What the evidence shows
The fixed April 2024 DOE release reported 1.5% battery replacements due to failure in an approximately 15,000-vehicle plug-in sample, excluding recalls. That dated snapshot challenges routine early replacement; it is not today’s US-wide lifetime failure rate.
Older vehicles or current generations?
The replacement snapshot spans model years 2011–2023, but has no battery specification or equal follow-up by age. It cannot isolate newer-pack performance from older vehicles or recall history.
We separate vehicles from before 2023 and 2023 onward. A report’s publication date does not establish a vehicle’s model year or a battery’s manufacturing date.
Mixed vehicle generationsUS plug-in-vehicle replacements due to failure
- Vehicles / sample
- About 15,000 US plug-in vehicles, including battery-electric cars and plug-in hybrids; recall replacements are excluded from the headline rate.
- Vehicle years
- 2011–2023, as reported by DOE; exposure and follow-up vary by model year, with the newest cohort least observed.
- Battery chemistry
- Not reported by DOE/Recurrent.
- Battery capacity
- Not reported.
- Battery manufacture
- Pack production dates not reported; model year must not be treated as pack build date.
- Observation period
- A cross-sectional sample summarized in April 2024; DOE does not specify the sample's full follow-up window.
Model or system evidenceLaboratory dynamic-cycling comparison
- What was measured
- 92 commercial EV-energy lithium-ion cells tested under 47 discharge protocols; these are cells, not complete production vehicles.
- Vehicle years
- Not applicable; the study does not use vehicle model years.
- Battery chemistry
- Silicon-oxide/graphite anode with nickel-cobalt-aluminium (NCA) cathode.
- Battery capacity
- Cell capacity not specified in the main study text.
- Battery manufacture
- Manufacture dates for the commercial test cells are not reported.
- Observation period
- More than two years of laboratory ageing experiments at 35 °C; publication date is not the cell production date.
“Not reported” means the source does not disclose it. Model year, first registration, vehicle assembly and battery manufacture are different dates. Unmatched studies cannot establish how much newer batteries improved.
Which brands do these results describe?
DOE reports plug-in battery replacements in aggregate, including BEVs and PHEVs, and excludes recall replacements. No make or model breakdown is published in the cited evidence, so a manufacturer comparison or brand attribution is unavailable.
All manufacturers are OEMs. These groups describe brand focus, not a quality ranking or country of origin. EV-focused brands can also sell plug-in hybrids. Results apply to the identified models, batteries and conditions.
Manufacturer and model sources (1)
Replacement is not the same as aging
A battery can gradually store less energy while continuing to power the car. Replacement due to failure is a separate event. A recall is different again: a manufacturer may replace a pack because of a safety defect even when that pack has not failed in everyday use. Combining these outcomes into one alarming number prevents readers from understanding what the statistic actually measures.
A report about an expensive repair is meaningful evidence about that owner’s experience. To make a claim about what normally happens, however, it needs to be placed against the number of comparable vehicles and the years they have been observed. A bill establishes cost; it does not establish frequency.
What the 2024 US snapshot showed
The Department of Energy’s April 2024 release summarized a Recurrent study of approximately 15,000 plug-in vehicles. It reported 1.5% replacements due to failure outside recalls, with model years 2016–2023 below 1%. Both battery electric cars and plug-in hybrids were included. This is a dated report of a sample, not a continuously updated estimate of all cars on US roads. [1]
The denominator and follow-up matter
Imagine one sample dominated by newly sold cars and another dominated by ten-year-old cars. Even if the batteries had identical aging behavior, the younger sample would have had fewer years in which to fail. That is why model-year percentages should be accompanied by age and follow-up. Likewise, excluding recalled packs answers a useful normal-use question but cannot describe the total number of replacement procedures owners experienced.
Another distinction concerns the observation unit. Counting owners who report a replacement differs from counting replacement procedures: one vehicle could undergo more than one procedure. A study may also observe only cars that remain in its monitoring service. Before treating a percentage as a population probability, check how vehicles entered the sample, how missing histories were handled and what qualified as a replacement. These are questions for assessing evidence quality, not allegations that the cited study is invalid.
What newer research adds
A peer-reviewed study published in December 2024 compared realistic dynamic discharge profiles with constant-current laboratory cycling. It found that the test pattern affects the inferred lifetime. That supports taking measurement methods seriously; it does not establish a guaranteed number of years for a production vehicle. NHTSA similarly describes battery lifespan as dependent on several factors rather than a fixed promise applicable to every pack. [2][3]
What this means for a used-car decision
Ask for a battery-health assessment, service and recall history, the remaining written warranty and a repair estimate for the specific vehicle. Compare those findings with the distance you need to travel. A pack retaining useful range may remain adequate even after measurable degradation. Conversely, a rare failure can still be expensive for the owner who experiences it. Both statements can be true without accepting the claim that frequent replacement is normal.
Warranty terms give useful context, but they are not failure-rate studies. The IEA’s May 2026 report describes typical EV battery warranties as guaranteeing 70% capacity retention after 8–10 years or 160,000 km. US Department of Energy guidance says comprehensive battery-failure data are not available and many US manufacturers offer 8-year/100,000-mile coverage; Australia’s national energy department says most EVs there have an 8-year/160,000-km battery warranty. Coverage thresholds, transferability and exclusions vary, so check the actual vehicle’s terms. [4][5][6]
Data period: Recurrent sample published March 2023 covers model years 2011–2023 and is summarized by DOE in 2024; battery-warranty guidance published or reviewed in 2026.
What this does—and doesn’t—tell us
- Recall replacements are excluded; the sample is not all US vehicles.
- Younger cohorts have shorter follow-up. These percentages are not a lifetime failure probability or an ICE comparison.
Routine replacement after a few years is not supported by the cited sample. A dated replacement rate is not a lifetime failure probability.
Found something that needs checking?
Identify the statement, explain your concern, and share original sources for our review.
Keep the conversation curious.
Questions and corrections are welcome. Link your sources, challenge ideas, and be kind to people.
Discussion actions are unavailable until sign-in is configured.
You can still read published comments and their sources.
Loading discussion…