TheDonEffect

Yeah, these types of lists are really really misleading. In the end, crash ratings mean more than statistics.

darcyw

those mustang deaths are actually pedestrian deaths.

darcy

jeffbrig

I actually remember a rental Avenger a couple of years ago. While the engine and transmission were absolutely anemic (floor it and wait 3 seconds for anything to happen), the ride was generally good/comfortable, the steering felt solid, and the brakes had a proper linear feel. In a way, I walked away kind of impressed at how competent even a shitbox economy rental car has become. None of that vague play in the steering wheel before the wheels actually turn, or a brake pedal that builds pressure before ANY braking begins, where pedal feel has no correlation with stopping force (my biggest pet peeve). I sure as heck would never buy one, but even cheap cars have vastly improved in my lifetime.

clawhammer

I disagree, I see it the other way around. The crash ratings are the equivalent of studying for a test, whereas the statistics are more representative of what actually happens in real life.

Chibo

You can normalize all you want, but if you have large differences in production numbers to build statistics off they are going to be skewed.

Example being a small production car that is only involved in a handful of accidents due to the number of cars on the road, compared to an extremely high volume car involved in thousands of accidents. Smaller production car may be more dangerous in a crash, but we don't have the data to see that. Normalizing the data draws down the average of fatalities in the high production car, but since the smaller production car never had a fatal accident it comes back as 0.

You also miss the traits that come with certain car ownership, and associated risk factors.

All data should be taken with a grain of salt, that's not to say that I don't believe the Kia Rio and Hyundai Accent are beercans.