Tesla FSD Safety Stats Misleading

[whitex]

This is with their robotaxi fleet in Austin, Houston and Dallas.
Why haven't they released it for all Tesla ? Probably will in time once they get more data as the miles accumulate. As Austin Texas is their first city with their robotaxi fleet, I'm sure they are being extra cautious. Can't predict everything that can go wrong and it would be much easier to correct it before it goes wide release.

Robotaxis are strictly geofenced, only allowed to drive where Tesla has extensively mapped the area and trained the cars. No different with Waymo by the way, they always start a new area with cars driven by humans.

Sponsored

 

[whitex]

As far as Tesla taking responsibilities for FSD at at service centers. Their cars drive off from the assembly plant in Texas, drive off by themselves across public highway, and self park in the delivery parking lot. I'm assuming if it got into a collision along the way, then Tesla would take responsibility.

Yes, they trust it in an extremely well controlled and mapped out environment. Apparently the effort to do the same for Service Centers is not worth the effort. Superchargers would be an even more useful scenario, use the supercharger snake (which Elon announced over a decade ago) and valet cars through superchargers to eliminate manual queuing, parking while idle, etc. Again, the system is not capable of this yet.

 

[Zcd1]

Because it's really expensive. At one point it was 15k. When you can get a Tesla for 35k-45k, paying and additional 15k is a no go. That's a 30-40% increase in price. Honestly I'm surprise that the uptake was 20%. I thought it would be 5% or less.

Yes, I’m aware.

My point was that despite the hype and everyone asking whether my Teslas have had FSD, the vast majority of Teslas don’t have it.

It will be interesting to see whether the subscription-only FSD offering that’s now in effect will move that needle…

 

[69Mach390]

This quote did not mention Musk, but others have and some of those had nothing to do Tesla, but rather his politics. That's why it's mention.

Again, there's a lot that Tesla can improve upon in their cars such as more buttons, better door handles, fancier interior. Etc. But a lot of arguments I hear from people that dislike Tesla are talking points, and has no correlation with reality. Most of these people have no first hand experience with Tesla or FSD.

It’s still a distraction and complete fallacy.

Other people don’t like Musk therefore this particular persons argument against FSD has no merit?

That’s completely illogical and NOT worth mentioning.

Same with the argument you keep making about first hand experience with FSD. I do have first hand experience, but that doesn’t matter. You don’t have to try something in order to be able to form an opinion on it.

Drugs are bad, I haven’t tried them. The Yankees suck, I don’t play baseball.

Learning through observation and other people’s experiences is quite a useful skill.

 

[AutoX]

It’s still a distraction and complete fallacy.

Other people don’t like Musk therefore this particular persons argument against FSD has no merit?

That’s completely illogical and NOT worth mentioning.

Same with the argument you keep making about first hand experience with FSD. I do have first hand experience, but that doesn’t matter. You don’t have to try something in order to be able to form an opinion on it.

Drugs are bad, I haven’t tried them. The Yankees suck, I don’t play baseball.

Learning through observation and other people’s experiences is quite a useful skill.

Sure, everyone can have their own opinions. But that doesn't make it true.

"Yankees suck" But they're in first place.
"FSD is dangerous" But it's 3x less likely to get into a collision than the avg driver.

See, you can have an opinion, but that doesn't mean that it's well informed.

 

[69Mach390]

Sure, everyone can have their own opinions. But that doesn't make it true.

"Yankees suck" But they're in first place.
"FSD is dangerous" But it's 3x less likely to get into a collision than the avg driver.

See, you can have an opinion, but that doesn't mean that it's well informed.

You are just racking up the logical fallacies.

Who said it wasn’t safer than the average driver? It wasn’t me. That’s a straw man argument.

We did say Tesla lied about the safety stats….. thus the whole point of this thread.

When the systems are used properly, they are safer.

I can summary my points if it helps:

I think FSD is actually a great system and makes the roads safer.

My problem is that Tesla LIES about the safety and markets it as “full self driving” which it is NOT.

This gives “the average driver” a false sense of security and they end up abusing the systems (driving drunk, thinking it is self driving, not paying attention, using weights to trick it etc).

The fault of the above is mostly on Tesla. Other manufacturers with similar level 2 systems don’t market the same and we don’t see the same abuse.

And who is at the core of this problem? Yeah, it’s their fearless leader. His narcissism bleeds into Tesla marketing. Root cause analysis…… a good product was made worse by a bad leader. So maybe you were on to something when you said it’s because people “don’t like Musk.”

 

[4424]

It's humorous reading all these posts from people who have never used FSD. Especially people who say they don't have to experience it to know how good/bad it is. Those of us who have it and have used it for thousands of miles objectively know much more about it than those who have watched videos or read articles about it. Keep on hating though...

 

[Vercingetorix]

Whats really humorous is the number of accidents that are being covered up. Hysterical in fact.

 

[4424]

Whats really humorous is the number of accidents that are being covered up. Hysterical in fact.

Based on what you read by people who hate Elon? I hate Elon too but FSD is solid.

 

[FlyingPoint]

Whats really humorous is the number of accidents that are being covered up. Hysterical in fact.

Unfortunately and unlike Waymo, there is no peer reviewed or insurance industry study for Tesla FSD crash statistics. Therefore, the data analysis is left to Tesla and like most companies they will color statistics.

Perhaps a more compelling peer reviewed accident statistic comes from an independent source and its insurance data from the Highway Loss Data Institute (HLDI), which is affiliated with the Insurance Institute for Highway Safety (IIHS).

HLDI found that Tesla vehicles generally have lower collision claim frequencies than the average vehicle in their class. For example:

• The Tesla Model Y had collision claim frequencies roughly 30–40% below comparable vehicles.
• The Tesla Model 3 also performed better than average in insurance claim frequency.
• Injury claim rates were generally below segment averages.

Fatal Crashes:
I could find no comparable independently and consistantly compiled data for fatal crash comparison. In an IseeCar study using NHTSA data, Tesla fatal crashes per registered vehicles is higher than the industry average. However, because this is per vehicle registered and not by total miles driven, it has no comparative value due to Tesla cars being driven more miles than the industry fleet. For example, a Tesla vehicle has a fatal crash with 100K miles driven and a Suburu has a fatal collision with 10K miles, in the ISeeCar study they would be treated as equal equal.

 

[69Mach390]

Unfortunately and unlike Waymo, there is no peer reviewed or insurance industry study for Tesla FSD crash statistics. Therefore, the data analysis is left to Tesla and like most companies they will color statistics.

Perhaps a more compelling peer reviewed accident statistic comes from an independent source and its insurance data from the Highway Loss Data Institute (HLDI), which is affiliated with the Insurance Institute for Highway Safety (IIHS).

HLDI found that Tesla vehicles generally have lower collision claim frequencies than the average vehicle in their class. For example:

• The Tesla Model Y had collision claim frequencies roughly 30–40% below comparable vehicles.
• The Tesla Model 3 also performed better than average in insurance claim frequency.
• Injury claim rates were generally below segment averages.

Fatal Crashes:
I could find no comparable independently and consistantly compiled data for fatal crash comparison. In an IseeCar study using NHTSA data, Tesla fatal crashes per registered vehicles is higher than the industry average. However, because this is per vehicle registered and not by total miles driven, it has no comparative value due to Tesla cars being driven more miles than the industry fleet. For example, a Tesla vehicle has a fatal crash with 100K miles driven and a Suburu has a fatal collision with 10K miles, in the ISeeCar study they would be treated as equal equal.

See? Actual data. 30-40% safer sounds believable. 1000%? Not so much.

I just wish they wouldn’t lie about stats and use misleading marketing to give people a false sense of security. The end result is people have misused the systems which can make them actually more dangerous.

And yeah, people are idiots. It’s why plastic bags have warning labels not to put them on your heads and try to breathe.

 

[FlyingPoint]

See? Actual data. 30-40% safer sounds believable. 1000%? Not so much.

Not so fast, this is Tesla cars compared to industry averages. This study does not differentiate if the Tesla accidents are with FSD engaged or not engaged. I suspect that it will be a long while before FSD safety statistics are compiled, if ever. Tesla would have to turn over ALL its data to an independent peer reviewer. As long as regulators ask for immature data, that is all we will ever have.

The only conclusions that I have come to are:

1. I am being ripped off paying for ALK and Innodrive. In fact, after the NE summer comes to an end, so does my subscription.
2. There are intangibles that will never be measured utilizing FSD in my lifetime. When I am feeling tired or stressed as I drive home from work, FSD is engaged routinely. When I am driving on a highway (day or night), FSD is engaged.
3. I feel safer using it, rather than not using it, especially on dark unfamiliar backroads where I am blinded by oncoming lights and FSD is not.
4. AS Nikki Lauder has said, I can tolerate 20% risk, nothing more. In certain conditions, human driving exceeds my risk tolerance.
5. I fully accept that I am a guinea pig, especially with being signed up for beta testing and advance updates. I have enough confidence in my skillset to overcome any FSD deficiencies I encounter.

Just my simple, uneducated view -- BTW, I am not a Tesla fanboy. I have owned one Tesla for a year now and many P-cars. I never appreciated how technilogically inferior P-cars were until I purchased a Tesla. Truthfully, given the price delta, it is almost criminal.

 

[69Mach390]

Not so fast, this is Tesla cars compared to industry averages. This study does not differentiate if the Tesla accidents are with FSD engaged or not engaged. I suspect that it will be a long while before FSD safety statistics are compiled, if ever. Tesla would have to turn over ALL its data to an independent peer reviewer. As long as regulators ask for immature data, that is all we will ever have.

The only conclusions that I have come to are:

1. I am being ripped off paying for ALK and Innodrive. In fact, after the NE summer comes to an end, so does my subscription.
2. There are intangibles that will never be measured utilizing FSD in my lifetime. When I am feeling tired or stressed as I drive home from work, FSD is engaged routinely. When I am driving on a highway (day or night), FSD is engaged.
3. I feel safer using it, rather than not using it, especially on dark unfamiliar backroads where I am blinded by oncoming lights and FSD is not.
4. AS Nikki Lauder has said, I can tolerate 20% risk, nothing more. In certain conditions, human driving exceeds my risk tolerance.
5. I fully accept that I am a guinea pig, especially with being signed up for beta testing and advance updates. I have enough confidence in my skillset to overcome any FSD deficiencies I encounter.

Just my simple, uneducated view -- BTW, I am not a Tesla fanboy. I have owned one Tesla for a year now and many P-cars. I never appreciated how technilogically inferior P-cars were until I purchased a Tesla. Truthfully, given the price delta, it is almost criminal.

Honestly I trust the “vs all cars” numbers much more than the cherry picked “only when using FSD” numbers.

FSD is going to be used the most in more likely safer conditions in general.

Like my elementary school science teacher taught me decades ago: you can’t draw conclusions if there are too many variables.

And as I learned way back in high school: there is a world of difference between correlation and causation.

Correlation- everyone who drinks water dies.

Causation- everyone who doesn’t drink water dies.

Sponsored