E-Bike Accident Injury Severity Prediction: Why 20 MPH Crashes Generate $2M Claims

The Left-Turn Collision That Rewrote a Carrier's Reserve

You're reviewing a file from last Thursday. A 34-year-old plaintiff was riding a Class 2 e-bike through a marked intersection at roughly 22 mph. A sedan making a left turn struck the e-bike broadside. The vehicle's front bumper shows a cracked fascia and a minor hood dent. Repair estimate: $4,200.

The rider hit the hood, then the windshield, then the pavement. Comminuted tibial plateau fracture. L1 burst fracture. Grade III AC separation. Two surgeries in the first week. A life care plan projects $1.4M in future medical costs alone.

The adjuster's first instinct is that vehicle damage this minor shouldn't produce injuries this severe. That instinct is wrong. And it's wrong because of physics that most claims professionals have never been trained on.

E-bike accident injury severity prediction is now a core problem for anyone handling BI claims in urban markets. The reason is straightforward: an unprotected human body absorbing crash energy behaves nothing like one strapped inside a 3,500-pound vehicle with a crumple zone, airbags, and a seatbelt pretensioner.

Why E-Bike Collisions Don't Follow Car-on-Car Rules

In a car-on-car rear-end at 15 mph, the struck vehicle's structure absorbs a large fraction of the impact energy. Crumple zones deform. The occupant's Delta-V might be 6 to 9 mph depending on mass ratio and structural stiffness. A seatbelt and headrest manage the occupant's kinematics. AIS 1 cervical strain is the typical outcome.

In an e-bike-versus-car collision at the same closing speed, none of those protections exist. The e-bike's mass is roughly 60 to 75 pounds. The rider's mass is the dominant factor. There's no crumple zone. No restraint system. A helmet protects the skull, sometimes. The rest of the body is exposed.

The result is a fundamentally different energy transfer profile. NHTSA's 2023 Traffic Safety Facts report on pedalcyclists documented 1,105 cyclist fatalities in 2022, a 13% increase over 2019. E-bikes are now a growing share of that population. The report doesn't break out e-bikes separately yet, but pedestrian and cyclist injury data consistently shows that impact speeds as low as 20 mph produce AIS 3+ injuries when the struck party has no vehicle structure around them.

This is the physics gap that makes e-bike accident injury severity prediction so different from anything in your standard BI playbook.

The Mass Ratio Problem

Consider the numbers. A 2023 Toyota Camry weighs approximately 3,500 pounds. A rider on a Class 2 e-bike weighs maybe 230 pounds combined (170 lb rider, 60 lb bike). That's a mass ratio of roughly 15:1.

In a collision between two cars of similar mass, momentum exchange is relatively balanced. Both vehicles decelerate. Energy distributes across both structures. But at 15:1, nearly all the velocity change transfers to the lighter object. The e-bike rider's Delta-V in a 20 mph impact can exceed the closing speed of the vehicle because the rider separates from the bike and enters a secondary impact sequence: hood, windshield, ground.

Each impact is a separate injury event. The initial strike might fracture the tibia or femur. The hood and windshield contact can produce thoracic and spinal injuries. The ground impact adds head injury, shoulder separation, wrist fractures. Three impacts, three body regions, three distinct AIS scores.

You can run these numbers yourself. Our free Delta-V calculator handles vehicle-to-cyclist scenarios and gives you a force profile for each phase of the collision sequence.

"The single biggest error I see in e-bike claims is treating the vehicle's damage as a proxy for occupant injury severity. In car-on-car, that correlation holds roughly. In car-on-cyclist, it's almost meaningless. The vehicle barely notices the impact. The rider absorbs nearly everything."
Senior biomechanical engineer, 22 years in crash reconstruction

Speed Classes and the 28 MPH Problem

Federal and state regulations define three classes of e-bikes. Class 1 provides pedal assist up to 20 mph. Class 2 adds a throttle, same 20 mph cap. Class 3 allows pedal assist up to 28 mph.

That 8 mph difference between Class 2 and Class 3 matters enormously. Kinetic energy scales with the square of velocity. A 28 mph impact delivers roughly twice the kinetic energy of a 20 mph impact. Not 40% more. Twice.

For injury severity prediction, this means a Class 3 e-bike rider struck at intersection speeds faces energy levels comparable to a pedestrian hit by a car traveling 35 mph. IIHS pedestrian research has shown that the fatality risk at 35 mph approaches 40%. The AIS distribution shifts from predominantly AIS 2 to 3 injuries up to AIS 4 to 5, including traumatic brain injury, pelvic fractures, and internal organ damage.

If you're handling a claim involving a Class 3 e-bike, the exposure calculation is categorically different from a Class 1. And most intake forms don't even capture e-bike class.

What the Vehicle Tells You (and What It Doesn't)

Here's where claims handling goes sideways. An adjuster sees a sedan with a cracked bumper fascia and a small windshield impact point. The vehicle damage score might be 15 out of 100. The reflex is to think: low severity crash, limited injury exposure.

But vehicle damage in a car-versus-e-bike collision is a measure of how much energy the vehicle absorbed. It is not a measure of how much energy the rider absorbed. Those are inversely related. The less the vehicle deforms, the more energy passed through to the rider.

This is the same principle that makes pedestrian impacts so injurious. A flat hood surface doesn't crumple on contact with a human body the way it crumples on contact with another vehicle's bumper beam. The human body is the crumple zone.

Silent Witness handles this by scoring damage and injury exposure independently. When you upload photos of a car-versus-e-bike collision, the system calculates the rider's Delta-V and injury probability using the vehicle's approach speed and contact geometry rather than relying on vehicle deformation alone. That's the only way to get an accurate e-bike accident injury severity prediction.

Injury Patterns That Repeat Across E-Bike Claims

After analyzing hundreds of e-bike collision files, certain injury patterns appear with high frequency at specific impact configurations.

Broadside (T-bone) impact, vehicle strikes rider's side: Tibial plateau fracture, lateral femoral condyle fracture, pelvic ring disruption, rib fractures on the impact side. The lower extremity takes the initial strike from the bumper. The torso impacts the hood edge. AIS 2 to 4 across lower extremity and thorax.

Frontal impact, rider strikes vehicle head-on: Bilateral wrist fractures (Colles or scaphoid) from bracing, anterior rib fractures, facial fractures if the rider goes over the handlebars into the windshield. AIS 2 to 3 upper extremity, AIS 2 to 4 facial and thoracic.

Rear impact, vehicle strikes rider from behind: Lumbar and thoracolumbar spine fractures, sacral fractures, traumatic brain injury from ground impact. This is the highest-severity configuration because the rider has no time to brace and the full closing speed translates to rider Delta-V. AIS 3 to 5 in spinal and head injury regions.

Each of these patterns generates a different claim value, a different life care plan, and a different litigation risk profile. You can see how attorneys use this data in demand packages by looking at the injury causation reports Silent Witness generates for each scenario.

Building the Demand Package (or Defending Against One)

If you're plaintiff's counsel on an e-bike case, your demand package needs to do something most car-on-car demands don't: explain why the vehicle damage is irrelevant to your client's injuries. Defense counsel and the carrier will anchor on that cracked bumper. Your job is to reframe the conversation around rider Delta-V, impact kinematics, and the biomechanical reality of an unprotected body absorbing 20+ mph of velocity change.

A crash reconstruction report that shows the rider's Delta-V was 22 mph, that the impact generated 35 to 40 g's at the rider's pelvis, and that the AIS probability distribution predicts a 74% chance of AIS 3+ lower extremity injury gives your demand an evidentiary foundation that a treating physician's records alone can't provide.

If you're defense counsel, the same data either confirms or undermines the claimed injury mechanism. A plaintiff claiming a lumbar burst fracture from a side-swipe where the rider's Delta-V was 8 mph has a biomechanical credibility problem. The force profile doesn't support the injury.

Either way, the physics is the physics. The question is whether you have it in the file or not.

Where the Market Is Heading

E-bike registrations in the U.S. grew 269% between 2019 and 2023 according to the Light Electric Vehicle Association. Urban delivery fleets (DoorDash, Uber Eats, Amazon) are deploying thousands of commercial e-bikes. Municipal bike-share programs are converting to e-bike fleets.

The claim volume is going in one direction. And unlike traditional bicycle claims, e-bike cases involve higher closing speeds, higher mass, and (in some jurisdictions) novel liability questions about whether the rider was operating a "bicycle" or a "motor vehicle" for purposes of comparative fault and insurance coverage.

For carriers, this means your severity models need a micromobility layer. For plaintiff's attorneys, it means these cases are worth more than your intake team might think based on the vehicle photos alone. For everyone, it means e-bike accident injury severity prediction requires tools built for the actual physics of unprotected-rider impacts.

Our methodology documentation walks through how we validate these models against NHTSA and IIHS crash test data, including pedestrian and cyclist impact configurations.

If you want to see how a specific e-bike collision scenario scores, the free Delta-V calculator takes a few photos and about two minutes.

This content is for informational purposes and does not constitute legal or medical advice.