The Real Cost of Broken Chargers: EV Charging Reliability Report 2026

Plug in at a public DC fast charger in 2026 and there's roughly a 1-in-8 chance you drive away without a charge — even though the network's own dashboard almost certainly told you the stall was "available." Run that failure rate through ChargeCostLab's cost model and unreliability quietly costs a public-charging-dependent driver about $280 and 9.6 hours a year; even a home-charging road-tripper loses close to $70 and 2.4 hours.

By Petra Halvorsen, Energy & E-Mobility Cost Analyst · Published 17 August 2026

---

This report synthesizes every major current US dataset on public charging reliability — J.D. Power's newly published 2026 study, ChargerHelp's telemetry analysis, UC Berkeley's field audit, Paren's live network monitoring and the federal and state uptime rules now in force — and then does the thing none of them do on their own: it converts failed sessions into dollars and hours per driver.

Reliability is also, right now, a moving target. J.D. Power's 2026 EVX Public Charging Study landed on 13 August 2026, four days before this report went to press, and it knocked Tesla Supercharger off the top of the DC fast charging leaderboard for the first time in five years. That single event forced a rewrite of the network-scorecard section below — proof, if any were needed, that "which network is most reliable" is not a settled question in this market.

Reported uptime vs. functional reality: the metric that's lying to you

"Uptime" is the number every network leads with, and it is also the one most likely to mislead you: a charger counts as "up" the moment its software reports itself online, whether or not a car can actually finish a charge on it. EVgo commits publicly to 98%+ network uptime; Electrify America's own California regulatory filings put its next-generation hardware at 97% [19][27]. Neither number tells you what happens when you tap your card.

Four independent ways of measuring the same population of chargers converge on the same conclusion: self-reported software status and driver-experienced outcomes are different things, and the gap between them is large. UC Berkeley's Rempel et al. tested 657 Bay Area CCS connectors in person and found only 72.5% functional — 22.7% failed outright on blank screens, payment errors or initiation faults, and another 4.9% had cables physically too short to reach the car [12]. ChargerHelp's 2025 Annual Reliability Report, built from over 100,000 sessions across 2,400+ chargers, found reported uptime running 98.7–99.9% against a first-time charge success rate of just 71% [9][10]. Paren's composite reliability score, which blends session outcomes, retries and downtime across more than 77,000 tracked DCFC ports, put the national average at 93.8 in Q2 2026 — better than the field-audit numbers, because it credits a station as "reliable" on a looser bar than "my car charged on the first try," but still nowhere near the 98%+ networks advertise [15][17].

Here's the honest read: none of these four numbers is wrong, exactly. They're measuring different things at different altitudes — software status, a driver's literal first attempt, a blended session-outcome score, a one-time physical audit — and a network can be telling the truth about "uptime" while still failing nearly three in ten drivers who show up expecting to charge.

The 2026 network scorecard: Tesla just got dethroned

Three OEM-backed networks — Ionna (807/1,000), the Mercedes-Benz Charging Network (797) and the Rivian Adventure Network (755) — beat every other DC fast charging operator in this year's satisfaction rankings, ending Tesla Supercharger's five-year run at #1 [1][2][3][4][5]. All three were newly award-eligible in 2026, and Ionna is itself backed by GM, BMW, Honda, Hyundai, Kia, Mercedes-Benz and Stellantis. Ionna's CEO, Seth Cutler, put it plainly: "Because we're not first to market, we've been able to learn from others, and we're determined to redefine public charging and quality" [3].

The 2026 study surveyed 6,594 BEV and PHEV owners between January and June, in partnership with PlugShare, and found DC fast charging satisfaction up 12 points year over year to a segment average of 666, while public Level 2 satisfaction fell 12 points to 595 on weaker payment and ease-of-charging scores [1][5]. Non-charging visits — J.D. Power's term for showing up and leaving without a charge — dropped to 12%, the lowest in the study's six-year history, down from 14% over the same half of 2025 [1][4]. Charger location matters more than most drivers assume: DC fast chargers at hotels (692), gas stations and convenience stores (689) and restaurants (688) all scored well above the segment average, while dealership-hosted chargers came in lowest at 570 — evidence that who operates and maintains the site around a charger shapes the experience almost as much as the hardware does [4].

What the 2026 release does not give you is a full network-by-network table. J.D. Power publicly disclosed only the three new award winners; Tesla, EVgo, Electrify America, ChargePoint and Blink's individual 2026 scores were not made public in the material available to us. So for a like-for-like comparison across the incumbents, the last fully disclosed scorecard remains 2025's:

Network	2025 DCFC satisfaction	Note
Tesla Supercharger	709	#1 for a 5th consecutive year — since dethroned in 2026
Red E	668	Strongest non-Tesla incumbent performer
Segment average	654	Down 10 points YoY in 2025, on cost dissatisfaction
ChargePoint	619
Electrify America	601
EVgo	579	Commits publicly to 98%+ network uptime
Blink	501	Lowest-ranked major DCFC network

Source: J.D. Power 2025 EVX Public Charging Study [6][7]. Read the full public network cost comparison for what each of these networks actually charges per kWh.

Tesla's dominance was never really about hardware sitting still — it was about vertical integration removing failure points (its own payment layer, its own back-end, plug-and-charge by default) that plague CCS networks stitched together from a charger vendor, a network operator and a card processor. Ionna, Mercedes-Benz and Rivian's 2026 scores suggest the newest OEM-backed entrants learned that lesson and applied it from day one, rather than retrofitting it onto an older fleet. It's a genuinely different competitive picture than the one that held from 2021 through 2025, and it is worth watching whether Ionna's numbers hold up once its network scales past its current early-stage footprint — a young network with fewer, newer stalls has an inherent reliability advantage over an older one, the same dynamic that inflates Tesla and Rivian's scores relative to Electrify America's more heavily built-out, harder-worked legacy fleet.

Context for scale: Tesla still operates 36,877 of the country's roughly 71,398 public DC fast ports as of April 2026 — 51.6%, more than the next nine networks combined — which is part of why its satisfaction numbers move markets even when they slip [35][36].

What the ChargeCostLab model says failed charging actually costs you

ChargeCostLab's model puts the cost of unreliable public charging at roughly $280 and 9.6 hours a year for a driver who depends on it entirely, and about $70 and 2.4 hours for a driver who mostly charges at home. No public dataset converts a failure rate into a dollar figure per driver, so this is ChargeCostLab's own estimate — every input below is either sourced or explicitly marked [ESTIMATE], and the output should be read as a floor, not a precise bill [38].

Sourced inputs:

• 12% of public charging visits ended with no charge in H1 2026 (J.D. Power) [1]
• 71% first-time charge success rate → roughly 29% of attempts need at least one retry (ChargerHelp) [9]
• 60% of failed visits in the 2025 study were caused by out-of-service equipment (J.D. Power) [6]
• Paren's Q2 2026 national average DC fast charging price is $0.538/kWh, up from the roughly $0.47/kWh figure that circulated through mid-2026 [15]

Assumptions [ESTIMATE]:

• Persona A — home charger plus road trips: 30 DCFC sessions a year
• Persona B — no home charging, apartment dweller: 120 DCFC sessions a year
• Retry friction (session fails once but the same stall works on a second attempt): 7 minutes, applied to the roughly 17% of sessions that need a retry but don't become a full failed visit (29% minus 12%)
• Full failed visit: 30 minutes average, covering troubleshooting, an app-support call and driving to another station
• Half of failed visits require relocating to a different station, adding 8 miles of detour at an all-in vehicle cost of $0.70/mile
• Value of time: $25/hour, between US DOT travel-time guidance and the median US wage

	Persona A (30 sessions/yr)	Persona B (120 sessions/yr)
Retry events/yr	5.1 (× 7 min ≈ 36 min)	20.4 (× 7 min ≈ 2.4 hrs)
Failed visits/yr	3.6 (× 30 min = 1.8 hrs)	14.4 (× 30 min = 7.2 hrs)
Total time lost	≈2.4 hrs/yr	≈9.6 hrs/yr
Time cost @ $25/hr	$60	$240
Detour miles cost	$10 (14 mi)	$40 (58 mi)
Total annual cost	≈$70/yr	≈$280/yr

The 12% failure rate is an improvement on the 14% the original 2025-vintage version of this model used, and it shows: both personas' totals fall by roughly 10% against last year's inputs, purely from measured reliability gains, before any change in assumptions. Framed differently, Persona B's roughly $280 in wasted time and miles sits on top of an annual public-charging spend that a driver taking 120 sessions a year at $0.538/kWh and a typical 35–45 kWh top-up would run to somewhere near $2,300–2,600 — call it a 10–12% hidden surcharge for unreliability alone, on top of a per-kWh rate that already runs nearly triple the US residential electricity average of about 18.5 cents/kWh [15][21][37]. Scaled across the millions of US EV drivers who lean on public charging, that plausibly adds up to hundreds of millions of dollars in wasted driver time nationally each year — a figure nobody publishes because nobody but a cost site has reason to calculate it [38].

The model deliberately excludes worst-case outcomes: a stranded battery requiring a tow, a missed flight or appointment, a hotel night burned waiting for a working charger. Read it as a floor.

Why chargers fail: five recurring causes, and a sixth that isn't failure at all

ChargerHelp's analysis of 19 million-plus data points found software and communication faults behind roughly two-thirds of documented charger failures, with more than 26% of stations carrying software inaccuracies — reporting "available" while unable to actually charge a car [11]. These faults cluster at the seams: between charger firmware, the network's back-end, and the vehicle's handshake protocol, none of which any single party fully controls.

Payment systems are the second-biggest culprit. UC Berkeley's Bay Area audit found payment failure among the most common causes of non-functional connectors, and J.D. Power's 2025 data showed payment-related satisfaction sliding even as raw reliability improved [6][12]. Card readers exposed to weather, cellular dead zones and app-authentication loops all register as "up" on a network dashboard regardless of whether a driver can actually pay.

Connectors and initiation errors are a close third. Berkeley documented blank screens, "connection error" messages and initiation failures on 22.7% of the connectors it tested, plus another 4.9% with cables physically too short to reach the car's inlet — a pure design failure that no amount of software patching fixes [12].

Aging is the most alarming pattern in ChargerHelp's data, because it runs directly against the industry's "we're improving" narrative: first-time charge success drops from 85% at newly installed stations to below 70% by a station's third year [9][10]. America's charging problem isn't just building more stalls — it's that stalls degrade without funded, sustained maintenance, and today's improving headline numbers partly reflect a young fleet rather than solved maintenance economics.

Concentration compounds all four. In 2024, just four networks held 25% of US ports but accounted for 75% of all down ports — reliability is heavily operator-dependent, not evenly distributed across the industry [11].

And then there's the failure mode that isn't a failure at all: congestion. A charger that's fully functional but occupied is unavailable to the next driver in exactly the way a broken one is, and it doesn't show up in any uptime statistic. One recombination of J.D. Power's 2025 data, from the charger-discovery app Parkopedia, estimates that up to 43% of public chargers are "effectively unavailable at any given moment" once you add roughly 25% genuinely broken to another 18% functioning but too congested to use in dense markets like San Francisco [29]. Parkopedia's response is an A-to-F reliability grade per location, scored dynamically on historical session-success rates rather than a static claim — a private-sector bet that drivers will trust a report card more than a network's own dashboard.

A structural issue sits underneath all six causes: no independent third party audits US network uptime claims. A network self-reporting 97–98% cannot be checked from the outside, which is exactly the gap the federal ChargeX Consortium — a two-year effort spanning Argonne National Laboratory, Idaho National Laboratory and NREL, working with more than 50 industry organizations — was stood up to close, publishing shared definitions for charging-experience KPIs, retry best practices and minimum error codes so failures can at least be measured the same way network to network [20][31].

Is it getting better? The four-year trend, and where the floor still sags

Yes, on the number that matters most to a driver standing at a stall: the share of US public charging visits ending without a charge has fallen every year since 2023, from 20% then to just 12% in the first half of 2026. That trend comes from J.D. Power's EVX Public Charging Study, which has now published four comparable readings: 20% in 2023, 19% in 2024, 14% in the first half of 2025, 12% in the first half of 2026 [1][6][8].

Paren's data tells a consistent, slower-moving story about the floor rather than the average: most states have climbed into a 90–95 reliability-score band, up from a broader 85–92 range a year earlier, driven largely by the sheer volume of newer, more reliable ports coming online — operators added 4,382 new DC fast ports across 806 stations in Q2 2026 alone, on top of 3,387 in Q1 [15][16][17]. But "most states" hides a real spread.

The gap between Washington, DC (97.7) and Oklahoma (78.5) is nearly 20 points on the same national scale — evidence that reliability tracks operator execution and market maturity, not some fixed national ceiling every state is bumping against [15]. A driver's actual experience still depends enormously on where they live and which network dominates their corridor.

Two countercurrents temper the good news. First, aging: ChargerHelp's 85%-to-under-70% three-year decay curve means every cohort of shiny new stations becomes next year's maintenance liability, so a rising fleet-wide average can coexist with individual stations quietly getting worse [9][10]. Second, a satisfaction paradox: even as measured reliability improved through 2025, J.D. Power's overall satisfaction score fell that year (DCFC down 10 points, L2 down 7), driven by rising prices — drivers were being asked to pay more for infrastructure that still failed on one visit in seven [6]. Whether 2026's satisfaction gains (DCFC +12, though L2 fell another 12) mark a durable turn or a one-year artifact of new OEM-backed networks entering the sample is a question next year's study will answer, not this one.

The policy layer: NEVI's rocky 2025–2026, and a new state-level standard

Federal rule 23 CFR Part 680 requires any charging port funded under the National Electric Vehicle Infrastructure program to average more than 97% uptime annually, calculated monthly, with narrow exclusions for vandalism and disasters [13][14]. It was, when finalized in 2023, the first enforceable US reliability standard of any kind. It's had a rough eighteen months.

On 6 February 2025, the Department of Transportation under Secretary Sean Duffy suspended approval of state NEVI deployment plans, effectively freezing the program. Seventeen states sued. In June 2025, a federal judge granted a preliminary injunction unfreezing funds for 14 of the plaintiff states; DOT issued new interim guidance that August, simplifying planning requirements [22]. On 30 January 2026, Judge Tana Lin issued final judgment in Washington v. U.S. Department of Transportation, ruling that the freeze violated the Administrative Procedure Act and calling it "arbitrary and capricious" — DOT then fully released obligated funds, clearing the way for states to proceed with solicitation rounds three through five [22].

That win was short-lived in dollar terms. The Consolidated Appropriations Act, 2026, enacted the following month, transferred roughly $503.8 million of unobligated NEVI formula funds to other highway programs, plus another $300 million from NEVI's competitive-grant set-aside and $75 million earmarked for the Joint Office of Energy and Transportation — about $879 million total redirected to programs including INFRA, Tribal Transportation and Reconnecting Communities [23][24]. As of mid-2026, states had obligated roughly $1.4 billion of the $4.4 billion still available for stations running through 2028, but only about 2% of that pool — some $94 million — had actually been spent, with roughly 121 NEVI-funded stations physically operational [22]. Read plainly: the 97% uptime rule remains fully in force for every port it covers, but Congress just shrank the number of ports it will ever cover, even as the litigation that unfroze the program was still being finalized in the courts.

Into that gap has stepped California. The state's Energy Commission adopted its own EV charger reliability rule in October 2025, effective 1 April 2026: any DC fast charger that received public incentive funding must maintain at least 97% average uptime, mirroring the federal NEVI floor but reporting hourly reliability data rather than an annual average, plus semiannual location and port-count disclosures [25][26]. Crucially, California's rule also mandates that network operators share availability and accessibility data with third-party software developers — the regulatory groundwork for exactly the kind of independent, app-based reliability check that Parkopedia and others are now building on top of J.D. Power's raw survey data [25][29]. It is the first state-level reliability mandate to extend meaningfully beyond NEVI's federal-corridor scope, and other states are watching it as a template.

For the federal side specifically, our NEVI state-by-state tracker follows deployment and funding status by state as it moves through 2026, and our explainer on whether NEVI-funded charging is actually cheaper covers the pricing side of the same program.

Reliability programs: what the networks are actually doing about it

EVgo's ReNew program, launched in 2022 and still the backbone of its public reliability commitments, is built on six pillars — prevention, diagnostics, rapid response, analysis, resilience and continuous customer service — aimed at halving average repair time, prioritizing spare-parts inventory for the highest-failure components (cables and connectors, overwhelmingly) and running a 24/7 bilingual support line. EVgo set a public target of exceeding 95% "One & Done" first-attempt success by the end of 2023 [19].

Electrify America's approach has been more hardware-driven: its next-generation cabinets, which the network has disclosed run near 97% uptime in its own California regulatory filings, generate roughly 80% fewer maintenance dispatches than the Gen 1 hardware installed in its 2017–2021 build-out. The catch is that the network-wide blended average, weighed down by that older legacy fleet, still sits in the 90–94% range rather than matching the newer cabinets' headline number — a live illustration of exactly the aging problem ChargerHelp's data describes at the industry level [9][27].

Beyond individual operators, the federally convened ChargeX Consortium is the closest thing the US has to an industry-wide reliability standards body: three DOE national laboratories working with more than 50 organizations across the charging industry, chartered to make "charge the first time, every time" an achievable operational target rather than an aspiration. Its published outputs so far include shared customer-experience KPI definitions, seamless-retry recommendations, a minimum set of standardized error codes, payment-system best practices, and an adapter-safety report — the un-glamorous plumbing work that has to happen before "reliability" can be measured consistently enough for a driver, a regulator or a competitor to hold any single network accountable [31].

Why unreliability is also an adoption problem, not just an inconvenience

Charging failures don't just cost the drivers who already own an EV — they're actively slowing down who buys the next one. NREL's own research on the topic states it bluntly: issues with public charging stations are consistently found among the top reasons potential buyers cite for not purchasing an EV, and the lab estimates roughly one in five public charging attempts still fails in some form [30].

AAA's most recent consumer survey found only 16% of US adults "likely" or "very likely" to buy an EV next — the lowest reading since 2019 — with 56% citing a lack of convenient charging stations and 55% citing range anxiety among their reasons for hesitating [28]. McKinsey's Mobility Consumer Pulse survey, covering 36,000 respondents across 15 countries, found 46% of current US EV owners saying they're considering switching back to a gas or hybrid vehicle, and the single most-cited US reason — ahead of cost, ahead of performance — was simply "cannot charge at home," with public charging friction close behind [32]. S&P Global Mobility's own 2025 Electrification Consumer Survey, run across roughly 8,000 respondents in eight countries, reached the same conclusion from a different angle: charging infrastructure is simultaneously the biggest driver of confidence in EVs and the single most persistent constraint on adoption, depending entirely on whether a given driver's experience of it has been good or bad [33].

Put together, these three surveys describe a market where the hardware failures documented earlier in this report aren't a side issue for people who already made the leap — they're a headwind for the wider EV transition, at a moment when Cox Automotive already logged EV market share sliding to 5.8% of new US vehicle sales in the first quarter of 2026, down from a 10.6% peak the prior autumn, following the September 2025 expiration of the federal EV tax credit [34]. Fix the charging experience and you don't just save existing owners $280 a year — you remove one of the two or three reasons cross-shopping buyers give for staying with gas.

How to protect yourself from a failed charging session

The levers available to an individual driver are limited, but they're real, and they map directly onto the failure modes above. First, treat a stuck handshake as a software problem before a hardware one: most networks' apps have a remote-restart or refresh function, and it clears the majority of first-attempt failures within the roughly 7-minute window ChargeCostLab's model accounts for separately from a full failed visit. Second, if a restart doesn't work, try a different stall at the same site before you drive anywhere — ChargerHelp and Berkeley's data both point to failures clustering on individual connectors rather than whole stations going down together, so the stall next to a broken one is very often fine [9][12].

Third, favor sites with more stalls and, where you have the choice, the networks and locations that score better in the data above: hotel-, gas-station- and restaurant-hosted DC fast chargers outperformed the segment average in 2026, while standalone dealership sites lagged [4]. Fourth, on a genuinely important trip, build in the buffer this report's own model assumes exists — a spare 30 minutes and a mental note of the next-nearest station — rather than arriving at 5% state of charge with no backup plan. None of this fixes the underlying reliability gap the policy and hardware sections above describe, but it does convert a stranding risk into, at worst, the kind of retry friction the cost model already prices in. Our detailed walkthrough covers what to do stall-by-stall, including which networks' support lines actually pick up.

---

Methodology and assumptions

Scope. What US public EV charging reliability data actually show in 2026 — reported uptime versus measured charge-success rates, network-by-network satisfaction, the federal and state policy layer, and what unreliable charging costs a driver in dollars and hours.

Reliability data. Four independent measurement approaches are cited throughout and are not directly interchangeable: self-reported network uptime, J.D. Power's driver-reported non-charging-visit rate [1][6][8], ChargerHelp's first-time charge success rate from charger telemetry [9][10], and Paren's composite reliability score across 95%+ of tracked US DCFC ports [15][16].

Cost model. Section 3's per-driver cost of unreliability is ChargeCostLab's own estimate; every input is either cited or labelled [ESTIMATE], and the outputs are directional floors, not precise bills [38].

Flagged uncertainty. J.D. Power's 2026 study publicly disclosed scores only for its three new OEM-backed award winners [1][2][3][4][5]; the full network scorecard here uses the last fully disclosed table, from 2025 [6][7]. Paren's Q2 2026 "national average" figure is cited via trade coverage that read the underlying report, while state-level and utilization figures are pulled directly from Paren's own published data [15][17][18]. Electrify America's Recurrent Auto-derived figures are cited via a secondary aggregation [27].

---

Frequently asked questions

How often do public EV chargers actually fail in the US in 2026? In the first half of 2026, 12% of public charging visits ended without a successful charge, down from 14% over the same period in 2025 and 20% in 2023, per J.D. Power's EVX Public Charging Study [1]. ChargerHelp's telemetry data separately puts first-time charge success at 71%, meaning close to three in ten attempts need a retry even when the visit eventually succeeds [9].

Which EV charging network is most reliable in 2026? J.D. Power's 2026 study found three OEM-backed networks — Ionna (807/1,000), the Mercedes-Benz Charging Network (797) and the Rivian Adventure Network (755) — outperforming the field, unseating Tesla Supercharger from the top spot it held for five straight years [1][2]. Tesla's exact 2026 score wasn't publicly disclosed; the last full scorecard, from 2025, still ranked Tesla first at 709 [6].

How much does charger unreliability cost EV drivers? ChargeCostLab's model estimates about $70 and 2.4 hours a year for a driver who charges mainly at home and fast-charges on road trips, rising to roughly $280 and 9.6 hours a year for a driver who depends entirely on public charging [38].

Why do networks claim 98% uptime if chargers fail so often? Uptime counts a charger as available whenever its software reports online, without verifying a car can actually start and finish a charge. Payment failures, handshake errors and technically-live-but-broken connectors usually don't register as downtime, and no independent third party audits network uptime claims in the US [20].

What is the NEVI 97% uptime requirement, and is it still in force? Federal rule 23 CFR Part 680 requires NEVI-funded ports to average more than 97% uptime annually [13]. A 2025 funding freeze was struck down in court in January 2026, but the February 2026 Consolidated Appropriations Act then redirected about $879 million of unobligated NEVI money elsewhere [22][23].

Is EV charging reliability actually improving? Yes on the headline trend — J.D. Power's non-charging-visit rate has fallen every year since 2023 [1][6][8] — but ChargerHelp's data shows first-time success dropping from 85% at new stations to under 70% by year three, so some of the improvement reflects a young fleet rather than solved maintenance economics [9][10].

What should I do when a public charger won't start a session? Try the app's remote-restart function first — most first-attempt failures clear on retry. If that doesn't work, move to a different stall at the same site before driving elsewhere, since failures tend to cluster on individual connectors rather than whole stations [9][12].

---

About the author

Petra Halvorsen is ChargeCostLab's Energy & E-Mobility Cost Analyst. She analyses retail power markets and electric-vehicle running costs, reconciling regulator data, charging-operator tariffs and real-world consumption into figures drivers can act on. She takes no payment from carmakers, charging networks or energy suppliers, and every calculation here is reproducible from the cited primary sources.

---

Sources

1. J.D. Power — 2026 U.S. Electric Vehicle Experience (EVX) Public Charging Study. https://www.jdpower.com/business/press-releases/2026-u-s-electric-vehicle-experience-evx-public-charging-study/
2. Electrek — IONNA just beat every major EV fast-charging network. https://electrek.co/2026/08/13/ionna-just-beat-every-major-ev-fast-charging-network/
3. GM Authority — IONNA Dominates J.D. Power 2026 U.S. EV Experience Public Charging Study. https://gmauthority.com/blog/2026/08/ionna-dominates-j-d-power-2026-u-s-ev-experience-public-charging-study/
4. CBT News — Public charging satisfaction rises 12 points, JD Power finds. https://www.cbtnews.com/public-charging-satisfaction-improves/
5. InsideEVs — Study: DC Fast Charging Satisfaction Is Up, And Ionna Takes The Top Spot. https://insideevs.com/news/804743/ev-charging-satifsfaction-2026-jdpower/
6. J.D. Power — 2025 U.S. Electric Vehicle Experience (EVX) Public Charging Study. https://www.jdpower.com/business/press-releases/2025-us-electric-vehicle-experience-evx-public-charging-study
7. EV Charging Stations — J.D. Power: EV Chargers Becoming More Reliable With Fewer Failed Charging Attempts. https://evchargingstations.com/chargingnews/j-d-power-ev-chargers-becoming-more-reliable/
8. J.D. Power — 2023 U.S. Electric Vehicle Experience (EVX) Public Charging Study. https://www.jdpower.com/business/press-releases/2023-us-electric-vehicle-experience-evx-public-charging-study
9. ChargerHelp — ChargerHelp Report Reveals Charge Success Rate and Not Uptime More Accurate Metric for EV Driver Experience. https://www.chargerhelp.com/post/news-and-events/chargerhelp-report-reveals-charge-success-rate-and-not-uptime-more-accurate-metric-for-ev-driver-experience/
10. ChargerHelp — 2025 Annual Reliability Report. https://www.chargerhelp.com/2025-annual-reliability-report/
11. Electrek — The real story behind EV charger-reported uptime vs. actual uptime. https://electrek.co/2024/08/05/ev-charger-reported-uptime-vs-actual-uptime/
12. Rempel, Cullen, Bryan, Cezar — Reliability of Open Public Electric Vehicle Direct Current Fast Chargers (arXiv / UC Berkeley / Cool the Earth). https://arxiv.org/abs/2203.16372
13. eCFR — 23 CFR Part 680, National Electric Vehicle Infrastructure Standards and Requirements. https://www.ecfr.gov/current/title-23/chapter-I/subchapter-G/part-680
14. Federal Register — National Electric Vehicle Infrastructure Standards and Requirements (final rule). https://www.federalregister.gov/documents/2023/02/28/2023-03500/national-electric-vehicle-infrastructure-standards-and-requirements
15. Paren — US EV Fast Charging — Q2 2026. https://www.paren.app/reports/us-ev-fast-charging-q2-2026
16. Paren — US EV Fast Charging — Q1 2026. https://www.paren.app/reports/us-ev-fast-charging-q1-2026
17. Electrek — US EV fast charging has entered "Charging 2.0" — here's what that means. https://electrek.co/2026/07/14/us-ev-charging-has-entered-charging-2-0-heres-what-that-means/
18. EV Charging Stations — Paren: US Added 800+ New DC Fast-Charging Sites in Q2 2026. https://evchargingstations.com/chargingnews/paren-dc-fast-charging-q2-2026/
19. EVgo — EVgo Advances Network Enhancements through EVgo ReNew Program, Releases Best Practices to Promote Greater Industry-wide Charger Reliability. https://www.evgo.com/press-release/evgo-advances-network-enhancements-through-evgo-renewtm-program-releases-best-practices-to-promote-greater-industry-wide-charger-reliability/
20. U.S. News & World Report — Racing to Fix America's Troubled EV Charging Network. https://cars.usnews.com/cars-trucks/features/ev-public-charging-reliability
21. EV Charging Stations — How Much Does It Cost to DC Fast Charge an Electric Car? https://evchargingstations.com/chargingnews/cost-dc-fast-charge-ev/
22. Eno Center for Transportation — Washington v. U.S. Department of Transportation and NEVI Progress Updates. https://enotrans.org/article/washington-v-u-s-department-of-transportation-and-nevi-progress-updates/
23. GCN — States obligated over $500 million to put fast chargers along every major highway corridor, and then Congress quietly cut half a billion more. https://gcn.com/states-obligated-over-500-million-put/20671
24. Congressional Research Service (Congress.gov) — Implementation of Electric Vehicle Charging Infrastructure Programs: CFI and NEVI. https://www.congress.gov/crs-product/R48996
25. California Energy Commission — EV Charger Data and Reliability Standards. https://www.energy.ca.gov/programs-and-topics/programs/clean-transportation-program/ev-charger-data-and-reliability-standards
26. California Energy Transition — California Sets Standards for EV Charger Reliability. https://www.californiaenergytransition.com/p/california-sets-standards-for-ev
27. Axis Intelligence — Electrify America Reliability 2026: Real Uptime Data, Failure Modes, and What the Numbers Hide. https://axis-intelligence.com/electrify-america-reliability-uptime-data/
28. AAA Newsroom — Americans Slow to Adopt Electric Vehicles Despite Widespread Availability. https://newsroom.acg.aaa.com/aaa-americans-slow-to-adopt-electric-vehicles-despite-widespread-availability/
29. The Weekly Driver — Charging Anxiety: Why Your Next EV May Grade Chargers A to F. https://theweeklydriver.com/2026/05/ev-charger-reliability-charging-anxiety-parkopedia/
30. National Renewable Energy Laboratory (NREL) — Impact of Electric Vehicle Charging Station Reliability, Resilience, and Location on Electric Vehicle Adoption. https://docs.nrel.gov/docs/fy24osti/89896.pdf
31. Joint Office of Energy and Transportation — National Charging Experience (ChargeX) Consortium. https://driveelectric.gov/chargex-consortium
32. McKinsey & Company — New twists in the electric-vehicle transition: A consumer perspective. https://www.mckinsey.com/features/mckinsey-center-for-future-mobility/our-insights/new-twists-in-the-electric-vehicle-transition-a-consumer-perspective
33. S&P Global Mobility — EV charging reality check: Readiness and consumer insights. https://www.spglobal.com/automotive-insights/en/blogs/2026/03/ev-charging-readiness-and-consumer-insights
34. Cox Automotive (Kelley Blue Book) — Q1 2026 EV sales report commentary. https://www.coxautoinc.com/insights/q1-2026-ev-sales-report-commentary/
35. US DOE Alternative Fuels Data Center — Electric vehicle charging stations: levels and US port shares. https://afdc.energy.gov/fuels/electricity-stations
36. EV Charging Stations (citing US DOE AFDC) — Largest DC fast-charging networks in the US, April 2026. https://evchargingstations.com/chargingnews/largest-dc-fast-charging-april-2026/
37. US Energy Information Administration — Electric Power Monthly, Table 5.3. https://www.eia.gov/electricity/monthly/epm_table_grapher.php?t=table_5_03
38. ChargeCostLab cost-of-unreliability model (this article, Section 3) — [ESTIMATE]. Original calculation from the sourced inputs above; not a third-party source.

---

© 2026 ChargeCostLab. Independent EV running-cost analysis. Figures reflect data available as of 17 August 2026 and will change as networks, prices and federal policy move. Informational, not financial advice.