Tesla Has the Highest Accident Rate of Any Auto Brand | Hacker News Reader

Tesla Has the Highest Accident Rate of Any Auto Brand

forbes.com

51 points7e39 comments

There is a reason accidents are done on a per mile basis, so let's make a quick example to show why.

There is two cars, car A and car B. Each driven by 1000 drivers each year.

Car A: The drivers drive a total 10000 km each year, and 20 people get into one accident each, averaging one accident every 500 km.

Car B: The drivers drive a total 1000000 km each year, and 40 people get into one accident each, averaging one accident every 25000 km.

Which car is safer? According to this opinion piece it's car A. But anyone that has even surface knowledge about statistics would realize that it's car B.

It would take only 500 km in car A for someone to have gotten into an accident on average, while for Car B it would be 25000 km. Drivers of car A are therefore on average 50 times more likely to get into an accident. Statistically, if car A was driven as far as car B, every single driver would have had two accidents each.

Conclusion: It is impossible to reach the conclusion in the title given only accidents per total number of drivers. More data is needed.

One small paragraph on what the headline suggests the article will be about and the rest of the article is talking about Tesla’s latest recall

But is it not reasonable to assume that all consumer car makes drive similar distances in average?

No. High-mileage drivers have a reason to prefer electric vehicles, since they're cheaper to operate per-mile than gasoline vehicles.

But range is less… we need more data before making assumptions.

A reason, but realistically, EV buyers are, anecdotally, buying them because they're greener and cooler than ICE vehicles. Very little people I know are doing it for the perceived cost savings and almost none of the EV owners I know are high-milage drivers.

True. Ideally high mileage drivers are electric, but that doesn't make it true.

This data from 2018[1] indicates that the most frequently driven car models are driven 25% more than average, and the least are driven 75% less than average. That is individual models, which you would expect to average out over multiple models. The large car companies like Ford and GM with a broad range of model types and customer base are in the middle of the pack in the lending tree report, as expected, while smaller companies tend to fall at the extremes. This could be explained by either differences in the types of customers they attract, or differences in driving distance.

For example RAM only sells trucks, and their drivers may not be any worse than other truck drivers, but they don't sell other vehicles to bring the average down.

Looking at the brands with the lowest crashes, they are one that are stereotypically owned by older drivers who drive significantly less miles.

All in all, enough suspicious correlations for me to take the data with a grain of salt.

[1]https://www.freep.com/story/money/cars/mark-phelan/2018/07/2...

A suspect there's also an aspect than longer journeys are safer to take into account. Merging onto a high speed road is a time where accidents are more likely, but you might only do it once between cities. Cruising at high speed on a good road0 is relatively safe.

Drive in a city and you may be merging 5-6 times as you move from road to road.

I'm just saying that assuming a linear relationship to distance is probably incorrect as well.

You can actually assume fewer Miles per year per car with EVs. 4500 Miles per year fewer:

https://cleantechnica.com/2023/11/13/new-study-finds-electri...

That makes this Dara even more damning

The author is not a reporter but the VP at the ARC Advisory Group. Here are their clients : https://www.arcweb.com/about/clients. (Exxon, BP, Mobile, Shell, etc. )

Nothing to see here

This article is just reporting on these findings:

https://www.lendingtree.com/insurance/brand-incidents-study/

I don't know why you would seemingly discredit the messenger but eschew from the main point that Tesla drivers have the highest accident rate.

Why is the VP of a consulting company writing articles reporting on smallish studies made by car lenders? Don't you find that odd?

The results should of course be taken for what they are, but the motives of the author are not insignificant.

I think it's relevant information in this case. If you saw a blog post on exxon.com saying Teslas are dangerous it would never be considered newsworthy and never make to YC.

ARC is paid by oil companies to write articles that make their position appear to be newsworthy. They are very upfront about article generation: http//www.arcweb.com/consulting-services/marketing-communications

I was involved in two fender benders. In both of them the other driver was at fault.

In both of them the other drivers asked me to settle without insurance and said they would lie to the insurance company if I chose to go that route. One was a powerful looking executive type who worked at Apple. She said to my face that she would lie. She even said it on camera audio (I was openly taking a video of our cars and the damage) and on Apple property in the presence of her coworker with a security person / witness nearby. She’s lucky I’m not the vindictive type.

I didn’t accept their offers to forego insurance, because I knew I had done nothing wrong and I had video footage from 8 cameras, because Tesla.

And guess what. I would get counted, twice, in the statistics with the approach in the article, because I went through insurance when a lot of people would not have done so, had they been driving any other car.

So... don’t believe this BS article. Also accidents should be counted per mile, not per person, and fault should be factored in imho. There are plenty of ways to slice data, but some are more misleading than others.

First of all, your point about accidents per mile seems fair. But your point about fault doesn't seem fair, because I can't think of any reason why Tesla drivers would be in more accidents despite being at fault less. A positive correlation between number of accidents and responsibility for those accidents seems obvious, while a negative correlation (or no correlation at all) seems extremely unlikely. Of course, there might be some minor variance related to geography or the type of vehicle, but it stands to reason that people who cause accidents often will be in more accidents than people who don't cause accidents often.

Anecdotally, all of the EV drivers I know are driving fewer total miles than ICE drivers. If you can find a study with a good sample size that says EV drivers are driving more miles on average, I'd be very interested to see it. As it stands, I think that correcting the numbers for miles driven would make Tesla look worse here, not better.

fwiw it came that the underlying insurance industry study here was based on the driving records of drivers (including drivers who have never been in a Tesla) who were shopping for insurance quotes for a new or upcoming Tesla purchase.

Useful anecdotal info: Most of the people I know who a) want Teslas, b) have the money, and c) haven’t bought yet, are just waiting to put a few more years on their current car so as to be wise their spending.

With that in mind, it’s clear what is going on. People who want a Tesla and who total their current car suddenly have the excuse they needed to pull the trigger. And surprise surprise, having just totaled a car (setting aside who is at fault), they have an accident on their record.

And that’s how you get skewed numbers. Because MANY people starting fresh with a new car want a Tesla. And some subset of those are only in their buying mode due to an accident.

Judging by the number of silly youtubes of bozos driving their Teslas while pretending to sleep, there's definitely a slight extra danger factor for a car that advertises as 'full self driving' but doesn't.

I think part of the reason is Tesla's have crazy fast acceleration and are also probably by far the largest market penetration of cars with that level of acceleration. People migrating from most conventional cars aren't used to driving sports cars. Also anecdotally given the acceleration, a few of my friends that drive teslas sometimes seem to ignore yellow lights.

From the same source, Ram drivers have the most incidents (all types), BMW the most DUIs by a wide margin

https://www.lendingtree.com/insurance/brand-incidents-study/

Also worth noting the worst is approximately 2x the best ~(15->30)/1000

> Also anecdotally given the acceleration, a few of my friends that drive teslas sometimes seem to ignore yellow lights.

They bought Tesla's, good decision making doesn't seem to be in their wheelhouse.

> Also anecdotally given the acceleration, a few of my friends that drive teslas sometimes seem to ignore yellow lights.

Can you uh, ask them to stop doing that?

Another day another "journalist" posts about Tesla in a misleading way.

>But it comes amid news that Tesla recently recalled more than 2 million Tesla vehicles over a safety issue related to its Autopilot software — specifically, a feature called Autosteer, which is part of the driver-assistance system. The recall affects nearly all the cars Tesla has sold in the United States.

The "recall" is an OTA update.

The "safety issue with autopilot" is that people apparently do not need enough IQ to trick it that you pay attention. In other words, people are the issue, stupid people.

Being an idiot driver seems to correlate pretty strongly with how much one uses autopilot.

I’ve known people that drive drunk and think it’s ok because auto pilot is on. Absolutely horrifying.

They don’t even have to necessarily use the autopilot often. I know people who are not confident in their own driving skills and therefore buying a Tesla is automatically a preferred option because of Autopilot. There isn’t even consideration to any other brands’ self driving capabilities.

> There isn’t even consideration to any other brands’ self driving capabilities.

or to Tesla’s self driving capabilities it seems.

Clicking through to the lending tree report, Tesla drivers have below average DUI rates: 1.02 per 1000 drivers ranking them 19/30, with BMW the worst at 3.13, and Mitsubishi the best at 0.89. I wonder how much of this is due to Tesla drivers drinking and driving less, and how much is due to Auto Pilot making them less likely to be caught unless they crash.

Even without autopilot a modern vehicle has lane following, adaptive cruise control, etc. so if you’re not drunk enough to fail to turn those one, you may just look like another driver.

> This was not a causal study; the study did not analyze the reason for an incident.

Therefore, this was not a study. It was counting something to be able to publish a number, write an article with a laughably indefensible conclusion in an attempt to generate viral results.

Silly rabbits.

There is a full court press against Elon. Not that it is all undeserved but I'm so sick of hearing about it. Every other day the media or the government has some fine or hit piece on him or his companies. Just read Brendan Carr's dissenting opinion on the FCC decision.

This is a good point. It is disgusting that lending tree would put out a report about accident statistics with Tesla at the top. Considering these attacks on mr musk they should have done the analysis different so Tesla is lower on the list

He seems to intentionally invite this, so suggesting it is a targeted attack is a bit unfounded.

I'm also sick of it. But is the appropriate response for everyone to ignore this behavior and just accept whatever he does next? Is that really good for anyone but him?

Well, yes, it's good for everyone that observes it and repeats it. He learned this trick, not invent it.

Multiple times on most days Musk puts out a "hit piece" on the government, a company he doesn't like, a person he disagrees with, whatever comes into his head. He bought Twitter so he would have a platform to exercise his "free speech" and decide what that means for every other user. He's a bully with a pulpit, not a victim.

Correlating driver accident rate to vehicle brand is stupid on so many levels.

And numerous U.S. states have announced future bans on the sale of new gasoline-powered cars. Those bans should be repealed.

If anything we should be decarbonizing faster and avoiding the situation where everyone just replaces their ICE vehicles with electric. Just because it's convenient to move 1-4 humans around in a couple of tons of metal doesn't mean it's feasible to allow at scale. Even allowing existing gasoline vehicles is continuing to let car manufacturers and fossil fuel companies externalize their costs to taxpayers.

Unfortunately, that would be difficult in much of the US. Things are just way too spread out for public transportation to be viable without unrealistically massive investment in high speed rail.

It will make the decarbonization slower, because people will be driving old cars for much longer.

Reply on news.ycombinator.com