Ugh. Another Model S fire - 2013-11-06

[MikeL]

Thanks doug, I'll perk up for that, and wake me up when Tesla's fire-per-million-miles-driven number is more than a small fraction of the ICE rate. It still isn't.

 

[GDH]

Awesome

 

[doug]

Is it suspicious that we go about a year without a single fire issue, then there are three within a month of each other?

Doug mentioned that too. I do think it's kind of weird that it's all bunched up like this, but maybe it's just bad luck.

I didn't say it was suspicious. I said it was "not good to have them clustered like this." I meant this in terms of people's perceptions. Clustering isn't that uncommon in statistics, and the fact that there are more cars out there now make it more likely.

We'll find out more soon.

 

[DaveT]

Here's an official statement from Tesla:

We have been in contact with the driver, who was not injured and believes the car saved his life. Our team is on its way to Tennessee to learn more about what happened. We will provide more information when we’re able to do so.

Thanks Doug. Great to hear that Tesla is on top of things and a team is on its way.

 

[ckessel]

Thanks doug, I'll perk up for that, and wake me up when Teslas fire-per-million-miles-driven number is more than a small fraction of the ICE rate. It isn't yet.

Well, after the first fire Elon noted you're still 1/5th as likely to have a fire in a Tesla based on miles driven. 2 accidents later we're at 3/5ths, I think that'd qualify as more than a "small fraction". And if you restricted it to fires in cars less than 1 year old, I think that fraction wouldn't be in our favor anymore.

Probably the skewing effect of small sample sets, but the fires per miles driven stat isn't something we can tout much at this point.

 

[SFOTurtle]

doesnt this pic show an injury on both sides of the rear tire?

The picture of the rear of the car shows some kind of collision, at least behind the right rear tire and on the rear fender.

 

[Shumdit]

Well, after the first fire Elon noted you're still 1/5th as likely to have a fire in a Tesla based on miles driven. 2 accidents later we're at 3/5ths, I think that'd qualify as more than a "small fraction". And if you restricted it to fires in cars less than 1 year old, I think that fraction wouldn't be in our favor anymore.

Probably the skewing effect of small sample sets, but the fires per miles driven stat isn't something we can tout much at this point.

I am putting as many miles on mine as I can to help that ratio!!!

 

[Discoducky]

Here's an official statement from Tesla:

We have been in contact with the driver, who was not injured and believes the car saved his life. Our team is on its way to Tennessee to learn more about what happened. We will provide more information when we’re able to do so.

TM is awesome!

 

[Shumdit]

TM is awesome!

I want to agree, but I think it's more likely that they know what this can do to their image/stock price and could trigger an NHTSA investigation so they are trying to diffuse the issue as much as they can. I REALLY hope they find a good reason for this one since I think both other ones were pretty much not the fault of the car.

 

[doug]

I've made the second post in this thread a wiki: Ugh. Another Model S fire - 2013-11-06

Please help keep it updated with facts as we learn them.

 

[hershey101]

Why do you keep saying this? Two of the three are not possibly P85+ cars based on the facts presented yet you have jumped to some crazy conclusion that the P85+ is the problem and needs to be recalled. I would ask that you retract your statement.

I didn't see those posts until after I had posted my statement. I agree, probably not a P85+ exclusive issue.... Still it has to have me wondering... Did they switch suppliers for some part? I don't think they make the 12v battery in-house do they?
I just find it difficult to believe its a coincidence. 20K cars on the road, 3 on fire ~= 0.015% a pretty low probability, but idk how it compares to other cars... How many Model S accidents have there been? 20? 30? 50? I have seen a bunch of reports, but if say even 3% of all accidents result in fires, thats pretty bad even if the car does save the person's life and no one has any injuries...

 

[eepic]

If it is due to an accident, the way I see it it's +2 for the driver

1) Safety, the occupant unharmed in accident
2) Insurance covers the car, so now he can get a new one with power folding mirrors

 

[MikeL]

Well, after the first fire Elon noted you're still 1/5th as likely to have a fire in a Tesla based on miles driven. 2 accidents later we're at 3/5ths, I think that'd qualify as more than a "small fraction". And if you restricted it to fires in cars less than 1 year old, I think that fraction wouldn't be in our favor anymore.

Probably the skewing effect of small sample sets, but the fires per miles driven stat isn't something we can tout much at this point.

I thought it was 1/15 (just from memory) so yes, at 1/5 it's not a small fraction, but even 2/5 (will not count Mexico) is still a good chunk less than 1.

 

[hershey101]

If it is due to an accident, the way I see it it's +2 for the driver

1) Safety, the occupant unharmed in accident
2) Insurance covers the car, so now he can get a new one with power folding mirrors

New ones have power folding mirrors? Hmm... Here is another consperecy theory... All old Model S drivers are pissed off that their mirrors don't fold and are crashing their cars on purpose, knowing fully well that they won't be harmed because of how amazing Tesla cars are, just so they can get new ones with the power folding mirrors... That gives me an idea j/king

 

[ckessel]

but even 2/5 (will not count Mexico) is still a fraction

Why doesn't Mexico count? If it'd been an ICE and caught fire, would that similarly get a pass as counting as a fire?

 

[tommolog]

Tommolog - Where did you find that first picture? It's not an Instagram photo... (Instagrams are square!)

You are right. There were other pictures of it I was looking at on Facebook and one linked to this one which was from Twitter:
Twitter / NASHVILLAIN_: I-24 #tesla #teslafire ...

 

[stopcrazypp]

Why doesn't Mexico count? If it'd been an ICE and caught fire, would that similarly get a pass as counting as a fire?

Mexico doesn't count because IIRC the statistic only counts US fires, not worldwide.

 

[Todd Burch]

ICE cars have one fire per ~20M miles driven. This isn't a super-useful statistic, unfortunately.

Model S is a performance car. They tend to be driven more aggressively, and therefore tend to be involved in more energetic accidents. This tends to skew statistics to make the Model S look better, as (I am assuming) performance cars have a higher fire-per-mile-driven rate than when considering all ICEs as a whole. (Not sure if those statistics are even available--again only an assumption).

On the other hand, Model S is a new car, which would tend to skew statistics to make it look worse: the fire-per-mile-driven rate is probably less for new cars than for all ICE cars as a whole.

So while comparing the S to the ICE fire rate is pretty much the only thing that can be done since other statistics (to my knowledge) are not available, it's a bit of an incomplete picture.

 

[joefee]

Elon needs to give a simple example: USA Model S fires 11/6 = 1, injuries = 0 :: ICE USA fires = x, injuries = y

 

[ckessel]

Mexico doesn't count because IIRC the statistic only counts US fires, not worldwide.

That's really splitting hairs. It's presumably a US car (since they're not shipping to Mexico) a bit over the boarder on a vacation trip. I'm pretty sure Tesla counts all miles driven...not just those in the US.