Road deaths only happen every hundreds of millions of vehicle kilometres travelled. It is only that so many vehicle kilometres are travelled that the number of road fatalities are as many of they are. And that makes it hard to dig into the numbers and draw conclusions in a country where there as few road deaths as Australia has. So few road deaths, so many brands and models, means you would be drawing dubious conclusions based on very small numbers of random events.
This study of what model, brands and vehicle types are the most dangerous was done in the US where the numbers are bigger.
Some of its results are entirely plausible. Like that you are more likely to die as an occupant in a small car than a large car, with large SUVs and trucks the safest place for an occupant to be. And the sheer number of Teslas on the road, and the small number of Tesla models, would tend to make the numbers for that brand and its models likely to be representative. Though it is harder to understand why a vehicle brand that does as well as it does in crash testing is as unsafe as these numbers say it is. Is it the fact that it is high-performance, and so attracts unsafe drivers and unsafe driver behaviour? Is it that it has autonomous features which drivers rely on to work, and that encourages them to rely on them? Certainly the fatality rate for Teslas is consistent with the cost of insuring them that indicates they are in a lot of crashes and/or in high severity crashes. It's also likely that you've got a high chance of dying in a small car like the Mitsubishi Mirage in a country where the vehicle you collided with was almost certainly a lot bigger.
But some of its results are harder to see a reason for. Why would fairly ordinary smallish SUVs like the Hyundai Venue and Honda CR-V Hybrid be way up at the high risk end of the scale? Are short narrow high smallish SUVs more inherently unstable, and so more likely to be tipped over in a crash, and does vehicle design and the types of crashes and crash protection crash testing does fail to take that into account? And why are there so many Hyundais and Kias at the high end of the risk scale that those brands rate as poorly as they do? Is it just that that's what those brands sell so many of?
And, yes:
- it was done in the US so it necessarily relates to the cars they drive there, but that's the only data available for the reasons I stated,
- it only relates to recent models, which is why it's a very different picture to what Elon paints because he compares Teslas to all the vehicles on the road, and
- it is based on the fatalities of occupants in that vehicle, not the number in the crash, or which vehicle and driver was responsible, so it only tells us how well the vehicle protects its own occupants, not how safe it is.
Honestly has nothing to do with the car. Just has something to do with the driver. Check out mkbhd