Your First Entry-Level E-Bike (2026): 10 Decisions

51 jurisdictions checked · 750 W federal line · 39 claims tested · 20 picks across 6 brands · verified August 24, 2026

Quick Answer: Buy for the trip you actually make. The average American commute is 13.4 miles each way, and a 26.8-mile round trip needs roughly 205–557 watt-hours if the motor does all the work. A 720 Wh battery covers that with room to spare, which means the 100-mile range on the box is answering a question you will never ask.

Then check three things the spec sheet buries: what the bike weighs when you have to lift it, what class it is where you live, and whether its battery carries a real certification. Start your entry level electric bike search from commuter and city or the full range. Everything else is preference.

The trap almost everyone falls into is treating advertised range as a plan. It is a ceiling measured under best-case conditions, and on our own storefront only 39 of 184 e-bikes publish enough for you to check it at all.

This page is one half of a pair. It starts from your body, your weather, and your trip; the spec-by-spec process (motor types, brakes, warranty terms) lives in our electric bike buying guide.

The rider pictured on this page is Cole Mercer, part of our fictional cast — meet the family and see how they are made. All cast photography is generated with Playcut.

How we verified this guide

Every number below was retrieved from a primary source during the week of August 18–23, 2026, and the retrieval is recorded. We read the federal e-bike definition at the Electronic Code of Federal Regulations, the injury and fatality data in the U.S. Consumer Product Safety Commission's own April 2026 report, the battery rulemaking in the Federal Register, and national trip-length data published by the Federal Highway Administration.

Corpus denominator. Eleven distinct sources anchor this guide and twenty-nine citations draw on them. Nine were retrieved and read in full. Two — the IEC 60529 standard text and the Journal of Applied Biomechanics article — are paywalled at the publisher, so we cite the publisher's own scope statement and a university-hosted copy respectively, and say so here rather than implying we hold both in full.

Where the questions came from. The questions this page answers were harvested rather than imagined: 4,149 observations across four independent sources — Google and Bing autocomplete, related searches, Semrush's question report, and the headings of pages currently ranking — yielding 3,909 distinct phrasings, of which 14 are answered here as FAQs and three themes were deliberately dropped (servicing, theft and exercise, each named with its reason in the page's demand ledger). One correction that came straight out of that harvest: "first electric bike" autocompletes mostly to India-market and history of the first e-bike queries, not to buying advice, which is the sort of thing an assumed keyword list never catches.

What is ours. Three findings are our own measurements rather than anyone's published claim: a count of all 51 US jurisdictions' e-bike frameworks, a sweep of all 184 e-bikes on our own storefront, and a physics model of range built on a peer-reviewed, experimentally validated equation. Each is labeled where it appears, and the scripts that produced them are named so the arithmetic can be re-run.

Source register

Every source this page rests on, with how much weight it can carry. T1 is a primary official publication, T2 an organization's own publication about itself, T3 peer-reviewed or standards-body work usable for context. We cite no T4 source alone and no T5 source at all.

Reviewed by the Redtail eBikes editorial desk · last verified August 24, 2026 · Corrections: hello@redtailebikes.com.

What's on this page

• Your first e-bike starts with four numbers
• Distance: the range math, done honestly
• What size battery you actually need
• Weight and storage: the spec nobody publishes
• Hills: what a grade demands, in watts
• Terrain: pavement, gravel, sand, singletrack
• Off-road: where you may actually ride it
• Weather: rain, and the rating that answers it
• Seasons: what cold actually costs you
• Age and body: what the injury data says
• Location: the class you buy is a map
• Class 1, 2 and 3, translated
• Matching a machine to a trip
• Price: what each band actually buys
• Battery safety: the 2026 federal rule
• Where buyers get caught out
• 20 entry level electric bikes, matched to 20 first-time riders
• The 12-point pre-purchase checklist
• Your first 30 days: the shakedown checklist
• The honest ledger: where an e-bike wins and loses
• Limitations and when this page expires
• Frequently asked questions
• The bottom line
• Primary sources

Your first e-bike starts with four numbers

Something changed in the e-bike market this summer. On June 24, 2026 the Consumer Product Safety Commission published a proposed rule at 91 FR 38162 that would make battery safety certification mandatory rather than voluntary, and the comment period closes August 24, 2026. Buy before the rule lands and you are buying under the old regime, where certification is a choice the manufacturer makes.

What you lose by choosing wrong is not usually money. It is use.

A bike too heavy to carry up your steps gets ridden twice. A Class 3 bought for speed can be barred from the path you meant to ride it on. A battery sized for a marketing number leaves you pedaling a 70-pound bicycle home.

So the calculation this guide walks you through starts with four numbers you already know, before any spec sheet is opened:

1. Your round trip in miles — the real one, on the day you actually ride.
2. The steepest hill on it, and roughly how long it lasts.
3. Where the bike sleeps — garage, hallway, third-floor apartment, balcony.
4. Your state, because the class you buy means different things in different places.

Nothing about that framing is new, but the regulatory backdrop is. The federal definition has stood essentially unchanged since 16 CFR § 1512.2 was amended in 2003 and again in 2011. What is moving in 2026 is the safety regime around the battery.

Enforcement sits with the Consumer Product Safety Commission, which regulates e-bikes as consumer products rather than as vehicles. The unsettled part is whether the states will follow. The federal 750 W line and the state three-class system already disagree in ten jurisdictions, and no authority is reconciling them.

Every section below turns one of those four into a specification. Nothing here asks you to trust a brand.

Distance: the range math, done honestly

Advertised range is not a lie, but it is not a plan. It is the distance achievable under conditions chosen to flatter the bike: minimum assist, flat ground, a light rider, no wind, a full charge, mild weather. The useful question is not "how far can it go" but "how many watt-hours does a mile cost me", because that number is physics and you can check it.

We built the answer from a published model rather than from anyone's blog. Martin and colleagues, writing in the Journal of Applied Biomechanics in 1998, derived and then experimentally validated an equation for road cycling power; their modeled values tracked meter-measured power at R² = 0.97 across six riders and six velocity bouts (Martin et al. 1998, 14(3):276–291). Three of its terms dominate steady riding:

• Aerodynamic drag rises with the cube of speed, which is why 20 mph costs so much more than 12.
• Rolling resistance rises with weight, and with fat low-pressure tires.
• Gravity is weight times gradient, and on a hill it swamps everything else.

Run those terms for a 250-pound combined load at a steady 15 mph on flat ground, and a mile costs 7.7 to 20.8 watt-hours from the battery if the motor is doing all the work, on Martin's Equations 6a, 9a and 15. The band is wide because two inputs are honest estimates rather than measurements, and we would rather show you the spread than invent a single confident number.

No agency polices range claims. We could not locate a federal test procedure for e-bike range comparable to the EPA's for cars, and we label that as our failure to find one rather than as proof that none exists. The statute giving CPSC authority over e-bikes runs to product safety, and its current rulemaking addresses batteries rather than performance claims.

What is certain is that manufacturers pick their own conditions, which is why the numbers are not comparable between brands. Whether a standardized test would even help is genuinely arguable: one test cycle would make claims comparable while still telling you nothing about your hill, your weight or your assist level. Until one exists, the arithmetic below is the only check a buyer has.

What we measured and what we assumed

That table is the part most guides leave out. Two of five inputs are ours rather than anyone's laboratory, and every figure below inherits that uncertainty — the two that are measured come from Martin et al..

What a mile actually costs, by weight and speed

Battery watt-hours per mile on flat ground, motor doing all the work:

Read the 20 mph column against the 12 mph column. Riding eight miles an hour faster roughly doubles the cost of every mile, because the drag term rises with the cube of velocity, and no battery upgrade buys back what speed spends.

The measurement we ran on our own catalog

Divide a bike's battery capacity by its claimed range and you recover the watt-hours per mile that claim quietly assumes, which is then checkable against the validated model. We ran that division across every e-bike on our own storefront on August 23, 2026 using scripts/build_first_bike_dataset.py. Two findings, both about our own shelves:

First, most bikes do not let you check. Of 184 e-bikes, only 39 publish both a battery capacity and a range claim. On the other 145 the arithmetic is simply unavailable to a buyer.

Second, the 39 that do publish both imply 5.0 to 31.2 watt-hours per mile, with a median of 14.4. The claims implying 5 to 10 Wh/mi sit at or beyond the optimistic edge of what the physics allows for a motor working alone — which is not fraud. It means the rider is assumed to be pedaling hard.

Each dot is one bike's own published claim, divided out. The two red dots on the left are claiming less energy per mile than a motor working alone can physically use.

One honest caveat about our own number: those 39 bikes are not a random sample. 23 are Eahora, 12 Velotric, 4 Himiway — a concentration that reflects which brands publish both figures rather than which brands are optimistic — and none of them is among the 20 we name below, each screened against 16 CFR § 1512.2. Read it as a finding about disclosure rather than a league table.

What size battery you actually need

Size the pack against your round trip, then add a third. The 2022 National Household Travel Survey puts the average American commute at 13.4 miles one way and the average trip of any purpose at 12.6 miles (FHWA/ORNL, Table 3-3). The survey counts e-bikes inside its "Bicycle" mode, which it defines as covering "bicycles of all speeds and sizes, including electric bikes", so the trip lengths are the right yardstick.

A 26.8-mile round trip — twice the 13.4-mile average American commute — needs 205 to 557 watt-hours with the motor doing everything, or 103 to 278 Wh if you pedal half the work. Set against the packs on the market:

The red line is the average American round-trip commute. Every pack on the chart clears it, even at the pessimistic end, even with the motor doing all the work.

Assumptions: 250 lb combined, 15 mph steady, flat, no wind, mild temperature. Hills, cold and headwinds all move these numbers down, and the sections below quantify by how much.

The practical reading is reassuring rather than alarming. A 720 Wh battery covers the average American round-trip commute at the pessimistic end of the band, with roughly 30% left. Buying 1,440 Wh for a 27-mile day buys weight and cost you will carry every single ride to solve a problem you do not have.

Two caveats on treating the NHTS figure as your figure. The 2022 survey measured trip distance by shortest-path routing rather than by asking respondents, a method change from surveys before 2017 that makes older comparisons unreliable.

The 2022 average trip length also rose to 12.6 miles from 10.7 in 2017, which the report attributes partly to people dropping short trips during the pandemic years. Whether that shift persists is unresolved. Your own measured round trip beats the national average every time.

Weight and storage: the spec nobody publishes

The heaviest thing about an e-bike is the decision to buy a heavy one. A bike you cannot lift is a bike that lives wherever you last left it, and for apartment dwellers that single fact decides whether the purchase works.

Our own sweep found something unflattering to the industry: of 184 e-bikes on our storefront, bike weight was published for only 28. Payload capacity is published far more often, because it is a bigger, better-sounding number, and neither figure appears in the federal definition that decides what the machine legally is. The two get confused constantly, and they are opposites: payload is what the bike carries, weight is what carries you.

Among the 28 that do publish it, the range runs 28.6 to 150 pounds, with a median of 80.7 pounds and 17 of 28 above 70 pounds. That weight is why the CPSC's 2024 study found a quarter of e-bike injuries happening on uphill or downhill ground, where a heavy machine is hardest to hold. An e-bike is not a bicycle with a motor added; it is a different object with different logistics.

What weight actually decides

If the answer to any of the last three rows is yes, start from the folding range rather than filtering a full-size catalog down. The removable battery is the underrated feature in this table. It turns one 80-pound carry into a 70-pound carry plus a separate 10-pound object, and it lets you charge indoors while the bike stays in the garage — which is the safer arrangement anyway.

Nobody enforces weight disclosure. Unlike motor wattage, which most state definitions test, vehicle weight appears in almost no US e-bike statute; Pennsylvania's 100-pound ceiling is the conspicuous exception among the states we read.

A manufacturer omitting the figure therefore breaks no rule, and the practice has if anything grown as bikes have got heavier. The proposed battery standard addresses electrical safety and says nothing about publishing mass, so this gap stays open.

Hills: what a grade demands, in watts

A hill converts your weight directly into watts, and the conversion is unforgiving. Gravity's share of the power equation is simply mass times gravity times gradient times speed, which means it does not care about your motor's marketing. It cares what you weigh and how steep the road is.

Running the same validated model for a sustained 8 mph climb:

The final column is the honest core of it. At a 6% grade and 250 pounds combined, 239 watts of the requirement is pure gravity — mass times gravity times gradient times speed, unavoidable by any tire, any bearing, any frame. Everything a manufacturer can engineer only affects the remainder.

The physics here is settled and old. Martin's gradient term is the potential-energy equation, and it has not changed since 1998 because it cannot.

The marketing convention built on top of it is another matter. No US authority defines how a manufacturer must measure "nominal" versus "peak" wattage or torque, so two bikes advertising 750 W may deliver materially different sustained output. The federal definition assumes a rated figure exists without saying how it is derived.

Why torque matters more than watts here

Watts describe how fast work gets done. Torque describes how hard the motor pulls at low speed, which is the condition a hill creates.

A 500 W motor rated at 80 N·m will start a loaded bike up a steep driveway more convincingly than a 750 W motor rated at 45 N·m, and both sit under the 750 W federal ceiling. Starting from near-standstill on a grade is a torque problem before it is a power problem.

This is also where mid-drive and hub motors genuinely diverge, because a mid-drive multiplies its torque through the bike's gears. We treat that comparison properly in the eBike motor guide, which measures 26 drive systems rather than asserting a winner.

A grade you can feel

Most riders badly misjudge gradient. A 3% grade is a long gentle rise you would barely name as a hill. A 6% is a noticeable climb that a fit rider on an unassisted bike works at, and a 10% is steep enough that walking becomes a live option.

If your commute has a sustained 10% section, buy for that number rather than the average, because the gravity term scales linearly with gradient.

Terrain: pavement, gravel, sand, singletrack

Tire width is a bigger decision than motor size, and it is the one most first-time buyers get backwards. fat tire electric bike is the single highest-volume search in this entire category, which means a great many people are choosing four-inch tires before they have asked what surface they ride on.

Fat tires buy flotation on loose ground and comfort over broken pavement. They cost rolling resistance every mile — our model uses an assumed 0.0140 for a low-pressure fat tire against a measured 0.0032 for a high-pressure road clincher, and that difference is most of the gap between the optimistic and pessimistic columns in every table on this page.

For loose surfaces and real trail riding the category to look at is mountain and off-road; for sand and snow it is fat tire. The honest summary: fat tires are a specialist tool that has been marketed as a default. If you ride pavement, a 2.4-inch tire will be faster, lighter, longer-ranged and more comfortable than a four-inch tire at the same price. If you ride sand or snow, nothing else works.

Fat tires arrived in the US market as a snow-and-sand specialism and became a default through fashion rather than any change in the engineering. No standard governs how tire width is advertised, and no agency tests rolling resistance for bicycles, so the trade-off is invisible at the point of sale.

The unsettled question is our own assumed coefficient. We use 0.0140 for a low-pressure fat tire against Martin's measured 0.0032 for a road clincher, having found no published measurement for a modern fat tire at a citable source. Treat the fat-tire column of every table here as our estimate.

Off-road: where you may actually ride it

Buying an off-road-capable e-bike and being allowed to ride it off-road are two different purchases. Capability is a specification. Access is a rule, set by whoever controls the land, and it varies between a national forest and the county park two miles away.

Federal land, state land, county land and city land each answer separately, and a bike welcome on a rail trail may be excluded from natural-surface singletrack in the same county. Nothing in 16 CFR § 1512.2 speaks to land access at all; it is a product-safety definition. Our state-by-state work on this lives in the 50-state eBike Trail Atlas, with the two deepest treatments at California trails and Texas trails.

There is a pattern worth knowing before you buy. In the 51-jurisdiction sweep behind this guide, natural-surface non-motorized trails are regulated separately from paved paths in state after state — the statute that grants your e-bike access to a bike path frequently says nothing at all about dirt — Arizona § 28-819 is typical in regulating natural-surface trails separately. That silence is usually resolved by the land manager rather than the legislature.

Trail access is enforced by the land manager rather than the police, and the penalty is usually ejection or a citation under the managing agency's own regulations. The rules moved sharply on 29 August 2019, when the Interior Secretary signed Order No. 3376.

That order directed the National Park Service, Bureau of Land Management, Fish and Wildlife Service and Bureau of Reclamation to define e-bikes consistent with 15 U.S.C. § 2085 and exempt them from their off-road and motor-vehicle definitions. Its operative sentence cuts both ways: "E-bikes shall not be allowed where other types of bicycles are prohibited."

States and counties have been revising their positions ever since, in both directions. The contested edge is Class 1 access to natural-surface singletrack, where land managers and access organizations still openly disagree, and where the answer differs between two adjacent parcels.

The question to ask before you buy for trails

Name the specific trail. Then find who manages it and read their posted rule. A guide cannot do this for you, because the answer is parcel-by-parcel, and a national article that claims otherwise is guessing.

Weather: rain, and the rating that answers it

Nearly every e-bike sold in the US is rain-capable and almost none are waterproof, and the difference is a two-digit code. After price, rain is the single most-asked question in this category — we counted roughly 1,100 monthly US searches across twelve phrasings of "can you ride an electric bike in the rain", which is more than "how to choose an electric bike" attracts by a factor of four.

The standard that answers it is IEC 60529, "Degrees of protection provided by enclosures (IP Code)", edition 2.2, published 29 August 2013 (IEC). It classifies enclosures for electrical equipment rated up to 72.5 kV, and it produces the IP number on your bike's spec sheet.

Drizzle is not the hazard; the pavement is. Braking distance grows on wet asphalt long before anything electrical notices the weather.

That last row is the trap. IPX4 does not mean the bike scored zero on dust; it means dust protection was never evaluated under IEC 60529. A marketing page that prints "IPX4 waterproof" is misusing two words at once.

IEC 60529 has carried its current structure since its 1989 second edition, amended in 1999 and again in 2013, so the code itself is stable. Its weakness for a buyer is that it is voluntary here: no US rule requires an e-bike to carry an IP rating, nobody audits the claim, and a manufacturer may simply not test.

The CPSC's proposed battery rule adds no ingress requirement, so an unrated bike stays perfectly lawful to sell. Reasonable people also differ on whether a rating is worth much on a machine whose most exposed component, the charge port, is one the rider opens by hand.

Rain is a traction and visibility problem long before it is an electrical one. Nothing here is unusual for a rated bike.

What rain actually does to an e-bike

Riding a rated bike in rain is fine. The failure modes are elsewhere: standing water deep enough to reach the hub motor's seals, pressure-washing the bike afterwards, and charging a wet connector. Wipe the charge port dry before plugging in, and never pressure-wash any e-bike regardless of its rating, since IEC 60529 rates jets only to the level its digits describe.

The rain risk that shows up in the injury data is not electrical at all. In the CPSC's 2024 e-bike special study, 10% of injured riders reported that visibility problems may have contributed, and of the injuries where it was dark or hard to see, fewer than half were using lights or reflective gear (CPSC, April 2026). Wet weather is a seeing-and-being-seen problem before it is a water-ingress problem.

Seasons: what cold actually costs you

Cold does not damage a lithium battery in ordinary use; it temporarily hides part of it. Capacity returns when the pack warms. Knowing roughly how much disappears is what stops a winter commute becoming a walk.

A 2025 controlled study in Machines ran charge–discharge cycles on 18650, 21700 and 26650 cells across two commercial chemistries at +25 °C and at −10, −15, −20, −25 and −30 °C (Machines 13(12):1114). Averaged across that sub-zero sweep, the measured capacity drops were:

The 21700 is the format in most current e-bike packs, and it holds up best of the three. None of this is permanent: the capacity comes back when the pack warms.

Two cautions on reading that table. These are averages across the whole −10 °C to −30 °C sweep rather than the loss at any single temperature, and they are cell-level laboratory results rather than whole-bike measurements. Treat them as the shape of the effect rather than a number to promise anyone.

The mechanism matters for one practical rule. The same paper notes that lithium plating during cold charging is a safety-critical constraint, and that charge-transfer resistance rises by a factor of two to four between 25 °C and −20 °C. Charging a battery that is genuinely cold is the one winter habit worth breaking.

Cold-weather derating is not a rule anyone enforces. It is a property of the chemistry, and no US regulation requires a manufacturer to state it. The understanding firmed up over the last decade as instrumented studies replaced anecdote, and the 2025 Machines study is part of that trend.

Two things remain genuinely open. The published figures are cell-level, and how a whole pack behaves inside an insulated enclosure on a moving bike has not been measured at any source we could retrieve. The studies also disagree on chemistry: the same paper found two chemistries in one 21700 format differing by nearly six percentage points.

The pack comes inside and warms up before it goes on the charger. That one habit is most of winter battery care.

The four winter rules that follow from the data

Rules one and two in a single frame: the pack lives indoors between winter rides, and it charges only after it has warmed.

1. Bring the battery inside between rides. It stores better and charges safely at room temperature.
2. Let a cold pack warm before charging it. An hour indoors is enough.
3. Plan for roughly a quarter less range in deep cold, and more on an older pack.
4. Charge before storing rather than straight after a cold ride, and store at partial charge over a long layup.

If you ride through real winters, the Minnesota and Colorado law pages cover two states where winter riding is normal and the rules are written accordingly.

Age and body: what the injury data says

The stereotype of the e-bike rider is young, and the federal data says otherwise. In the 2022 NHTS, 31% of US bicycle riders were 55 or older and 39% were 35 to 54 (FHWA/ORNL). This is a middle-aged and older activity, and the equipment decisions should follow that rather than fight it.

Frank Kowalski, 63, of our fictional cast, on the setup this section keeps arriving at: a low step-over, an upright posture, and no hurry.

The CPSC's April 2026 report is the authoritative US source on what goes wrong. Across 2017–2024 it records 310 e-bike fatalities out of 533 across all micromobility products, and an estimated 155,200 emergency-department visits (CPSC, Tables 1.1–1.3).

Read by age, the pattern is stark. Of the 281 e-bike fatalities where age was recorded, 160 — 57% — were riders 45 or older. Riders 65 and older account for 75 of those 281 deaths, or 26.7%, against a 17% share of the US population. And among all micromobility deaths of people 65 and older, 75 of 92 involved an e-bike specifically, which is 82%.

The CPSC compiles these figures under its statutory injury-surveillance mandate, and its own caution matters: reporting for 2023 and 2024 is still open and counts may rise in future editions.

The trend line is steep, with e-bike fatalities rising from zero in 2017 to 97 in 2024. So did the number of e-bikes on the road, and the report publishes no exposure denominator. Without miles ridden or bikes in use, these counts cannot tell you whether riding became more dangerous or simply more common. Anyone citing them as a rate is overreading them, ourselves included if we did.

The gold ticks are what each age band would account for if deaths tracked population. Riders 45 and over sit above their share; riders under 25 sit well below it.

A step-through frame removes the leg-swing over a high top tube — the moment where low-speed falls actually happen.

What that does and does not mean

It does not mean older riders should not ride, and 31% of US bicycle riders are already 55 or over. It means the failure modes shift, and so should the bike. The same report's 2024 special study found 25% of e-bike injuries occurred on unlevel — uphill or downhill — surfaces, 38% of injured riders were carrying or holding something, and only 40% were wearing a helmet.

Those three findings point at the same equipment answers: a frame you can get on and off without swinging a leg over a high top tube, a rack so cargo is on the bike rather than in your hand, and stability at low speed where mounting and dismounting happen.

Location: the class you buy is a map

Which class you buy decides where you may legally ride, and the answer changes at state lines. We counted this rather than estimating it. Across all 51 US jurisdictions — 50 states and the District of Columbia — using our own statutory sweep retrieved 2026-08-19:

• 41 of 51 use the three-class system.
• 10 do not, and their alternatives differ from each other as much as from the norm.

That second group is where first-time buyers get hurt, because a national guide that says "check your class" is useless in a state with no classes. Each of the ten was read at its own legislature's text:

If you live in one of those ten, a product page's "Class 2" badge is a manufacturer's description rather than a legal status where you ride. The only definition binding nationwide is the product-safety one at 16 CFR § 1512.2.

Where a Class 3 costs you path access

The most expensive consequence of buying the fastest class is losing the shared-use paths. In our sweep, at least 17 of 51 jurisdictions keep a Class 3 off shared-use or bicycle paths by default, or let a local authority bar it outright while Class 1 and Class 2 get notice-and-hearing protection first.

The statutory language is consistent enough to quote. Arizona § 28-819(D)–(E) keeps a Class 3 off paths unless it is "within or adjacent to a roadway". Connecticut § 14-289k bars one from a bicycle or multiuse trail outright.

Oklahoma 47 O.S. § 11-1209(E)(2) uses Arizona's wording almost exactly. Washington RCW 46.61.710 excludes Class 3 from shared-use paths and sidewalks by default, making wider access a local opt-in. Vermont 23 V.S.A. § 1136a lets an authority bar Class 3 outright while Class 1 and 2 require notice and a safety finding first.

Being precise about the denominator matters here: 10 jurisdictions have no Class 3 category to restrict, and for 24 the statutory text we read did not address Class 3 on paths at all. Those 24 are unresolved by this dataset rather than cleared. We are not going to convert silence into permission.

A correction, made while checking this page's own links on 23 August 2026. Our earlier sweep recorded Oklahoma as a state that pointedly did not single Class 3 out, and we had written that here. Opening 47 O.S. § 11-1209(E)(2) showed the opposite: "A Class 3 electric-assisted bicycle shall not be ridden on a bicycle or multiuse path unless it is within or adjacent to a highway or roadway or unless the local authority or state agency having jurisdiction over the path permits that operation." Oklahoma belongs in the restricted column, which moves the count from 16 to 17 and leaves no jurisdiction in the "explicitly not restricted" column at all. The error came from trusting a summary line instead of the section.

For the jurisdiction-level detail, our 50-state e-bike law hub carries the full table, with deeper pages for California, New York, Texas, Florida, Pennsylvania, Washington, New Jersey, Connecticut and Minnesota.

The three-class model is a spreading convention rather than settled national law, and it is still being rewritten where it exists. By August 2019 the Interior Department could say only that "a majority of States have essentially followed" the federal definition "in some form".

California's own definition at Veh. Code § 312.5 was last amended by Stats. 2024, Ch. 791 (SB 1271), effective 1 January 2025. That churn is why ten jurisdictions have not arrived at the model and why New Jersey has moved away from it.

State and local police enforce these rules, and the practical penalty is usually a citation. Our own coverage is the contested part: for 24 of 51 jurisdictions we could not resolve the Class 3 path question from the statutory text we read, and we would rather publish that gap than round it away.

Class 1, 2 and 3, translated

The federal government defines an e-bike once, for product-safety purposes, and the states define it again for road rules. Confusing the two is the most common error in this entire subject.

The federal definition sits at 16 CFR § 1512.2(a)(2), which makes a low-speed electric bicycle a "bicycle" for Consumer Product Safety Commission purposes. Its operative words:

"A two- or three-wheeled vehicle with fully operable pedals and an electric motor of less than 750 watts (1 h.p.), whose maximum speed on a paved level surface, when powered solely by such a motor while ridden by an operator who weighs 170 pounds, is less than 20 mph."

That language traces to Public Law 107-319, adding 15 U.S.C. § 2085, and the regulation carries its amendment history on its face: 43 FR 60034 (Dec. 22, 1978), amended at 68 FR 7073 (Feb. 12, 2003) and 76 FR 27888 (May 13, 2011).

The three classes are a state overlay on top of that, and where a state uses them:

Two things are worth knowing about the age of this definition. The 750 W and 20 mph figures were set by Congress in Public Law 107-319, signed 4 December 2002, and written into the regulation at 68 FR 7073 in February 2003. They reflect the market as it was more than twenty years ago, and nothing in the current rulemaking proposes to revisit them.

Enforcement is split. The CPSC polices whether a product may be sold as a bicycle, while the states police how it may be ridden. That split is the contested ground: a machine can be lawfully sold and still be unlawful to ride on the road in your state.

The trap: wattage and class are independent. A 1,000 W motor limited to 20 mph is not made a Class 2 e-bike by that limiter, because most state definitions test the motor rating as well as the speed. Our eBike motor guide works through that distinction across 26 real drive systems.

Matching a machine to a trip

The category names on a product page describe frames rather than uses, and buying the wrong category is how people end up with a bike that does nothing well. Start from the trip.

Two categories deserve a warning for first-time buyers. Moped-style bikes look like the most bike for the money and are frequently rated well above the 750 W federal ceiling, which puts them outside the federal definition and, in many states, outside the e-bike rules entirely. Dirt-bike-style machines usually have no pedals at all, which excludes them from the federal definition by its plainest term: "fully operable pedals".

Neither is a bad machine, but a 3,000 W dirt bike with no pedals is not an e-bike under any US definition, and in several states a moped-style machine of that description is licensed as a moped. Both are the wrong first purchase for someone who has not yet checked what their state permits.

Price: what each band actually buys

Below about $800 the compromises stop being cosmetic. Above about $2,500 you are buying refinement rather than capability. The middle is where most first bikes should sit, and knowing what each band buys stops you overpaying and underbuying at once.

If budget is the binding constraint, our current reductions are the honest place to start, because a discounted bike from the middle band usually beats a full-price bike from the bottom one.

The cost that surprises people is not the bike. It is the replacement battery, which commonly runs several hundred dollars and is the most expensive service item an e-bike has. Check that a replacement pack for your specific model can actually be ordered, because a bike whose battery cannot be replaced has a fixed lifespan.

Prices in this category have been unusually unstable, moving with tariffs, freight costs and a shakeout among direct-to-consumer brands. Any band stated here is a snapshot rather than a rule.

No authority regulates e-bike pricing, and the only price-adjacent enforcement is the Federal Trade Commission's general jurisdiction over deceptive advertising. Whether the sub-$1,000 tier is a durable segment or a temporary product of oversupply is genuinely disputed, and we do not know the answer.

Our best place to buy electric bikes guide compares the purchase channels honestly, including the cases where a local shop beats us on service.

Battery safety: the 2026 federal rule

Battery fires are real, rare, and concentrated in a specific and avoidable pattern. The honest numbers matter here, because this topic attracts more fear than evidence.

Of the 310 e-bike fatalities the CPSC recorded from 2017 through 2024, 19 were associated with lithium-ion battery fires. That is about 6%. Collisions with motor vehicles caused 170 — roughly 55% (CPSC, Table 1.4).

Within those 19 fire deaths, the concentration is the useful part: 7 involved homemade battery packs or bike repair shops, and 2 involved flame eruptions while charging. Uncertified, rebuilt and modified packs are doing a large share of the damage.

This is the fastest-moving subject on the page. Voluntary certification became a de facto expectation after the CPSC called publicly on manufacturers, retailers, importers and distributors to comply with the UL standards on 20 December 2022.

New York City then made it law, requiring third-party certification for e-bikes sold in the city from September 2023 (UL Solutions). The federal rule now proposed would generalize that. The contested question is scope: the proposal reaches chargers, conversion kits and aftermarket packs, and where the line falls for bikes already in garages is unsettled.

What is changing, and when

On June 24, 2026, the CPSC published a notice of proposed rulemaking at 91 FR 38162 proposing a new 16 CFR part 1265. Proposed § 1265.2(a) would require that "each eBike must comply with all provisions of UL 2849-20 that apply to the product", and proposed § 1265.1(c) sets a 180-day effective date after any final rule. Written comments are due August 24, 2026.

The Commission's own assessment is worth quoting because it cuts against the marketing: CPSC found that the performance requirements in UL 2849-20, UL 2272-24 and UL 2271-23 "are inadequate to eliminate or adequately reduce the unreasonable risks of injury" on their own, and proposes modifications including tamper-resistant enclosure requirements and post-discharge charge testing.

What UL 2849 does and does not cover

UL 2849 is the Standard for Electrical Systems for eBikes, and UL Solutions describes its scope as examining "e-bikes' electrical drive train system, battery system and charger system combinations" (UL Solutions). The related standards are UL 2271 for the battery pack and UL 2272 for other personal e-mobility devices.

The limitation is stated by UL itself, and almost never repeated by anyone selling bikes: "The Standard does not evaluate for the operator's ability to maintain control while riding." UL 2849 is a fire and electrical standard. It is not a statement that the bike handles well or brakes adequately.

Where buyers get caught out

Every trap below is one we have watched people walk into. They are collected here rather than scattered, so you can read them in one pass before you spend anything.

Treating advertised range as a plan. It is a best-case ceiling. Divide watt-hours by claimed miles and see what the claim assumes.

Buying peak watts. Manufacturers quote peak because it is bigger. Laws and physics both work on the nominal rating.

Assuming a speed limiter fixes the class. In most states the motor's rated wattage is part of the definition, so a 1,000 W bike capped at 20 mph is still a 1,000 W bike.

Confusing payload with weight. A 400 lb payload figure tells you nothing about whether you can lift the bike.

Buying fat tires for pavement. They cost range, weight and speed, and buy you nothing on tarmac.

Buying Class 3 for the speed and losing the paths. In at least 16 jurisdictions that is a direct trade.

Reading "IPX4" as a waterproof rating. The X means untested rather than passed.

Assuming trail access follows road legality. Different authority, different rule, decided parcel by parcel.

Forgetting the battery is a consumable. Check a replacement is orderable for your model before you buy.

Charging a frozen battery. Let it reach room temperature first.

Skipping the helmet. Only 40% of injured e-bike riders in the CPSC's 2024 study were wearing one.

Buying a moped-style machine as a first e-bike without checking whether the state treats it as an e-bike at all.

20 entry level electric bikes, matched to 20 first-time riders

There is no single best first e-bike, and any guide that names one is guessing about your life. There is a best bike for a 6-mile flat commute with a third-floor walk-up, and a different one for a hilly 20-mile round trip with a child on the back. Below are twenty machines we stock, each matched to the situation it actually suits.

Read the two tiers before the bikes. Everything in Tier 1 is rated at or under 750 W and held to 20 mph, which keeps it inside the federal definition and inside the class system in the 41 jurisdictions that use one. Everything in Tier 2 is faster or more powerful. Tier 2 bikes are not worse machines, and several are far more machine for the money. They do narrow where you may legally ride, and on a first bike that is usually the wrong trade.

The whole set at a glance

Brand spread, stated plainly: nine Eunorau, four Samebike, three Qiolor, two Eahora, one Ridstar, one Riding'times. We stock no first-bike-appropriate Vtuvia. Two brands we do sell, Himiway and Velotric, are excluded from every recommendation on this site.

Tier 1: the twelve we would hand a genuine beginner

These stay inside 20 mph and inside the class system. If you are new, start here, and leave only when a specific need pushes you out.

Tier 2: more machine, and the trade that comes with it

These eight are rated above 750 W, run faster than 20 mph, or both. Each is honestly the better object for the money, and each narrows where you may legally ride. Buy from this tier once you have read your own state's rule and know the answer.

What we did not do. We did not fill this list from one brand: six brands appear and none supplies more than nine of the twenty. We did not put a Himiway or a Velotric on it. And we did not quietly promote Tier 2 bikes into Tier 1 to make the list look stronger, because the split is the advice.

The 12-point pre-purchase checklist

Work down this list with the product page open. Anything you cannot answer is a question for the seller, in writing, before you pay.

1. Round trip measured on a map, never estimated from memory.
2. Battery watt-hours stated — volts times amp-hours if the page does not print Wh.
3. Implied watt-hours per mile calculated from their own claim.
4. Nominal motor rating in watts, distinguished from peak.
5. Torque in newton-meters, if you have hills.
6. Vehicle weight with battery, in writing.
7. Removable battery, yes or no.
8. Class, and whether your state uses classes.
9. Path access for that class where you ride.
10. UL 2849 or UL 2271 certification, for this model.
11. Replacement battery confirmed orderable, with its price.
12. Return window and warranty term, read rather than assumed.

Your first 30 days: the shakedown checklist

Day one is not a long ride. Find an empty car park and spend twenty minutes at low speed: starting, stopping, and turning tight circles. An e-bike carries more weight and more speed than a bicycle, and the muscle memory is genuinely different.

Week one: brakes and tires. Practise hard braking from 15 mph on a clear surface until you know how the bike behaves. Check tire pressure, because it changes rolling resistance more than any other thing you control.

Week two: your real route. Ride the actual commute once with no time pressure. Note the battery percentage at the turnaround point, which is the only range measurement that matters.

Week three: load it. Ride it as you actually will, with the rack and bag and whatever you carry — if the load turns out to be the point, a cargo frame carries it better than panniers on a commuter. Recall that 38% of injured riders in the CPSC's 2024 study were holding something.

Week four: decide within the return window. Most returns are possible for a limited period, and a bike that is not working for you after four honest weeks will not start working in month three.

The honest ledger: where an e-bike wins and loses

An honest ledger belongs in any guide that also sells the thing.

Where an e-bike wins. It flattens hills and headwinds, which is what stops most people cycling. It converts a 13-mile commute from an athletic event into transport. It carries loads a bicycle cannot. It parks anywhere and costs pennies per charge. It gets people riding who had stopped.

Where it genuinely loses. It is heavy, and that weight shows up every time you lift, carry, rack or store it. Its battery is a consumable with a finite life and a serious replacement cost. It is a theft target and needs a lock proportional to its value. Servicing is harder than for a bicycle and fewer shops do it well. In deep cold it loses a meaningful share of its range. And it is exposed in traffic, which the fatality data states plainly: motor-vehicle collisions caused 170 of 310 recorded e-bike deaths.

Who should not buy one yet. Anyone whose only trip is under a mile, where a bicycle is simpler and cheaper. Anyone with nowhere secure to store it. Anyone buying a moped-style machine before checking whether their state treats it as an e-bike, a moped or a motor vehicle.

Limitations and when this page expires

Stating these is the difference between a guide and an advertisement.

Author and interest. Redtail eBikes sells e-bikes, including all 20 named above. That is a commercial interest and you should weigh it. We have tried to earn trust by publishing the method, disclosing that most of our own catalog fails the screen this guide applies, and printing the implied watt-hours per mile of our own picks even where it flatters them.

Data limitations.

1. Our range model uses two assumed inputs, drag area and fat-tire rolling resistance. Both are labeled at the point of use and both are why the tables show bands.
2. The model omits two terms from Martin et al.'s Equation 15 — wheel-bearing friction and kinetic-energy change — which are small at steady speed but not zero.
3. Our catalog sweep covers our own storefront only, 184 e-bikes, and the 39 with checkable claims skew to three brands. It is not a market-wide survey.
4. The cold-weather figures are cell-level laboratory averages across a −10 °C to −30 °C sweep rather than whole-bike measurements at a stated temperature.
5. The class-3 path finding is "at least 17 of 51." For 24 jurisdictions the statutory text we read did not address the question, and we have not counted that silence as permission. One of the 17, Oklahoma, was misclassified in our own sweep until 23 August 2026 and is corrected on this page.
6. CPSC fatality reporting for 2023–2024 is ongoing and the Commission states counts may change in future reports.
7. Bike weight was published for only 28 of 184 models, so the weight distribution rests on a small and self-selected subset.

Expiry conditions. This page needs re-checking when any of these happens:

• The CPSC issues a final rule on 16 CFR part 1265, whose comment period closed 2026-08-24; a final rule would carry a 180-day effective date.
• NEISS 2025 data is finalised, expected by the CPSC in spring 2026, which will revise the injury estimates.
• Any state changes its e-bike framework, as New Jersey did in January 2026.
• The next NHTS wave replaces the 2022 trip-length figures.

Frequently asked questions

How do I choose an electric bike as a first-time buyer?

Start with your real round-trip distance and the steepest hill on it, then where the bike will be stored and which state you ride in. Those four answers determine battery size, torque, weight and class — the last set by your own state rather than by the federal definition. Only then compare models. Choosing from spec sheets first is how people buy a bike that does not fit the life they actually live.

How far can an electric bike really go on one charge?

Divide the battery's watt-hours by roughly 8 to 21 watt-hours per mile for a 250 lb combined load at 15 mph on flat ground, motor doing all the work. A 720 Wh pack gives roughly 35 to 94 miles on that basis, and considerably more if you pedal. Advertised ranges assume best-case conditions and are a ceiling rather than a plan.

How much do electric bikes weigh?

Among the 28 models on our storefront that publish a weight, the range is 28.6 to 150 pounds with a median of 80.7, and 17 of the 28 exceed 70 pounds. Most product pages publish payload capacity instead, which is a different and larger number, and neither appears in the federal definition. Ask the seller for vehicle weight with the battery installed.

Can you ride an electric bike in the rain?

Yes, if it carries an IP rating from IEC 60529 — IPX4 covers splashing from any direction and is the common e-bike rating. Avoid standing water deep enough to reach the motor, never pressure-wash the bike, and dry the charge port before plugging in. The bigger wet-weather risk is visibility: CPSC's 2024 study found fewer than half of riders injured in poor light were using lights or reflective gear.

What class of e-bike should I buy?

Class 1 or Class 2 for most first-time buyers, because both assist to 20 mph and generally keep the widest path access. Class 3 assists to 28 mph but in at least 17 of 51 US jurisdictions it is kept off shared-use paths by default or may be barred outright. Check your own state first, because 10 jurisdictions do not use the class system at all, and the federal rule sets no classes either.

Do I need a license or insurance for an electric bike?

In most US jurisdictions, no — e-bikes meeting the state definition are generally exempt from driver licensing, registration and insurance requirements. The exemption depends on the bike actually meeting that definition, which is where high-wattage and moped-style machines fail — most state definitions borrow the federal 750 W and 20 mph limits.

The standing exception is New Jersey: under P.L.2025, c.285, signed January 19, 2026 with its grace period ended July 19, 2026, every e-bike there needs a license and registration, and throttle-equipped ones need insurance — the full rules live on our New Jersey page. Our 50-state law hub carries the rest of the jurisdiction-level detail.

How long does an e-bike battery last?

Lithium packs are consumables measured in charge cycles rather than years, and capacity also falls temporarily in cold before recovering. Be suspicious of pages promising a hard number: even Bosch, the largest drive-system maker, publishes no cycle count — its own battery guide says a pack "also ages over time, even when not in use" and names the real drivers: heat, time, and storage habits (it recommends 32–68 °F at roughly 30–60% charge for long storage).

The more useful question before buying is whether a replacement pack for your specific model can be ordered and what it costs, because that determines the bike's usable lifespan. Confirm it in writing before you buy.

How much does a good first electric bike cost?

Most good first e-bikes sit between $800 and $2,500. Below about $800 the compromises stop being cosmetic: smaller packs, mechanical brakes, and batteries whose certification you must check hardest. The $800–$1,500 volume band buys 500–750 W, 480–720 Wh and increasingly hydraulic brakes, while $1,500–$2,500 adds torque sensing and published certifications.

Budget for the surprise item too: a replacement battery commonly runs several hundred dollars. And treat any band as a snapshot, because tariffs and a brand shakeout keep prices moving.

Are cheap electric bikes under $800 worth it?

Sometimes, if the compromises are ones you can live with and the battery carries a real certification. Below roughly $800 the savings usually come from brake type, battery size and component quality. The one place to spend regardless is battery certification, since uncertified and rebuilt packs feature disproportionately in the CPSC's fire-fatality records.

What size electric bike do I need?

Match your height to the maker's published rider-height range, not to a wheel size. If a listing does not state one, treat it the way this guide treats every undisclosed spec: as a no. Then check the stand-over height against your inseam, leaving about an inch of clearance so you can plant both feet at a stop.

And if swinging a leg over the frame is the hard part, choose a step-through, which the fit table above recommends for that exact situation.

Do I need a fat-tire electric bike?

Only if you ride sand, snow or genuinely soft ground. Fat tires cost rolling resistance, weight and range on every paved mile, and on pavement a 2.4-inch tire will be faster, lighter and longer-ranged at the same price, because rolling resistance scales with the coefficient their low pressure raises. It is the most over-bought feature in the category.

How much torque do I need for hills?

At a 6% grade with a 250 lb combined load, gravity alone demands about 239 watts at 8 mph, and the total requirement is roughly 263 to 314 watts. Torque matters more than wattage for starting on a grade, so treat 60 N·m as a sensible floor for hilly terrain and read the nominal rather than peak rating, since the federal test is the rated figure.

Does cold weather damage an e-bike battery?

Ordinary cold temporarily reduces available capacity rather than damaging the pack, and it returns when the battery warms. A 2025 controlled study measured average capacity drops of 22.4% and 28.1% for 21700 cells across a −10 °C to −30 °C sweep. Charging a genuinely cold battery is the real risk, because lithium plating during cold charging is a safety-critical constraint.

Are electric bikes safe for older riders?

Riders 55 and over are 31% of US bicycle riders, so this is a mainstream activity rather than an exception. The CPSC data shows the risk profile shifts with age: 57% of e-bike fatalities with recorded ages were riders 45 or older. The equipment answers are a step-through frame or three wheels, a rack so nothing is carried by hand, and Class 1 or a limited Class 2.

What is UL 2849 and does my bike need it?

UL 2849 is the Standard for Electrical Systems for eBikes, covering the drive train, battery and charger as a combined system. A CPSC proposed rule published June 2026 would make compliance mandatory. It is a fire and electrical standard only: UL states it "does not evaluate for the operator's ability to maintain control while riding."

Should my first e-bike have a throttle?

A throttle is genuinely useful for pulling away at junctions, on steep driveways and when your legs are tired, which is why Class 2 suits many first-time buyers. It also consumes markedly more battery than pedal-assist, so a throttle-heavy rider should size the pack against the pessimistic end of our tables. Class 1 and Class 2 access rules are usually identical.

The bottom line

Buy for the trip you actually make. For most first-time buyers in the United States that is a round trip well under 30 miles on pavement, which a 500–750 Wh battery covers comfortably, on a Class 1 or Class 2 bike weighing what you can actually lift. The 100-mile range and the 2,000-watt motor are answers to questions you have not asked, and they cost you weight, money and in many places path access.

Who should buy now. Anyone with a commute between two and fifteen miles each way, anyone who stopped cycling because of hills, and anyone replacing car trips for errands. The economics and the enjoyment both work.

Who should wait. Anyone without secure storage, anyone whose only trip is under a mile, and anyone drawn to a moped-style machine who has not yet read their own state's definition.

What to do today, with no pressure to buy. Measure your round trip, then filter the full range by the battery size that trip actually needs. Read your state's rule on our 50-state law hub. If a test ride would settle it, our store directory lists verified shops in Portland, NYC, Los Angeles, San Diego and Denver — including shops that are not us. Our refund policy is published in full, and you can reach a person at hello@redtailebikes.com before you spend anything.

You do not need to decide quickly. You need to decide with the right four numbers in front of you, and now you have them.

Image credit: all photography of the Mercers and Kowalskis on this page is generated with Playcut (playcut.ai), using identity-locked references from its custom-actor pipeline for face consistency. The data charts are our own, built from the sources below.

Primary sources

Retrieved and read during 2026-08-18 to 2026-08-23. Full bibliographic records, including archive snapshots, are in this article's citations.csv.

1. U.S. Consumer Product Safety Commission. "Micromobility Products-Related Deaths, Injuries, and Hazard Patterns 2017–2024." April 2026. Tables 1.1–1.4; § III-B. Accessed 2026-08-22. https://www.cpsc.gov/s3fs-public/Micromobility_Products-Related_Deaths_Injuries_and_Hazard_Patterns_2017-2024.pdf
2. U.S. Consumer Product Safety Commission. "Safety Standard for Lithium-Ion Batteries Used in Micromobility Products and Electrical Systems of Micromobility Products Containing Such Batteries." Notice of Proposed Rulemaking, 91 FR 38162, 24 June 2026. Proposed §§ 1265.1(c), 1265.2(a). Comments close 2026-08-24. Accessed 2026-08-22. https://www.federalregister.gov/documents/2026/06/24/2026-12749/safety-standard-for-lithium-ion-batteries-used-in-micromobility-products-and-electrical-systems-of
3. Office of the Federal Register / National Archives. Electronic Code of Federal Regulations, Title 16 § 1512.2, "Definitions." Paragraph (a)(2). Amended 43 FR 60034 (1978-12-22); 68 FR 7073 (2003-02-12); 76 FR 27888 (2011-05-13). Accessed 2026-08-22. https://www.ecfr.gov/current/title-16/part-1512/section-1512.2
4. United States Congress. Public Law 107-319, "An Act to amend the Consumer Product Safety Act to provide that low-speed electric bicycles are consumer products subject to such Act," H.R. 727, 116 Stat. 2776, signed 4 December 2002. Adding Consumer Product Safety Act § 38, codified at 15 U.S.C. § 2085; subsection (b). Accessed 2026-08-23. https://www.govinfo.gov/content/pkg/PLAW-107publ319/html/PLAW-107publ319.htm
5. U.S. Department of the Interior. Secretary's Order No. 3376, "Increasing Recreational Opportunities through the use of Electric Bikes." Secs. 4(a), 4(c), 5. Signed 29 August 2019. Accessed 2026-08-23. https://www.doi.gov/sites/default/files/elips/documents/so_3376_-_increasing_recreational_opportunities_through_the_use_of_electric_bikes_-508_0.pdf
6. Oklahoma Legislature. Oklahoma Statutes, Title 47 — Motor Vehicles. § 11-1209(E)(2), "Electric-assisted bicycles and operators – Rights and restrictions – Label requirements"; added by Laws 2019, c. 43, § 4, effective 1 November 2019. Read in the Oklahoma Senate's hosted Title 47 compilation. Accessed 2026-08-23. https://oksenate.gov/sites/default/files/2019-12/os47.pdf
7. California Legislature. California Vehicle Code § 312.5, "Electric bicycle." Subsections (a)(1)-(a)(4); as amended by Stats. 2024, Ch. 791 (SB 1271), effective 1 January 2025. Accessed 2026-08-23. https://leginfo.legislature.ca.gov/faces/codes_displaySection.xhtml?lawCode=VEH&sectionNum=312.5
8. Federal Highway Administration and Oak Ridge National Laboratory. "Summary of Travel Trends: 2022 National Household Travel Survey." Table 3-3; mode definitions. Accessed 2026-08-22. https://nhts.ornl.gov/assets/2022/pub/2022_NHTS_Summary_Travel_Trends.pdf
9. Martin, J.C., Milliken, D.L., Cobb, J.E., McFadden, K.L. and Coggan, A.R. "Validation of a Mathematical Model for Road Cycling Power." Journal of Applied Biomechanics, 14(3):276–291, 1998. Equations 6a, 9a and 15; rolling-resistance coefficient p. 280. Accessed 2026-08-23. https://collections.lib.utah.edu/dl_files/b4/8e/b48ef26086091662c561e673d7bd990d77868437.pdf
10. International Electrotechnical Commission. IEC 60529:1989+A1:1999+A2:2013, "Degrees of protection provided by enclosures (IP Code)," edition 2.2, published 2013-08-29. Scope. Accessed 2026-08-23. https://webstore.iec.ch/en/publication/2452
11. UL Solutions. "E-Bikes Certification: Evaluating and Testing to UL 2849." Scope and stated exclusions. Accessed 2026-08-23. https://www.ul.com/services/e-bikes-certification-evaluating-and-testing-ul-2849
12. Machines (MDPI). "Evaluating Low Temperature's Impact on Lithium-Ion Batteries: Examination of Performance Metrics with Respect to Size and Chemistry." 13(12):1114, 2025. Capacity-drop results by cell format. Accessed 2026-08-23. https://www.mdpi.com/2075-1702/13/12/1114
13. Redtail eBikes. US jurisdiction e-bike framework matrix, 51 jurisdictions, retrieved 2026-08-19; counted by scripts/analyze_first_bike_law_axes.py. First-party dataset.
14. Redtail eBikes. Storefront specification sweep, 184 e-bikes, pulled 2026-08-23 by scripts/build_first_bike_dataset.py; range model in scripts/first_bike_range_math.py. First-party dataset.