Yep I believe that is a pretty well established cognitive bias called Illusion of Control [1]. Going down the rabbit hole of cognitive biases is a fascinating journey, and lot of them are relevant when driving a car.
One I found especially interesting is specifically about how improving safety features may not reduce accidents as much as it could, because when people feel safer, they tend to take greater risks [2]
I generally don't fear other people driving drunk (except on holiday evenings when I refuse to be on the road.) My personal risk of dying in an alcohol-related crash is massively reduced by my decision to never drink and drive, or get into a car with a driver who's been drinking. The risk is never zero, but it's low enough that I generally don't worry about it.
Do I have absolute complete control over it? No. Do I have a whole lot of control over it? You betcha. The control is not an illusion.
>Are you a good enough driver to save yourself in this situation?
Are you a good enough pedestrian to dodge a drunk that jumps the curb?
That's a rhetorical question. My point here is that the standard you're trying to apply is an asinine one. Unless you live beside the train station you have some exposure to drunk drivers.
No. Nothing the camera car driver could've done to see or avoid that one.
But surely you realize that's not a very common scenario crash. Watch the endless youtube dashcam videos of crashes, the vast majority are simple scenarios where the driver could've done something smarter than what they did.
> Yep I believe that is a pretty well established cognitive bias called Illusion of Control
It is not an illusion that the car driver is in control of the car and their decisions.
It is an euphemism to call car crashes "accidents" because the vast majority were not an accident that simply randomly happened, it was an event the driver could have avoided by making better choices (e.g. call that cab instead of driving drunk, etc).