed coiling becomes impractical. Minimum bend radius nearly trebled over the same range, from 92mm to 268mm, which is why a cable that coils neatly into a boot in September will not go back in the same space in January.

Jacket material predicts cold behaviour better than price, rating or brand. Of the 28 TPU-jacketed cables tested, 78.6% remained coilable at -10°C, at a mean bend force of 54N. Of the 11 PVC-jacketed cables, 9.1% did, at 96N, and 27.3% of them showed visible cracking below -15°C. TPE sat between the two and the five rubber-compound cables performed close to TPU. No TPU or rubber cable cracked at any temperature tested. This is the one place in the whole study where the specification of the cable, rather than the house or the car, is decisively the thing that matters.

The connector figures put the electrical side of cold weather in proportion. Even at -10°C, connector contact resistance costs 0.81W at 32A and accounts for 3.6% of total cable loss. That is a real measurement and a small number. The 9.3% cold-weather shortfall is overwhelmingly the vehicle and the battery rather than the cable, and a driver replacing a cable to fix slow winter charging is solving the wrong problem, though replacing a PVC cable with TPU will fix the one about not being able to coil it.

10A, 13A, 16A and 32A in the real world#

A 32A cable delivers 2.04 times the power of a 16A cable in real use, not the 2.0 its rating implies. EV Cable Hub's 2026 testing recorded 6.76kW from 32A cables against 3.31kW from 16A, and found 16A sessions are more often limited by the vehicle than by the supply.

Doubling the current more than doubles the delivered power because the two ratings fail for different reasons. A 16A session is usually capped by something upstream of the cable: the vehicle's onboard charger in 34.6% of cases, the supply in 21.4%. A 32A session is capped by the supply in 74.2% of cases. Only 8.4% of 32A sessions ran genuinely unconstrained, against 32.3% of 16A sessions. The larger cable is closer to the limits of the house it is plugged into.

In terms a driver can use, the ladder runs like this. A 10A Mode 2 setting adds 7.7 miles of range an hour. 13A adds 10.0, 16A adds 12.2 and 32A adds 25.0. On a three-phase supply, 16A per phase adds 36.8 and 32A per phase adds 72.9. The step from 16A to 32A is the one that changes daily behaviour: it is the difference between a charge that fits in an overnight window and one that does not.

The 10A versus 13A decision on a Mode 2 charger is a genuine nightly choice and it is poorly documented. The 13A setting delivered 2.71kW against 2.08kW on 10A, a 30% improvement in range added, and that is a real gain on a slow charger where every hour counts. What it costs is socket temperature: EV Cable Hub's 2026 bench programme recorded 38.6% of domestic sockets above 50°C at 13A against 4.2% at 10A, and 14.8% of 13A sessions triggered a thermal derate against 2.1% at 10A. On an old socket, in a cold garage, on a circuit shared with other loads, 10A is the setting that finishes.

Who actually needs 32A is a narrower question than the market implies. A driver covering 8,000 miles a year at 3.7 miles per kWh needs about 2,160kWh, which a 16A cable will deliver in roughly 650 hours of charging across a year, comfortably inside overnight charging alone. The case for 32A is not annual throughput; it is recovery after an unusual day, and the ability to complete a charge inside a narrow cheap-rate window. Both are real, and both are about the shape of the demand rather than its size.

The common error runs the other way. Buying a 32A cable for a supply that cannot deliver 32A produces exactly the delivered power a 16A cable would have produced, at a higher price and a heavier cable to coil. Our 16A against 32A comparison and the guide to charging amps set out how to check what a supply can actually carry before buying for it.

Your car is often the limit, not your cable#

44.1% of UK drivers own a charging cable rated below their vehicle's maximum AC intake, and are charging more slowly than their car allows. A further 28.6% own a cable rated above what their car can accept, and are paying for headroom they will never use.

Only 27.3% of the drivers in EV Cable Hub's 2026 owner survey had a cable matched to their vehicle. The mismatch is not random: it clusters heavily on one side, because 7.4kW is the default single-phase cable and a large share of the current UK fleet can accept 11kW on AC. A driver with an 11kW car and a 7.4kW cable loses a mean of 2.68kW of intake and, across a year of home charging, about 68 hours of charging time. That is time rather than money on most tariffs, and it becomes money on a timed one.

Onboard charger derating is the mechanism behind the vehicle side of this. A car's published AC maximum is what its onboard charger can convert under good conditions. In practice it converts slightly less, and it converts less again when the battery is cold, when the pack is near full, or when the car is running thermal management. Across the 62 vehicles in the table below the observed mean AC draw ran consistently below the published maximum: an 11kW car typically drew between 9.84kW and 10.24kW, and a 6.6kW car between 6.09kW and 6.18kW.

The AC and DC figures are the largest source of confusion in this category and they are not comparable. A car advertised as charging at 150kW is describing DC rapid charging at a public site, where the charger feeds the battery directly and the onboard charger is bypassed entirely. The AC figure (7.4kW, 11kW or 22kW) is what the car can accept at home through a cable, and it is set by the onboard charger's own hardware. No home cable, of any rating, will make a 7.4kW onboard charger accept 11kW.

There is a caveat on the table below that should be read before it is quoted. It lists 62 vehicles against the cable rating owners of each typically buy, which is 7.4kW across the board; it is a reference for matching a car to a cable, not a measure of the surveyed base. The 44.1%, 28.6% and 27.3% shares come from the 2,140-driver survey and are weighted to the UK electric vehicle parc, which is why they do not match the row-by-row outcomes in the vehicle list. Our UK cable statistics study covers the ownership side in more depth, and the charging cable range is organised by rating and length.

There is a version of this finding that is worth resisting. It would be easy to read 44.1% and conclude that four drivers in ten should replace their cable, and the data does not support that. A driver with an 11kW car, a single-phase supply and a 7.4kW cable is not under-equipped, because 11kW is unreachable at their property at any price. Nor are the 28.6% who over-buy making a costly mistake: a 7.4kW cable on a 6.6kW car delivers everything the car can take, and the cable is simply not the constraint. Only the subset of the 44.1% with a three-phase supply describes a mismatch a cable purchase would fix.

Granny chargers and Mode 2 reality#

Mode 2 granny chargers delivered 2.08kW against a 2.3kW rating on the 10A setting and 2.71kW against 3.0kW on 13A, shortfalls of 9.6% and 9.7%. Both sit above the 8.6% average and above every single-phase cable rating measured.

The proportional gap is worst here for a straightforward reason: a Mode 2 charger is working against a domestic socket and a ring main that were never designed for a four-hour continuous load. The absolute loss is tiny (0.22kW on the 10A setting), but the base is tiny too, so the percentage looks alarming. Among the six ratings tested only the three-phase ones were proportionally worse, at 9.6% on 11kW and 10.5% on 22kW.

Socket temperature is the finding that matters most in this section, and it is a safety observation rather than a performance one. At 6A the mean socket temperature after four hours was 28.4°C. At 10A it was 41.2°C with a peak of 52.8°C. At 13A it was 52.6°C with a peak of 68.4°C, 38.6% of sockets exceeded 50°C and 14.8% of sessions auto-derated on the charger's own thermal sensor. That last figure is the equipment doing its job, and it is also the clearest signal that 13A on an unknown socket is the top of the envelope rather than a comfortable setting.

The 10A versus 13A decision should therefore be made on the socket, not on the car. On a modern, dedicated, unshared outlet, 13A adds 80.2 miles over twelve hours against 61.6 at 10A, and the thermal margin is adequate. On an outdoor socket of unknown age, on an extension, or on a circuit shared with anything else, 10A is the setting that will still be charging in the morning. The lower settings are genuinely useful too: 6A delivered 1.24kW and 8A delivered 1.66kW, and both ran at socket temperatures under 42°C throughout.

Realistic overnight range is the number to plan around. Twelve hours at 10A adds 92.4 miles and at 13A adds 120.4. A 36kWh charge takes 17 hours 18 minutes at 10A and 13 hours 17 minutes at 13A. For a driver covering 25 miles a day that is comfortably enough overnight; for a driver covering 80 it is not, and no setting on a Mode 2 charger will make it enough.

The honest framing is that a granny charger is a backup and an occasional-use tool rather than a primary charging solution. It is the right piece of equipment for a second home, a relative's driveway, a campsite or the week a wallbox is being replaced. Used every night on the same socket for years, it is asking a domestic outlet to do something it was not specified for. Our Mode 2 explainer and the Mode 2 against Mode 3 comparison cover where the line sits, and the granny charger range lists the settings each unit offers.

The 6A and 8A settings deserve more attention than they get. Both are present on most Mode 2 chargers, both ran at socket temperatures below 42°C throughout four-hour testing, and 8A adds 73.7 miles over twelve hours, enough for the great majority of daily driving. For a driver using a granny charger nightly on a socket of unknown provenance, 8A is a materially safer setting than 13A at a cost of about 47 miles a night, and almost nobody is told this.

Single phase versus three phase#

Three-phase cables lose proportionally more power than single phase, at 10.1% against 8.4%, because phase imbalance adds a loss path single-phase installations do not have. Only 3.7% of UK homes in EV Cable Hub's 2026 survey had a three-phase supply.

Three phase means three live conductors arriving at the property instead of one, offset in time so that power is delivered more evenly. It is the standard arrangement for industrial and commercial premises across the UK and Europe, and it is what makes an 11kW or 22kW AC charge possible at all. Almost no UK housing has it. The 3.7% figure in this survey is consistent across property types and skews towards large detached properties and converted commercial buildings.

Phase imbalance is the specific mechanism behind the wider shortfall. The three phases rarely carry identical current, and delivered power is governed by the weakest of them. EV Cable Hub's 2026 session panel recorded a mean imbalance of 4.2% across 22kW sessions and a worst case of 11.8%. A single-phase installation cannot have this problem, which is why its mean shortfall is 1.7 percentage points lower despite being the technically cruder arrangement.

In absolute terms three phase still wins comfortably and it is important not to lose that in the percentages. An 11kW cable delivered 9.94kW and a 22kW cable delivered 19.70kW, against 6.76kW on the best single-phase option. A 22kW cable adds 72.9 miles of range an hour against 25.0 on 7.4kW. The larger proportional shortfall is a footnote against roughly three times the delivered power.

Whether to upgrade is a question with a clear shape. The mean quoted cost of a three-phase supply upgrade in the 2026 survey was £3,840, and 61.4% of the drivers who went through with it reported it worthwhile. That is a solid majority and it is not overwhelming. The case is strongest for high-mileage drivers, homes with two electric vehicles, and properties where a three-phase supply is already at the boundary. It is weakest for a single car covering ordinary mileage, where a 7.4kW cable overnight already finishes the job before morning. Our single phase against three phase guide covers the installation side.

How long it really takes to charge#

A 60kWh EV charging from 20% to 80% on a 7.4kW cable takes 5 hours 20 minutes in real use, against the 4 hours 52 minutes its rating implies. EV Cable Hub's 2026 charging time matrix covers every combination of battery size and cable rating measured.

Every figure in the matrix below is derived from measured delivered power rather than from the rating on the cable. The arithmetic is deliberately simple and open so it can be checked: energy needed divided by the mean delivered power recorded for that rating. A 60kWh battery charging from 20% to 80% needs 36kWh; at the 6.76kW measured on 7.4kW cables that is 5 hours 20 minutes, where the 7.4kW rating would suggest 4 hours 52 minutes.

Read the matrix down a column to see how battery size scales and across a row to see what a cable rating buys. The 2.3kW and 3.0kW columns are the ones that reframe the discussion: a 77kWh car needs 22 hours 13 minutes on a 10A granny charger to go from 20% to 80%, which is not an overnight charge under any definition. The 7.4kW column is where almost all UK home charging actually happens, and every battery size up to 100kWh clears 20% to 80% inside nine hours on it.

The absolute penalty of the shortfall shrinks as the cable gets faster while the proportional penalty grows. On a 60kWh battery the gap between rated and real is 1 hour 39 minutes on a 2.3kW charger and 12 minutes on a 22kW cable. In proportional terms the 22kW figure is the worse of the two, but 12 minutes at the end of a 1 hour 50 minute charge is a different kind of problem from 99 minutes at the end of a 17-hour one.

These are measured rather than calculated times, and there is one thing they do not include. Charging slows near the top of the pack on AC as on DC, so a 20% to 80% window is a fair test and a 20% to 100% window would run longer than a straight-line extrapolation suggests. The matrix stops at 80% for that reason. It also assumes an uninterrupted session; the 4.2% of sessions that were interrupted and the 2.1% that needed a manual restart are excluded from these times.

For a driver working out whether a charge fits an overnight window, the useful test is the one in the cost section that follows. A 7.4kW cable inside a six-hour cheap window delivers about 40.6kWh at measured rates, which covers a 20% to 80% charge on anything up to about a 68kWh battery. Above that the charge overruns the window and the last part of it completes at the day rate.

What the gap costs you#

For a driver on a timed overnight tariff the 8.6% shortfall costs an average of £64 a year, and £108 a year on a four-hour window. On a flat-rate tariff it costs nothing at all in money, because the energy is charged at the same price whenever it arrives.

The mechanism is worth setting out before the numbers, because it is not the one people expect. The shortfall does not waste energy. A car that needs 36kWh takes 36kWh whether the cable delivers it in five hours or six. What the shortfall does is push part of that energy outside the cheap window, where it is charged at the day rate instead of the off-peak rate. On a flat tariff there is no window to overrun and the cost is zero; the shortfall costs time only.

The arithmetic behind each row is open and can be checked. On a four-hour window the shortfall pushes 552 kWh a year to the day rate, and the gap between the 26.4p day rate and the 6.8p off-peak rate is 19.6p, which gives £108. On a six-hour window at 25.8p against 7.9p it is 342kWh at 17.9p, which gives £61. The £64 headline is the mean of the five fixed overnight windows in the table, from four hours to eight.

Window length matters far more than the cable does. Going from a four-hour window to an eight-hour one takes the annual cost of the gap from £108 to £30, a larger swing than anything a driver could achieve by changing cable. The dynamic half-hourly structure sits at £94, higher than most fixed windows despite the lowest average off-peak rate at 4.1p, because the cheap periods are scattered rather than contiguous and a long charge cannot sit inside them.

Cost per mile puts the whole thing in perspective and it is the number worth quoting. On a flat 24.8p tariff every cable rating costs the same 6.7p a mile, because the price does not vary by time. On a six-hour overnight tariff a 7.4kW cable costs 2.1p a mile and a 2.3kW Mode 2 charger costs 3.1p, because the slow charger cannot finish inside the window. On a dynamic tariff the same comparison is 1.4p against 2.4p. The gap between tariff structures is three times the gap between cable ratings.

The practical conclusion is unglamorous. For a driver on a flat tariff, the shortfall is a time cost of about 42 hours a year and no money at all. For a driver on a narrow timed window, moving to a longer window, or scheduling the charge to start earlier, will recover more than a faster cable would. The cable is worth changing when it is genuinely the binding constraint, which the calculators further down this page will say plainly one way or the other.

The 42-hour annual time cost is the figure that travels furthest and it is worth putting in context. Spread across a year of overnight charging it is about seven minutes a session, almost all of it while the driver is asleep. It becomes real when a charge has to fit a window (a narrow cheap-rate period, or an hour between two journeys), and it is almost entirely notional the rest of the time. Both of those statements are true at once, and reporting only the first would overstate the finding.

Cable durability and failure#

4.8% of charging cables had failed within three years in EV Cable Hub's 2026 owner survey, and connector latch failure accounted for 38.2% of those faults. Mean time to first fault across the surveyed base was 3.4 years.

Cables fail at the ends. Connector latch failure and jacket abrasion together account for 62.8% of all faults reported, and both are mechanical rather than electrical. The latch is the small plastic catch that holds the connector in the inlet; it is cycled twice a day for years and it is the part that gives up first. Internal conductor faults, which is what most people imagine when they think of a cable failing, account for 14.1%.

Bench cycle testing found the equipment considerably tougher than its own specifications claim. Cables reached a mean of 4,120 insertions before any measurable contact degradation and 11,840 before failure, against a mean manufacturer-stated rating of 10,000 cycles. Some 71.4% of cables exceeded their stated rating in testing. At the 412 coil-and-uncoil cycles a year the average driver reports, a cable's mating life is not the constraint on its useful life.

What actually shortens that life is the driveway. EV Cable Hub's 2026 survey found 31.6% of drivers drive over their own cable at least weekly, and the bench programme found 22.2% of cables showed measurable damage after 500 vehicle crossings. A cable is well able to survive being run over occasionally and much less able to survive it twice a day for three years. This is the single cheapest fault to avoid and the most commonly ignored.

Ingress protection is the area where published ratings and measured behaviour diverge most. Mean rating across the cables tested was IP55, 34.9% carried a rating of IP67 or above, and 41.3% passed a 30-minute immersion test. That last figure sits above the share carrying an IP67 rating, which means a meaningful number of cables are more water-resistant than their rating claims. It also means well over half failed an immersion test, and a connector lying in a puddle overnight is a realistic scenario on a British driveway rather than an unfair one.

• Cables failed within 3 years: 4.8%
• Cables failed within 5 years: 11.2%
• Mean time to first fault: 3.4 years
• Connector latch failure share of faults: 38.2%
• Cable jacket abrasion share: 24.6%
• Internal conductor fault share: 14.1%
• Control pilot circuit fault share: 12.8%
• Water ingress share: 6.4%
• Other faults: 3.9%
• Bench cycle test, insertions to first contact degradation: 4,120
• Bench cycle test, insertions to failure: 11,840
• Manufacturer-stated cycle rating, mean: 10,000
• Cables exceeding their stated cycle rating in testing: 71.4%
• Mean coil and uncoil cycles per year per driver: 412
• Drivers who drive over their cable at least weekly: 31.6%
• Cables showing measurable damage after 500 vehicle crossings: 22.2%
• Mean IP rating across cables tested: IP55
• Cables rated IP67 or above: 34.9%
• Cables passing a 30-minute immersion test: 41.3%

Connector types and their real ceilings#

Type 2 connectors carried a mean of 6.76kW in EV Cable Hub's 2026 UK home testing, against a theoretical single-phase ceiling of 7.4kW. Every connector standard on UK roads underdelivers against its published maximum.

Type 2 is the settled UK standard and the numbers say so without much room for argument: it accounted for 94.1% of UK cables sold in 2026. It carries seven pins, three live, neutral, earth, and two communication pins that carry the control pilot and proximity signals. It is also the only AC connector on the UK market that supports three phase, which is what gives it a 22kW theoretical ceiling against Type 1's 7.4kW.

Type 1 survives on older imported vehicles and accounted for 3.8% of cables sold. It has five pins, no three-phase capability, and a measured mean of 6.61kW against the same 7.4kW theoretical ceiling as single-phase Type 2, slightly worse on a much smaller sample. A driver with a Type 1 inlet is not disadvantaged at home in any meaningful way; the constraint appears at public sites, where Type 2 sockets are now effectively universal.

The DC standards belong in a different conversation and are included here only to stop the comparison being made wrongly. CCS Combo 2 adds two DC pins below the Type 2 pattern and covers 25kW to 400kW. CHAdeMO covers the same theoretical range, is most commonly seen at 50kW in the UK, and accounted for 8.2% of UK public connectors in 2026. Neither figure has anything to say about home charging, because DC charging bypasses the onboard charger entirely and no home cable carries it.

At the bottom of the ladder, a three-pin BS 1363 plug has a 3.0kW theoretical ceiling at 13A and measured 2.71kW. The connector measurements underneath all of these are the part worth keeping: mean contact resistance rose from 0.42 mΩ new to 0.71 mΩ after 5,000 mating cycles, mean mating force was 62N and mean withdrawal force 48N. A connector that has been used twice a day for seven years is measurably worse than a new one, and the degradation is gradual enough that nobody notices it happening. Our connector types guide covers which vehicles use what.

• Type 1 (SAE J1772) pins: 5
• Type 1 theoretical AC maximum: 7.4 kW
• Type 1 measured mean in UK testing: 6.61 kW
• Type 1 share of UK cables sold 2026: 3.8%
• Type 2 (IEC 62196-2, Mennekes) pins: 7
• Type 2 single phase theoretical maximum: 7.4 kW
• Type 2 measured mean single phase: 6.76 kW
• Type 2 three phase theoretical maximum: 22 kW
• Type 2 measured mean three phase: 19.70 kW
• Type 2 share of UK cables sold 2026: 94.1%
• CCS Combo 2 additional DC pins: 2
• CCS theoretical range: 25 kW to 400 kW
• CHAdeMO theoretical range: 25 kW to 400 kW
• CHAdeMO most common UK rating: 50 kW
• CHAdeMO share of UK public connectors 2026: 8.2%
• Three-pin BS 1363 theoretical: 3.0 kW at 13A
• Three-pin measured mean: 2.71 kW
• Mean connector contact resistance, new: 0.42 mΩ
• Mean connector contact resistance after 5,000 cycles: 0.71 mΩ
• Connector mating force, mean: 62 N
• Connector withdrawal force, mean: 48 N

Vehicle to load and discharge rates#

V2L-capable vehicles delivered a mean of 2.94kW against published ratings averaging 3.2kW in EV Cable Hub's 2026 testing, a shortfall of 8.1%. V2L adapter demand grew to 18.4% of EV Cable Hub orders in the first half of 2026.

Vehicle to load runs the process in reverse: the car's inverter supplies mains-voltage AC through an adapter on the charging inlet, turning the traction battery into a very large portable generator. The 8.1% shortfall against published ratings is close to the 8.6% measured on charging, which is a useful cross-check. The same conversion and thermal management losses appear whichever way the energy is moving.

Across the seven capable vehicles tested the spread was narrow. The three 3.6kW Korean models delivered between 3.28kW and 3.34kW. The 3.7kW Renault 5 E-Tech delivered 3.38kW, the 3.0kW BYD Atto 3 delivered 2.74kW and the Volvo EX30 2.76kW, and the 2.2kW MG4 delivered 2.02kW. Not one reached its published figure, and not one fell more than 9% short of it.

What that actually powers is the useful question. At 2.94kW mean output a V2L adapter will run a kettle, a fridge, lighting, a laptop and most power tools, but not a domestic immersion heater or an electric shower. A 2kW load drawn from a 60kWh battery at 80% charge ran for a mean of 19.2 hours before hitting the mean 20% state-of-charge cutoff at which vehicles stop discharging. That is a genuinely useful reserve during a power cut and a genuinely poor way to run a house indefinitely.

The demand growth is the commercial finding. V2L adapters went from 9.7% of EV Cable Hub orders in the first half of 2025 to 18.4% in the first half of 2026, growth of 89.7% year on year. What buyers say they want it for is mostly leisure rather than resilience: 41.2% cited camping, 28.6% power-cut backup, 19.4% work tools and 6.1% charging another electric vehicle. The V2L adapter range covers the vehicles listed here.

• V2L adapter share of orders, H1 2026: 18.4%
• V2L adapter share of orders, H1 2025: 9.7%
• Year on year growth in V2L adapter orders: 89.7%
• Mean published V2L rating across capable vehicles: 3.2 kW
• Mean measured V2L output: 2.94 kW
• Mean V2L shortfall: 8.1%
• Kia EV6 published V2L 3.6 kW, measured 3.31 kW
• Hyundai Ioniq 5 published 3.6 kW, measured 3.28 kW
• Kia EV9 published 3.6 kW, measured 3.34 kW
• MG4 published 2.2 kW, measured 2.02 kW
• BYD Atto 3 published 3.0 kW, measured 2.74 kW
• Renault 5 E-Tech published 3.7 kW, measured 3.38 kW
• Volvo EX30 published 3.0 kW, measured 2.76 kW
• Minimum battery state of charge for V2L, mean cutoff: 20%
• Mean runtime for a 2kW load from a 60kWh battery at 80% charge: 19.2 hours
• Share of V2L buyers citing camping as primary use: 41.2%
• Share citing power cut backup: 28.6%
• Share citing work tools: 19.4%
• Share citing EV to EV charging: 6.1%
• Share citing other uses: 4.7%

What UK drivers actually own and buy#

34.2% of UK EV drivers own a 10m charging cable, making it the most common length by a clear margin, and 22.8% own 5m. EV Cable Hub's 2026 survey of 2,140 UK drivers found 18.7% of 5m owners went on to buy a longer cable.

The ownership pattern is more layered than a single cable per driver. Only 42.8% own exactly one cable; 38.4% own two and 18.8% own three or more, for a mean of 1.84 cables per driver. Separately, 61.2% own a granny charger, and 34.7% of those have never used it. That last figure is the clearest statement in the dataset about what a Mode 2 charger is for: it is insurance, and most insurance is never claimed on.

Repeat purchase behaviour is where the order data earns its keep. Mean spend on a first cable was £118 and on a replacement £146, and the mean gap between the two purchases was 19 months. Of drivers who bought a second cable, 68.4% bought a longer one, 22.1% bought a higher rating and 14.8% bought coiled. The direction is almost entirely towards length, which is consistent with the regret data: 14.2% report their cable is too short, against 22.6% who say it is too heavy.

The complaint that dominates everything else is winter stiffness. Some 48.1% of drivers, nearly half the surveyed base, report their cable is too stiff in cold weather, more than twice the share reporting weight and more than three times the share reporting length. That is a materials question rather than a specification one, and the jacket material findings earlier on this page are the direct answer to it.

Storage and handling produce the last set of figures, and one of them is a safety point. Most drivers (71.4%) store the cable in the boot, 18.2% use a wall holder and 10.4% leave it plugged into the charger. Some 38.6% use a cable bag. And 31.6% have driven over their own cable while 12.8% have tripped over it. Theft is rare at 2.1%, and cable locks are correspondingly rare at 8.4%, which is a proportionate response to a small risk rather than a gap in the market.

Two figures in this section sit awkwardly together and the tension is the interesting part. Some 68.4% of replacement buyers went longer, yet only 14.2% of owners report their cable is too short. The gap is explained by the ones who already solved it: a driver who bought a second, longer cable no longer reports the first one as too short. Ownership surveys measure the state after the fix, which is why the order data and the survey data have to be read together rather than separately.

• Drivers owning exactly one charging cable: 42.8%
• Drivers owning two cables: 38.4%
• Drivers owning three or more: 18.8%
• Mean cables owned per driver: 1.84
• Drivers who also own a granny charger: 61.2%
• Drivers who have never used their granny charger: 34.7%
• Mean spend on first cable: £118
• Mean spend on replacement cable: £146
• Drivers who bought a longer cable as replacement: 68.4%
• Drivers who bought a higher-rated cable as replacement: 22.1%
• Drivers who bought a coiled cable as replacement: 14.8%
• Mean time between first and second cable purchase: 19 months
• Drivers who report their cable is too short: 14.2%
• Drivers who report their cable is too heavy: 22.6%
• Drivers who report their cable is too stiff in winter: 48.1%
• Drivers who store their cable in the boot: 71.4%
• Drivers who store it in a wall holder: 18.2%
• Drivers who leave it plugged into the charger: 10.4%
• Drivers who use a cable bag: 38.6%
• Drivers who have tripped over their cable: 12.8%
• Drivers who have driven over their own cable: 31.6%
• Drivers who have had a cable stolen: 2.1%
• Drivers who use a cable lock: 8.4%

Coiled versus straight cables#

Coiled cables delivered 3.1% less power than straight cables of the same rating and conductor size in EV Cable Hub's 2026 testing, at 6.55kW against 6.76kW, because the coiled section retains heat and derates sooner.

The mechanism is thermal rather than electrical. A coiled cable has the same conductor and the same resistance as a straight one of the same length, but the coil holds the heat that resistance generates instead of shedding it into the air. Mean temperature rise at 32A after four hours measured 31.4°C on coiled cables against 24.8°C on straight, and a hotter conductor derates earlier and further.

The second thing to know about coiled cables is that the stated length is not the usable length. Mean extended length across the coiled cables tested was 84.2% of the figure printed on the packaging, because a coil that is pulled completely straight is a coil that will not retract. A 10m coiled cable is, in practice, about 8.4m of reach. Mean retraction force measured 34N, which is the pull a driver has to overcome every time they extend it.

The trade-off is genuine and it is about handling rather than performance. A coiled cable stores itself, does not lie in a puddle, and does not need to be wound. Some 62.4% of coiled buyers cited tidiness as their reason for choosing one. What they give up is 3.1% of delivered power, about 16% of nominal reach, and a cable that is harder to manage in cold weather, since the retraction spring stiffens with everything else.

Coiled cables account for 11.8% of UK orders and 58.1% of coiled owners said they would buy coiled again, the lowest repeat-purchase intention of any format in the survey, and the figure that best summarises the trade-off. It is a real preference held by a real minority, and it comes at a measurable cost. Our coiled against straight guide sets out who each suits.

• Coiled cable mean delivered, 7.4kW rated: 6.55 kW
• Straight cable mean delivered, 7.4kW rated: 6.76 kW
• Coiled cable penalty: 3.1%
• Coiled cable mean temperature rise at 32A after 4 hours: 31.4°C
• Straight cable equivalent: 24.8°C
• Coiled cable extended length as share of stated: 84.2%
• Coiled cable mean retraction force: 34 N
• Coiled share of UK orders 2026: 11.8%
• Coiled buyers citing tidiness as reason: 62.4%
• Coiled owners who would buy coiled again: 58.1%

Tethered versus untethered#

Tethered and untethered setups delivered within 0.4% of each other in EV Cable Hub's 2026 testing, at 6.79kW against 6.76kW on a 7.4kW rating. The choice is a convenience decision, not a performance one.

A difference of 0.03kW is inside the measurement noise of the session panel and should be treated as no difference at all. This settles an argument that is had constantly and has no basis in the data: a cable permanently attached to a wallbox and a cable carried in the boot deliver the same power to the same car on the same supply, because the conductor does not know which end is fixed.

What does differ is length. Mean installed tethered cable length was 6.2m, shorter than the 10m most untethered buyers choose, because a tethered cable has to store on the unit. That is the one place where the format has a real performance implication, and it works in the tethered cable's favour: a shorter run drops less voltage. It also constrains where the car can park, permanently.

UK homes split 58.7% tethered to 41.3% untethered, and both groups contain a meaningful minority who wish they had chosen the other. Some 18.4% of tethered owners wanted untethered and 24.6% of untethered owners wanted tethered. The regret runs slightly heavier among untethered owners, who have to fetch and stow the cable every time. A third of them (34.1%) carry a spare cable as well, which is the untethered format's actual advantage in practice: the cable goes with the car. Our tethered against untethered guide covers the installation side of the decision.

• Tethered mean delivered, 7.4kW: 6.79 kW
• Untethered mean delivered, 7.4kW: 6.76 kW
• Difference: 0.4%
• UK homes with tethered chargers: 58.7%
• UK homes with untethered chargers: 41.3%
• Tethered owners who wanted untethered: 18.4%
• Untethered owners who wanted tethered: 24.6%
• Mean tethered cable length installed: 6.2 m
• Untethered owners carrying a spare cable: 34.1%

Charging losses, wall to battery#

11.4% of the energy drawn from the wall does not reach the battery on a typical UK home charging session. EV Cable Hub's 2026 measurements put cable losses at 1.9 percentage points of that, with the remainder in the onboard charger and battery thermal management.

This is a different measurement from everything above it and the distinction matters. The delivered-power figures on the rest of this page are measured at the vehicle inlet, so they describe what arrives at the car. The wall-to-battery figure describes what happens after that, inside the vehicle, and the two should be added rather than compared. A driver paying for 100kWh at the meter gets 88.6kWh into the pack.

The breakdown is dominated by conversion. The onboard charger, which turns mains AC into the DC the battery needs, accounts for 6.2 percentage points, more than three times the cable's contribution. Battery thermal management accounts for 2.4 points, cable resistance 1.9 and standby and communication overhead 0.9. Best recorded wall-to-battery efficiency across the panel was 93.1% and the worst was 81.6%, a spread of over eleven points between otherwise ordinary home setups.

Temperature moves the whole figure substantially. Mean loss was 10.8% at 20°C, 14.2% at 0°C and 17.6% below -10°C. That is the same cold-weather mechanism described earlier, seen from the meter rather than from the cable, and it is why winter charging costs more than winter mileage alone would suggest. Nearly a fifth of the energy bought on a freezing night never reaches the battery.

In money the numbers are modest and worth stating precisely so they are not overstated. The average driver wastes 312kWh a year to these losses, which costs £77 at a 24.8p flat rate and £25 at a 7.9p overnight rate. Nothing on this page will recover most of it, because conversion losses are a property of the vehicle. Charging in a garage rather than on an open driveway, and charging when the battery is warm from driving rather than cold from standing, are the two things that measurably help.

There is a reason this measurement is reported separately rather than folded into the headline. Adding wall-to-battery losses to the rated-against-delivered gap would double-count, because the two are measured either side of the same point. The cable's 1.9 percentage points here and its 19% share of the 8.6% shortfall above are the same physical loss described in two different denominators, and quoting them together as if they were separate would overstate the cable's contribution twice over.

• Mean wall-to-battery loss: 11.4%
• Cable resistance share: 1.9 percentage points
• Onboard charger conversion share: 6.2 pp
• Battery thermal management share: 2.4 pp
• Standby and communication share: 0.9 pp
• Best recorded wall-to-battery efficiency: 93.1%
• Worst recorded: 81.6%
• Mean loss at 20°C: 10.8%
• Mean loss at 0°C: 14.2%
• Mean loss below -10°C: 17.6%
• Energy wasted per year, average driver: 312 kWh
• Cost of that waste at 24.8p flat rate: £77
• Cost of that waste at 7.9p overnight rate: £25

Regional and seasonal patterns#

Scottish drivers recorded 4.2% lower mean delivered power than drivers in southern England, at 6.51kW against a southern mean of 6.80kW, and the difference is almost entirely explained by ambient temperature. June sessions delivered 5.9% more than January sessions.

The regional table reads as a temperature map rather than an infrastructure map. The four lowest regions (Scotland at 6.51kW, Northern Ireland at 6.61kW, the North East at 6.58kW and the North West at 6.62kW) are the four coldest, and the four highest sit in the south. The spread from top to bottom is 4.8%, which is smaller than the gap between a pre-2000 terrace and a post-2000 detached house in the same street.

It is worth being explicit that this is not a finding about regional infrastructure or regional cable quality. The same cables, on the same ratings, delivered less in colder places for the same reason they delivered less in colder months. A Scottish driver with a 7.4kW cable is not disadvantaged by anything except the weather, and the practical consequence is roughly seven minutes on a typical overnight charge.

The seasonal series runs from 6.44kW in January to 6.82kW in June, a spread of 5.9% measured against the January figure. The study window closed at the end of June, so the second half of the year is not covered and the full annual swing will be wider at both ends. The January and February figures also carry the smallest sub-zero subsample in the dataset (107 sessions in total below 0°C) and should be read with that in mind.

Session timing is the last part of the regional and seasonal picture and it is the part that connects to cost. Mean session start time across the panel was 00:34, and 68.4% of sessions started inside a cheap window. But 41.2% of sessions overran the window they started in, which is the behavioural fact underneath the whole cost section: the shortfall becomes expensive at precisely the point where a charge that was supposed to finish cheaply does not.

• Scotland mean delivered, 7.4kW: 6.51 kW
• North East England: 6.58 kW
• North West England: 6.62 kW
• Yorkshire and the Humber: 6.64 kW
• West Midlands: 6.71 kW
• East Midlands: 6.72 kW
• Wales: 6.68 kW
• East of England: 6.78 kW
• South West England: 6.81 kW
• South East England: 6.82 kW
• Greater London: 6.79 kW
• Northern Ireland: 6.61 kW
• January mean delivered: 6.44 kW
• February: 6.48 kW
• March: 6.62 kW
• April: 6.74 kW
• May: 6.81 kW
• June: 6.82 kW
• Seasonal spread, January to June: 5.9%
• Mean session start time: 00:34
• Share of sessions starting inside a cheap window: 68.4%
• Share of sessions overrunning the cheap window: 41.2%

Safety, protection and standards#

14.8% of Mode 2 sessions on the 13A setting triggered a thermal derate in EV Cable Hub's 2026 testing, and 38.6% of domestic sockets exceeded 50°C at that current. Every figure below is measured rather than specified.

The thermal derate figure should be read as protection working rather than as equipment failing. A Mode 2 charger that senses its plug getting hot and reduces current is doing exactly what it was designed to do, and the 14.8% of sessions where that happened at 13A finished, more slowly, without incident. The number worth attention is the 38.6% of sockets above 50°C, because a socket does not have a sensor and a socket that runs warm every night for years is the failure mode that matters.

Residual current protection is close to universal in the cables tested but not uniform in type. Some 89.4% carried an integrated RCD, and 10.6% carried none. Of those that did, 62.8% used Type A protection and 26.6% used Type B or an RDC-DD device capable of detecting the smooth DC residual currents an electric vehicle can produce. Mean recorded trip time was 24 milliseconds and 96.8% of devices tripped within 40 milliseconds. The protection is fast; the question is whether it is the right type for the fault it may have to see.

The installation side of the survey found 71.2% of homes charging on a dedicated EV circuit and 28.8% charging from a general ring main, which is the arrangement Mode 2 charging implies. Only 34.6% had Type B protection at the consumer unit. Mean earth loop impedance measured 0.42Ω across the surveyed homes, with 6.4% exceeding 0.8Ω. That is a small minority, but it is the group for whom fault protection is least likely to operate as intended.

The cables themselves were in good order electrically. Of the 63 tested, 98.4% passed insulation resistance testing at 500V, with a mean insulation resistance of 412 MΩ. No cable in the programme presented an electrical safety defect. The measured risks in this section sit at the socket, at the consumer unit and in the choice of protection device, not in the cable, and it is worth saying that in a section that could easily be written to imply otherwise.

• Mode 2 sessions triggering thermal derate at 13A: 14.8%
• At 10A: 2.1%
• Domestic sockets exceeding 50°C at 13A: 38.6%
• At 10A: 4.2%
• Peak socket temperature recorded: 68.4°C
• Cables with an integrated RCD: 89.4%
• Cables with Type A RCD protection: 62.8%
• Cables with Type B or RDC-DD protection: 26.6%
• Cables with no integrated protection: 10.6%
• Mean RCD trip time recorded: 24 ms
• Cables tripping within 40 ms: 96.8%
• Homes with a dedicated EV circuit: 71.2%
• Homes charging from a general ring main: 28.8%
• Homes with a Type B RCD at the consumer unit: 34.6%
• Mean earth loop impedance recorded: 0.42 Ω
• Homes exceeding 0.8 Ω: 6.4%
• Cables passing insulation resistance at 500V: 98.4%
• Mean insulation resistance: 412 MΩ

The 2026 cable performance league table#

The highest-performing cable in EV Cable Hub's 2026 testing delivered 7.28kW against its 7.4kW rating, a shortfall of just 1.6%. The lowest delivered 5.41kW, a shortfall of 26.9%, and the spread between the two is 34.6%.

The 63 cables are reported by specification rather than by brand, because the finding is about specification. Conductor cross-section correlated with delivered power at 0.78, a strong relationship. Length correlated at -0.71, in the expected direction. Price correlated at 0.21, which is close to no relationship at all. The mean price of cables in the top performance quartile was £142 and in the bottom quartile £129, a difference of £13 across a 34.6% performance spread.

That is the most counterintuitive result in the study and it deserves to be stated without hedging: paying more does not reliably buy a faster cable. What buys a faster cable is heavier conductor and shorter length, and neither of those is reliably reflected in price. A well-specified 6mm² cable at £120 will outperform a thinner one at £180, and nothing on either product page makes that obvious to a buyer who does not already know what to look for.

The distribution is tighter than the extremes suggest. Some 22.2% of cables came within 5% of their rating and 47.6% sat between 5% and 10% off, so roughly seven cables in ten performed within 10% of what they claimed. A further 25.4% were between 10% and 20% off, and 4.8% (three cables) were more than 20% off. Median shortfall across all 63 was 8.4%, close to the 8.6% session mean.

The practical guidance that falls out of this is short. Check the conductor cross-section before the price. Buy the length you need and not more. On a single-phase supply, a 32A cable is the top of what is useful. And be sceptical of any performance claim on a cable that does not state its conductor size, because the one variable that predicts performance is the one least often published. The conductor and gauge explainer covers what to look for, and the long cable guide covers sizing beyond 15m.

• Best performing cable, delivered: 7.28 kW (1.6% shortfall)
• Worst performing, delivered: 5.41 kW (26.9% shortfall)
• Spread between best and worst: 34.6%
• Median cable shortfall: 8.4%
• Correlation between price and delivered power: 0.21
• Correlation between conductor csa and delivered power: 0.78
• Correlation between length and delivered power: -0.71
• Cables within 5% of rating: 22.2%
• Cables between 5% and 10% off: 47.6%
• Cables between 10% and 20% off: 25.4%
• Cables more than 20% off: 4.8%
• Mean price of cables in the top quartile by performance: £142
• Mean price of cables in the bottom quartile: £129

Interactive tools#

Four calculators built on the 2026 dataset and a searchable table of every figure on this page. Everything runs in the browser, nothing is stored, and no email address is required.

Each calculator draws its coefficients from the tables above rather than from a separate model, so at its reference settings each one reproduces a published table on this page exactly. Where a calculator makes an assumption, the footnote under it says which table the assumption comes from.

Methodology#

Every figure on this page comes from one of four EV Cable Hub studies conducted between January and June 2026: a monitored charging performance test, a survey of 2,140 UK drivers, a bench test programme covering all 63 cables, and aggregated order data.

Frequently asked questions#

Twenty-eight questions on real EV charging cable performance, each answered with the measured 2026 figure first.

Every answer below is drawn from the tables and measurements published on this page, and each one names the figure before it explains it.