Tesla just had their third car fire in a 6 week period. The data isn’t really volumous enough to tell yet, and so it could just be rather bad luck. However I did want to play Texas Sharpshooter and point something out.
According to the NFPA, in 2010 there were 33 vehicle fires per hour. Assuming about 250 million vehicles on the road, that equates to a .0133% chance of being involved in a car fire over a 6 week period.
Comparatively, over the last 6 weeks, with approximately 16,000 Model S on the road, there have been 3 vehicle fires. This equates, to a .0188% chance of being involved in a vehicle fire. Over the last 6 weeks, it was statistically more likely to get in a fire while driving the world’s safest car.
Is this just a statistical anomaly of no consequence?
To test, I doubled the time period to see how sensitive the outcome is. In a 12 week period, the likelihood of being involved in a car fire in a standard ICE vehicle is .027%.
Over the last 12 weeks, there have still been only 3 Model S fires. However, there also were 5,500 less Model S’ on the road 12 weeks ago. When you hold the 3 fires but reduce the number of vehicles, I’m getting a .029% chance of being involved in a vehicle fire with a Model S, still higher than the ICE vehicle…
So double the time period to 24 weeks; you get .053% odds of being in an ICE vehicle fire, to .055% odds of being in a Model S fire. The lower number of Model S on the road keeps that number above the ICE. Again, Tesla loses, at least for now.
The trouble here again is the small data, and Musk could just be experiencing the worst luck in the world. However, at this point in time it’s worth noting we already have 8,000 vehicle years of data to look at. Depending on what you think the life expectancy is of one of these vehicles, that’s between 200-500 vehicle lifetimes.
I’m assuming the risk of a critical fire (being a pretty extreme event, not like regular wear and tear) stays about uniform throughout the life of the vehicle, not getting better or worse with age.
Now that’s not as good as say, 1,000 lifetimes of data. But because we’re so early in the Model S production schedule, keep in mind that every four months the number in percent of these vehicles on the road grows by leaps and bounds. After the next 3 months, we should already have between 500-1,000 lifetimes of data.
So maybe Musk is just really unlucky and the number settles back down with time. The difference between the rate soaring or dropping like a stone is as little as one major accident. Consider if we had stayed at 2 accidents, the Model S would easily be much safer.
However, from a vehicle usage standpoint, we are increasingly seeing plenty of data to make a judgement about accidents of the Model S in the day to day hands of consumers.
So the discrete nature of the accidents is still making this murky. But because we have so many Model S on the road, the impact on the standard deviation of the marginal accident diminishes pretty quickly. We could have fairly concrete data on the probability of a fire in a Model S as early as the fifth or maybe tenth fire.
The number of Model S’ on the road is going up, not down, so increasingly these vehicles are more likely to be seen in accidents. In another year or two, Musk’s claims of vehicle safety could get reconfirmed by a relatively safe lull. Maybe his car is just being bought by a rather more reckless segment of the population – unsafe first adopters – or bad chance? But at least as of this point in time, 9 months in, it really doesn’t appear that the Model S is all that much safer, from a fire standpoint, than a car full of flamable liquid. In fact, if things don’t turn around, it could be judged as worse.
If you enjoy the content at iBankCoin, please follow us on Twitter
Mr. Cain ‘Texas Sharpshooter’ Thaler
hahaha
It seems like these fires occur when there are significant impacts. Since there were no reports of fires from the car overheating, which is the case with other car fires, and since the IIHS does not test at most highway collision speeds, I think a safer assumption would be that Tesla’s batteries probably cannot handle impacts in the 60mph+ speed.
Which would be a travesty to the shares.
I intend to avoid leaping to conclusions, however, and will keep an eye on the empirical data
I’m still waiting for the first electrocution death as a result of these cars. Some of my family is mechanics and they all hate hybrid cars because the risk of electrocuting and killing yourself is significantly higher than any other models
the time is almost right to buy TSLA shares.
Tesla fires are a popular story, but the moral i’m getting is that the car is well designed to protect human life.
reading over financials or setting up blogs and writing reports, getting dirt on a company takes too much time and effort
short the stock, pour some gas and burn their cars..
disclosure: I would never do that and its just a joke
Very Funny Blog! When the S ran over a trailer hitch and it went into a ball of fire, that would make one think you must avoid bumps as well as trees in the Tesla. Still, the stock will recover.