Tesla received a cease-and-desist letter from NHTSA over Model 3 safety claims
cnbc.com
cnbc.com
> The Vehicle Safety Score represents the “relative risk of injury with respect to a baseline of 15%,” according to NHTSA. Model 3 achieved a Vehicle Safety Score of 0.38, which is lower than any other vehicle rated in NHTSA’s public documents.
How can the NHTSA create a safety score regarding risk of injury, and then say that you can't use it to make a statement about the risk of injury?
Government is highly averse to saying or doing anything that could possibly come back to bite them (with exceptions beyond the scope of this comment) hence they will compile stats but are non-committal about drawing any conclusions from or interpreting those stats.
1. NHTSA released data showing the Model 3 has the best score in their dataset.
2. Tesla happily publishes this to show off.
3. NHTSA gets grumpy.
4. (???)
NHTSA is taking issue with how the results are being framed. The ratings are only valid across classes in some cases and for stuff like front impact rating the numbers are only valid for cars with very similar weights, Tesla is using the numbers to say they have the lowest probability across all cars. See [0] page 7 for the original letter, the NHTSA has very specific guidelines for what their numbers actually mean and how they can't be compared in all cases across different vehicles, Tesla has violated those by comparing their score to all vehicles which definitely includes vehicles outside the +/- 250 lbs weight difference the NHTSA feels is appropriate.
[0] https://www.plainsite.org/documents/fnrhg/tesla-nhtsa-foia-r... (still reading but see bottom of page 3 and top of page 4 and the NHTSA letter on page 7)
edit: To expand having read more they specifically say that in a collision between vehicles of different classes the numbers become largely meaningless and that the larger vehicle like an SUV will have the lower probability of injury compared to the smaller vehicle.
I’m not making any claim about Tesla’s cars (because of its low center of gravity and relatively large weight, they likely are quite a bit safer than the average car in collisions) but if brand B has distracting UI, bad steering, or whatever, that causes it to slam into walls twice as often as cars of brand C, being safer in a single collision may not offset that.
The methodology simply doesn't provide the data for making the type of global claims that Tesla make.
A minivan has a higher center of gravity so in an example where a minivan is cornering aggressively, say to avoid a collision, the rollover likelihood is higher than that of a sports car. As I understand it safety ratings are generally after occupancy protection and aren't concerned with the class of vehicle which goes against what you've claimed.
From the NHTSA website [0]:
"Can I compare vehicles from different classes?
Side crash rating results can be compared across all classes because all vehicles are hit with the same force by the same moving barrier or pole.
Rollover ratings can also be compared across all classes. Frontal crash rating results can only be compared to other vehicles in the same class and whose weight is plus or minus 250 pounds of the vehicle being rated. This is because a frontal crash rating into a fixed barrier represents a crash between two vehicles of the same weight."
And again they are, generally, looking for impact to occupant, for example for the "Side Pole Crash Test Scenario" under "Test Details" they are looking at "Evaluation of injury to the head, chest, lower spine, abdomen, pelvis".
Just because a sports car is a different class of vehicle does not make it inherently less safe. The driver input can change that for any class of vehicle.
A minivan weighs over 4500 lbs, while a sports car probably weighs 3000 lbs. In most collision situations, the heavier car is safer.
Frontal collision tests (and frontal overlap tests) are against stationary objects. So a minivan is tested with 4500+lbs of weight, while a sports car is only tested with 3000 lbs of weight.
Some tests are weight dependent. Others have a static weight.
-----------
I know some friends who survived getting hit by a bus. Riding in a 5000lb+ vehicle helps in these situations. There is a huge degree of safety that is afforded by just having a heavier vehicle.
As an aside I have lots of friends who own fast cars who haven't gotten into accidents by avoiding them. Thankfully they were driving lighter vehicles.
I won't berate survivorship bias here.
As in driving aggressively and avoiding accidents they were about to create ? Statistically most accidents are rear ends and side collisions, both of which you most likely won't see coming with enough time to react let alone move your car. And getting plowed by an SUV when you're sitting in a sport car means your head is on the absolute worst position you could think of.
Bikers say the same "I'm small, agile and fast, no accidents for me", yet they're disproportionally represented in crash stats.
Again, with the biker statement, since you'll never find a data set that would show accidents avoided on the particular vehicle there's a huge swath of underrepresentation of the reality. And one can argue proportionally bikers are in more crashes because there are less of them overall on top of the fact that most drivers of cars and trucks do not pay considerate attention to smaller vehicles.
And finally my comment about the "friends with fast cars" was a tongue in cheek response to the fact that the parent I was replying to made a large jump by saying SUVs are safer because of size and weight.
Or you know, not being on distracted autopilot: https://www.theguardian.com/technology/2016/jul/01/tesla-dri...
People get distracted, and when they're distracted they'll crash into things.
An alternative solution is "don't be distracted", but that doesn't seem to work in practice.
And we all have friends who regularly avoid accidents in motorcycles. The difference is that you HAVE to avoid an accident to survive them, especially for motorcyclists (the extreme case of smallest but most agile vehicle). While if you're in a big car, you can have an accident and survive.
Its a very big difference. If you ever are a bit tired or distracted, the "trust my reactions to save me" technique will fail.
But if you're tired / distracted, but armored up in a 5000lb or 6000lb vehicle, you'll probably survive whatever hits you.
The survival rate of 18-wheelers (even when they're distracted and tired from driving 10+ hours a day) is quite high. Multi-ton armor just makes things safer.
You're heavier than the cars around you which helps a lot, but the extra weight also smashes through a lot of obstacles.
Electric cars actually weigh more due to batteries, but that’s not particularly helpful.
That is incorrect. Your statement is only true in the case of single vehicle inelastic collisions, which is the only type tests NHTSA conducts.
What I wrote is in quotes and directly taken from the NHTSA site. Those are not my words. Try taking a look at the site first.
---
> The documents, obtained through a Freedom of Information Act request, also include orders for information that NHTSA sent to Tesla following several crashes, including a fatal March 1 crash involving a Model 3 operating on Autopilot. Tesla shares fell 0.5% in New York pre-market trading to $229.51.
NHTSA seem rather unkeen for Tesla to say the good without the bad, and are more than happy to try and prove that the marketing fluff Tesla has forwarded isn't a complete enough story for the public to make a decision.
---
[0] https://www.bloomberg.com/news/articles/2019-08-07/tesla-dra...
Notice the teslas avg weigh more than everything but the semi's. 3: 3,552lb-4072lb | 60D 5,072 lb (2,300 kg) | 70D 5,072 lb (2,300 kg) | 75D 5,140 lb (2,330 kg) | 90D 5,271 lb (2,390 kg) | P90D 5,381 lb (2,440 kg) | 100D 5,421 lb (2,459 kg) | P100D 5,531 lb (2,509 kg)
Sounds like something Trump would say.
Pretty much anyone can be made to sound like anything when you take a sound bite.
I’ll take this blog post from Tesla over, for example, the Ford commercial showing their concept electric truck pulling a train, any day.
[1] - https://www.tesla.com/blog/model-3-lowest-probability-injury...
And, maybe Ford also sounds like Trump. I don't know. Maybe all salesmen sound alike. I don't much care. "But some other guy did it before" is a weak argument.
Sounds like you have some pretty serious bias. Meanwhile, Tesla claims their non-existent truck concept can pull 300,000lbs. Has Ford ever claimed something similar?
It just isn't as big an accomplishment as it looks, since it was on a railway.
Tesla Model 3 does have the lowest weighted probability of an injury of any car ever tested by NHTSA. The Ford prototype did pull 1 million pounds of railway cars.
In the case of Tesla, it’s important to understand the limits and intent of the test and that it’s the lowest probability of injury when crashing into certain types of stationary objects or cars of roughly the same mass. That’s an important caveat but the Tesla did score much better than any other car ever has, and I think it’s fair to brag about that.
In the case of the Ford, it turns out any vehicle that can pull with ~2,000 lbs of force could do the same thing. To me, that makes the demo basically a gimmick and “misleading” to the general public.
By comparison, the world record holder for pulling a train with his teeth pulled over 575,000 pounds. Ok, now that’s impressive! [1]
On its own, though, that's not very relevant to the Tesla story. It might be made more substantive by linking it to a larger issue, such as how we think about the line between acceptable puffery [1] and unacceptable false advertising, and how that interacts with the mindsets of sales versus product definition versus implementation.
NHTSA describes that they have a specific guidelines document (which has a rule about comparing vehicles 250 lbs apart in weight) which Tesla apparently hasn't followed, so they're not making up rules selectively against Tesla.
All I am reading is that the NHTSA has bad tests?
If they are so mad Tesla is saying "We have the best crash test rating of any car", why not change the test to include those issues?
If someone was mad that I said "My code passes all of the tests", but there were still bugs in the code, the issue isn't my statement. The issue is that there is incomplete coverage.
But they are upfront in their guidelines about what exactly their tests do and do not say. Their tests are good for comparing cars in the same weight class. They are not good for saying what is the world's safest car.
Stars is way too marketable and deceptive to begin with. Make it Pass / Fail. 4 stars to actually failing is "fail". 5 stars is "pass". Actually failing being a rejection of approval and such a car wouldn't get a rating.
"Our car passes all necessary safety tests" ... good?
The last one might need a little elaboration. The NHTSA battery of tests is relatively small and limited. The lower the risk of injury in those tests is, the more dependent the actual real-world risk becomes on all the scenarios they didn't test. This isn't a hypothetical problem. The Model S also got the best score ever in the NHTSA tests and Tesla did a similar press release about that, but it didn't get the IIHS's highest rating because it performed badly in small-overlap collisions which weren't part of the NHTSA tests.
Elon is saying more than "My car passes all of the tests"
Or, they can do what they have done and tell Tesla to stop exaggerating the claims, which is a pattern with this company.
Seems like a testing policy designed to keep people from avoiding buying small cars ever, and/or to keep people from agitating to ban or otherwise legally discourage heavy road vehicles.
The problem with making comprehensive tests would mean small cars would be disadvantaged. You wouldn't pair up a heavyweight MMA fighter with a strawweight to figure out who is best, and likewise, you wouldn't have a bus and a small car t-bone each other to determine which is safer. (If you think a bus is a bad comparison since they're not usually sold to the general public, just try a pickup truck vs a small car.)
Because testing cars is expensive, and why should the association be beholden to a manufacturer's desire for marketing?
I believe the Tesla marketing would have been more true if it had included "... in an NHTSA-style incident" or some such. But that's not how marketing works.
I am suspect of the safety of all cars that don't have rally/track style roll cages and four point harnesses (with fire extinguisher in easy reach). But I would go in a Tesla.
We have all seen the crash test videos, we have all been driving at twice the speeds in those crash test videos. We have all stepped it on a little bit more to be travelling past the speed limit.
We have all 'rubber necked' at some ghastly accident on the other carriageway and wondered how so much carnage can happen on a perfectly straight road. A million people get killed on the roads globally every year.
On a motor racing event the medics and fire crew are minutes away, ready. There is no on-coming traffic. There are no people drunk or partially sighted on the road. The barriers and run-off areas have been tested. Helmets are mandatory. The road ahead is the same circuit it was two minutes ago, last time it was lapped. Features in the circuit are there to restrict top speed - chicanes etc.
Meanwhile, on the autobahn you can do the same speeds. Or, elsewhere you can find an overtaking car in your carriageway doing 70+ mph, combine that with your 'safe' 50 mph and imagine how that head on crash compares to the test videos that are at a fraction of the speed. Imagine how well your brain copes smashing into your skull in the impact. Just because everyone is driving along in all weather conditions tailgating the guy in front every day does not make it wonderfully safe. So yes I am serious about the need for a roll cage or a Tesla.
So literally every car then.
Yes, you can modify a stock car until it becomes a race car and then also drive that car on the road. But since no car ships from the factory like that it's insignificant enough to be completely discounted.
Stars are how well the car contains the passengers during an accident in a variety of situations. 5 stars does not mean "low risk of injury" at all. Infact, a 1 star Sterling box truck will probably cut a Tesla with a 5 star rating in half. Guess who's walking away almost 100% of the time?
The star rating is more a rating of how well the car holds together, prioritizes it's occupants, and the effectiveness of it's safety features. They mostly don't consider factors such as the mass of the vehicles involved or other important factors that become non-arbitrary as soon as you enter the real world.
So what is the average risk of injury in all of the following situations for each of these vehicles (Sterling box truck, Tesla Model 3)?
- 50 mph against a thick wall
- 50 mph against a Sterling box truck
- 50 mph against a random car
- 50 mph into a dense forest
- 50 mph falling off a 50 foot ledge
And what's the average risk of getting into these situations, for weighting?
My point is that a Sterling box truck might crush a Tesla Model 3, but the vehicle's own mass isn't always helping and with no airbags and crumple zones the risk of injury is still higher in many common situations. So let's not overrate one particular situation vs a large truck.
50mph Model 3 vs a 50mph Sterling Box Truck.
Situations when OTHER objects run into you. Its unfortunate, but people "armor up" and buy heavier vehicles for safety in these situations.
Everybody buys 5 star cars. So there's an abundance of them. When a Chevy Spark crashes into a Toyota Corolla it's a good representation of what crash tests simulate. Sure they could make every car crash into a semi truck and give them a rating but none of the dummies would survive. None of them. Even the mighty Tesla. But trucker Bob goes home to his family at the end of every shift.
They would be grading flaming chunks of steel and plastic all day. That's why I drive an SUV. Because I'm broke and can't afford a new "safe" car, but in all liklihood I'm bigger than the median vehicle size so that gives me comfort.
Not all crashes involve other vehicles, you have to take into account the frequency of crashes vs. the environment.