An exclusive in-depth account of the most recent Chevy Bolt EV fire, plus what you can do to stay safe

MikeyBolt said:

That would certainly help but I'm not holding my breath for that to occur. They seemed to think the temp fix was enough to prevent fires while they worked on the perma-fix so saying they know with 100% certainty that the perma-fix would have prevented this latest fire is also a sort of admission that they (at least by now) know that the temp fix was inadequate and could still result in fires. And the other side of the coin, if they were not aware the temp fix wasn't good enough, is an admission that there are still things they don't understand about these fires. It's a lose-lose for them either way.

Mike

Click to expand...

One of the new recall docs is the dealer service bulletin:
https://static.nhtsa.gov/odi/rcl/2020/RCSB-20V701-5306.pdf

On page 6, after a defect has been diagnosed, there's a warning that makes me wonder if the temporary fix would really prevent a fire:

While waiting for a replacement battery cell module row to arrive, do not charge the vehicle or drive more than a half mile.

Click to expand...

Bad news: If having the defect means it's unsafe to drive half a mile or charge at all, then limiting the charge to 90% wouldn't be sufficient to guarantee safety, unless charging to 100% actually causes the defect to manifest in some way.

"Good" news: This could explain the latest fire even if the temporary recall had been installed, and also means it's at least possible that GM's final fix would still be effective. Unlike the temporary fix, the final procedure includes a process to diagnose the defect, replace affected battery modules, and install new diagnostic software that could prevent the car from being driven or charged at all, just like the warning in the service bulletin.

Edit: Part of me feels like GM is having a Pinto moment here. If limiting charge makes it more safe, but doesn't guarantee that a fire can't happen, then it feels like someone has done the risk calculation of the number of fire lawsuits they'd have to pay for vs the cost of taking all affected Bolts off the road immediately (buy back, loaners, whatever).

desiv said:

I'm sure the 7 people whose cars caught fire feel much better knowing that.
And they probably have never even flipped a coin to heads 13 times in a row!!!

I had a stats teacher who said you flip a coin 49 times and it is heads every time, what are the odds that for the 50th?
The class said "50/50."
He said "No way, you better believe there is something wrong with that coin!!!

Click to expand...

The point is there is a difference between how policy-makers need to assess risk, and how individuals need to assess risk. You asked a specific question about how worried you should be. The answer is, not much at all. That's not the same thing as saying NHTSA or GM should not be worried - they absolutely should be worried. The difference is that NHTSA and GM are dealing with large numbers (flipping 59,000 coins), while you are dealing with a very small number (flipping 1 coin).

Put another way, if 59,000 people flip a coin 13 times, roughly 7 of them will get all heads. If you flip a coin 13 times, you will almost certainly NOT get all heads. So NHTSA and GM need to act because there are 59,000 affected Bolts out there, and there have been 7 fires. How you choose to act should be based on your personal assessment of the likelihood of getting 13 heads in a row yourself, as an individual.

As I mentioned earlier, people have a difficult time assessing very small probabilities of very large outcomes. The large outcome - either good or bad - distorts our perception of risk. Trying to place the risk in the context of other activities with similar probabilities, like flipping a coin, can help.

And all that being said, I'm still planning to buy a MegaMillions ticket before Tuesday.

desiv said:

But that isn't how statistics actually works and you aren't counting unknown variables.
If it has happened 7 times so far out of 59,000 vehicles, that doesn't mean the odds are 7 out of 59,000.

Click to expand...

I'll freely admit I'm not a statistician. I am planning to buy a lottery ticket, after all.

That being said, I'm looking at this like epidemiology - 7 cases out of 59,000 possible cases. The advantage of epidemiology is that you don't have to define all the pathways and causes - you look at the overall outcome. Having said that, with such a low number of cases, you are correct in that we can't say that the probability is exactly 7 out of 59,000 with a high degree of confidence. When something occurs rarely (like a rare cancer), the confidence interval will be large. However, it's still a rare cancer.

And that isn't the end either. What if it turns out that it is also tied to how people have charged? Then, if your charging pattern matches the triggers, then your odds will be much higher.

Click to expand...

I think that's a fair concern, but I don't think your odds will be much higher, because it's still actually dependent on you having the defect combined with having a rare charging behavior.

In the NHTSA recall documents, GM / LG have estimated about 1% of the vehicles have the battery defect. That's roughly 590 Bolts. So right there, you have to assume you got one of the unlucky 590 out of 59,000 Bolts. But we haven't seen 590 fires - we've only seen 7. Let's pretend that a particular charging behavior, in combination with the defect, caused the 7 fires. That means those 7 owners did something that the other 583 owners did not - their charging behavior is something only 1 out of 84 owners do.

So if you want to reduce your risk, charge the way most Bolt owners charge. Don't do weird things that only 1 out of 84 Bolt owners do.

If your charging behavior is rare - like 1 out of 84 rare, AND you have the rare defect, like 1 out of 100 rare, then you may get a fire. But your overall risk - the combination of those factors - is still low, like a rare cancer.

Edit:

p7wang said:

Odds of winning jackpot is tiny and so is odds of fire. I guess one should go buy a lotto ticket. If one were lucky, one would win both. Car fire would be covered by insurance. Lotto win is yours to keep.

Click to expand...

To be fair, the probability of a Bolt fire is much, much higher than the probability of winning the lottery. If the risk was literally like winning the lottery, there wouldn't be a recall at all.

This is a serious problem that NHTSA and GM absolutely should address, just like the medical community absolutely should address rare cancers. It's just that as an individual, we shouldn't go crazy worrying about our personal risk of a Bolt fire, just like we shouldn't worry about getting a rare cancer. Take some common sense measures - like those recommended by GM and your doctor - and let yourself go to sleep at night.