Safety of Car Models

There is safety data on cars by model and types of cars. There is insurance and regulator data for whether particular car models are involved in more accidents and if the drivers of particular cars have more fatalities.

The Lexus CT has the most at-fault accidents. Drivers of the Lexus CT get involved in no-fault crashes at the highest rate in the nation. 11.62 percent of CT drivers have reported a no-fault accident on record within the past seven years. This is 69 percent greater than the national average. The IIHS (Insurance Institute for highway safety).

Sporty cars dominate the top ten most accident-prone models. Luxury vehicle are among the models with the lowest crash rates in the country. Drivers of the Audi A4, the Tesla Model Y, the Lincoln Navigator, and the Porsche 911 all cause accidents at bottom-ten rates. they have a combined 38 percent lower car accident involvement rate than the national average.

There is also data on which car models are the most dangerous for the drivers of those models.

The Ford F-Series pickup is the most-often crashed vehicle in America. The top-selling pickup was involved in over 10,000 fatal crashes over our 5-year study period. ValuePenguin analyzed fatal crash data from the National Highway Traffic Safety Administration (NHTSA) to find which cars, trucks and SUVs are most likely to be in fatal crashes.

According to the Bureau of Transportation Statistics, there were 8.7 million motorcycles on the road out of 273.6 million total. That means that motorcycles account for just 3% of vehicles but are involved in 10% of fatal crashes.

Furthermore, the typical number of motorcycle occupants killed per crash is a staggering 0.98. This means that if a motorcyclist is involved in a fatal crash, it’s the motorcyclist or their passenger who is killed nearly every time.

Older cars (12-14 years old) from 2008-2009 were involved in the most accidents. This could be explained by improved safety standards, like collision avoidance technology and backup cameras. Drivers could also be more safety-conscious when driving a new car.

Tesla has far lower involvement in accidents but Tesla has far newer cars. About half of all Tesla’s ever made were produced in the last 12-18 months.

In the 4th quarter of 2021., Tesla recorded one crash for every 4.31 million miles driven in which drivers were using Autopilot technology (Autosteer and active safety features). For drivers who were not using Autopilot technology (no Autosteer and active safety features), Telsa recorded one crash for every 1.59 million miles driven. By comparison, NHTSA’s most recent data shows that in the United States there is an automobile crash every 484,000 miles.

SOURCES – IIHS, Insurify, Tesla, Bureau of Transportation Statistics, Value Penguin
Written by Brian Wang,

8 thoughts on “Safety of Car Models”

  1. Why is the Tesla so outstanding? Since most crashes are caused by driver error, is the cabin (driver) monitoring in the Tesla the reason? There are other cars with driver monitoring, eg some Subaru. If driver monitoring is the reason for such a dramatic benefit, should it be mandatory on all cars?

  2. A very important factor is driver monitoring. Driver error is the cause of the majority of crashes. Tesla has an in cabin camera that picks up things like driver inattention. Some other models have this eg some Subaru models. It would be very interesting to see if models with this feature do better. Is this the reason why Teslas are so much safer?

  3. The statistics given in this post seem to show that Teslas driven using “autopilot” have approximately a tenth the rate of crashes than the national average (1 crash every 4.3 million miles driven for Teslas using “autopilot” while the figure is 1 crash every 0.48 million miles driven for the national average).

    The way I would interpret that is that Teslas driven using “autopilot” are approximately 10 times safer than the average car and driver in the USA.

    Given that apparent much safer driving record for driving using “autopilot”, why is some government safety agency (I don’t remember which one) currently doing a level 2 investigation of Tesla over the safety of “autopilot”, potentially leading to a recall order? Is this clear evidence of gross bias against Tesla / Musk? Or is there something about those statistics I am not understanding that makes my conclusion wrong that Teslas driven with “autopilot” are much safer?

  4. Next to useless. You must take into consideration how many vehicles there are of a particular model at bare minimum, better would be the number of miles driven in that model…relative to the fatality rate.
    And the number and rate of fatalities is not the only concern. Injuries that leave people incapable of working or at far diminished capacity is a very serious thing for those people, their families and the economy.
    Automakers designing cars that avoid killing but leave people maimed is still unacceptable to me. They make the cars to fit the test. But the test is at a slow speed. If you use all the crumple zone in the slow test, you get your 5 stars, because of the lower G-forces. But when those vehicles crash at higher speeds, bad things happen.
    I’d like to see vehicles passing the tests at 55 mph, with no spinal or head injuries, and no crushing of the passenger compartment, at least where occupants would likely be, in frontal and rear impacts. Side impacts at 55 mph are virtually impossible to protect people in. 45 mph there would be good. Cars would need to be at least double the width for 55 mph. Not sure how you would even open a 3-foot thick door. Or all vehicles would have to have a wedge in front, to scoop under another vehicle in a side accident, throwing them into the air instead. But it is not realistic to change all the existent vehicles.
    Just 60 mph to 0 mph in a fraction of a second is already very difficult to avoid brain injury, regardless of how well protected. It is like trying to engineer a light helmet to take the impact of a heavy, fast moving baseball bat to the head. You can prevent the scull from fracturing, but keeping the brain healthy is next to impossible without increasing the size of the helmet to the size of a beach ball.
    Race cars often protect drivers even in some very high speed accidents, so protection is not inconceivable. Some of that is 5-point safety harnesses and helmets. I don’t think we could ever manage to get everyone to wear helmets, but safety harnesses might be doable of some type. The shoulder and lap belts just are not enough in my opinion. Of course, airbags are very beneficial, and accident avoidance sensors and software. Traction control can be beneficial.
    I also think we need better breaking standards. 60-0 in 130 feet or less. Current regulation is 323 feet. I think a large number of accidents could be prevented or reduced in their severity with better breaking.
    The roads themselves could be much safer. Where I live, they don’t even have curbs except where there are 4 lanes roads and in some commercial areas. And even in some 4-lane areas, there are no streetlamps.
    True full self-driving? I still think this is years away…perhaps even a decade. Good enough for most situations is not an acceptable standard. It has to manage all situations…except maybe a sinkhole opening under the car. And it is going to need a microphone and speaker to communicate with police or roadwork crews. Does it even hear other horns? Can it recognize a disabled vehicle. Can it recognize a flat tire, steam or smoke coming from under the hood, broken glass on the street, smashed cars? Does it know when another driver left their vehicle? If it does not recognize these things, it can’t leave itself enough room to go around them. It will end up too close to the back of these vehicles, and it will take forever to get around it, if it even figures out that it needs to, unless it is going to wait an hour for the tow truck to remove that car.

  5. Need to adjust accidents by number of vehicles sold. Chevy Silverados seem to be involved in more fatal crashes than F-150s when adjusted for sales. Trucks also are used in different ways than cars. Dirt roads, towing, etc. lead to more possibilities of fatal accidents.

    Chevy Impala is significantly more likely to correlate with a fatal accident than every other vehicle. Probably an interesting story behind that correlation.

    Cars that are 12-14 years old are more likely to have been resold. Second owners are probably not as safe as original owners. Also possible that in the absence of proper maintenance that there are issues that lead to fatal crashes in older cars.

    • Many people with 12-14 year old cars are living in areas where there are more reckless drivers, and less safe roads. Poorer areas have less streetlamps, more drunks walking around in the dark, less sidewalks, less traffic lights where they are needed, more feral animals to doge, and more intoxicated drivers. People with less happy, secure lives tend to use more, so more intoxicated drivers.
      Tires can also be more worn, breaks more worn. And he is right about safety features being better in newer cars. Though, I doubt many people die from lack of backup cameras, pets maybe. Some sure. I just doubt it makes much of a statistical effect. Other stuff like side impact airbags, blind spot warnings, stronger bodies, but also most of the new vehicles are not sedans, compacts and hatchbacks, they are larger and mostly safer except in rollovers.

    • It’s not just number of vehicles that should be factored in, but mainly number of miles driven by the vehicle. I would think a delivery vehicle would be much more likely to be involved in a crash than grandma’s Towncar just because the Towncar sits in the garage six days a week while the delivery vehicle is on the road eight hours a day.

  6. The chart of which vehicle has the most occupants killed per fatal crash seems a bit obvious, 0.2 for bus, 0.98 for motorcycles. Duh. More interesting would be how many occupants died per mile per crash.

Comments are closed.