3 ms·

I decided to just look into an assertion from the first paragraph of this post: "NHTSA testing shows that the #1, #2 and #3 cars with the lowest likelihood of i

I decided to just look into an assertion from the first paragraph of this post: "NHTSA testing shows that the #1, #2 and #3 cars with the lowest likelihood of injury in a collision are the Tesla Models 3, X and S in that order." This is linked to a news article on "CleanTechnica," sourced to the Tesla corporate blog. Notably, the blog does not claim that these vehicles are #1,2,3 in the rankings, but rather that in prior tests the X and S models were found to have "the lowest and second lowest probabilities of injury of all cars ever tested." But it turns out NHTSA disputes even Tesla's more mild assertions: "A five-star rating is the highest safety rating a vehicle can achieve," the agency said in the statement, which did not name Tesla. "NHTSA does not distinguish safety performance beyond that rating, thus there is no 'safest' vehicle among those vehicles achieving five-star ratings." https://autoweek.com/article/car-news/nhtsa-downplays-teslas-statements-model-3-crash-ratings https://autoweek.com/article/car-news/nhtsa-downplays-teslas... In 2017, just over 30 cars attained 5-star ratings: https://www.safercar.gov/Vehicle-Shoppers/my2017_5_star_tests https://www.safercar.gov/Vehicle-Shoppers/my2017_5_star_test...

kbenson 8y ago
> But it turns out NHTSA disputes even Tesla's more mild assertions: I think it's worth noting that neither are contradicting each other, as they are saying different things. Tesla says NHTSA testing shows something, which may have nothing to do with specifically what the NHTSA says or how it rates. If they have access to the actual test data (or the tests are publicly available), they can make assertions off that data. If that was the case, the NHTSA is clarifying its position, in that it is not making those claims so them saying "NHTSA does not distinguish safety performance beyond that rating, thus there is no 'safest' vehicle among those vehicles achieving five-star ratings." is a way of clarifying their position and staying neutral, since they haven't made that assertion for any other vehicles in the past. In other words, it's possible for them both to be be correct and accurate, but talking about slightly different things. This doesn't even require intent to deceive from either party. I, of course, have no idea if that's true and this is the case. It is one possible way to interpret the statements though, so a little additional investigation might be warranted before we determine that either statement is false or intentionally misleading.
- singingboyo 8y ago
  This was my impression as well, in that what I've read suggests Tesla is making claims on the basis of percentage chance of injury from NHTSA data, but NHTSA does not take a position on whether differences in those percentages is significant. I can't say I've verified the percentages myself though, nor do I have the requisite understanding to say whether there are margins of error present (which would make Tesla's claims correct but potentially statistically meaningless.)

kbenson 8y ago

> But it turns out NHTSA disputes even Tesla's more mild assertions: I think it's worth noting that neither are contradicting each other, as they are saying different things. Tesla says NHTSA testing shows something, which may have nothing to do with specifically what the NHTSA says or how it rates. If they have access to the actual test data (or the tests are publicly available), they can make assertions off that data. If that was the case, the NHTSA is clarifying its position, in that it is not making those claims so them saying "NHTSA does not distinguish safety performance beyond that rating, thus there is no 'safest' vehicle among those vehicles achieving five-star ratings." is a way of clarifying their position and staying neutral, since they haven't made that assertion for any other vehicles in the past. In other words, it's possible for them both to be be correct and accurate, but talking about slightly different things. This doesn't even require intent to deceive from either party. I, of course, have no idea if that's true and this is the case. It is one possible way to interpret the statements though, so a little additional investigation might be warranted before we determine that either statement is false or intentionally misleading.

singingboyo 8y ago
This was my impression as well, in that what I've read suggests Tesla is making claims on the basis of percentage chance of injury from NHTSA data, but NHTSA does not take a position on whether differences in those percentages is significant. I can't say I've verified the percentages myself though, nor do I have the requisite understanding to say whether there are margins of error present (which would make Tesla's claims correct but potentially statistically meaningless.)