[helen___keller]

> Even people who don't have a statistical understanding of power law vs. normal distribution usually have a good intuition of it, which is why most people worry about terrorism more than one would if they were just looking at just the total number of deaths vs. something car accidents or smoking.

I think this is a poor example, by any measure terrorism is an irrational concern for Americans, it's nothing like a pandemic that could actually affect the whole country.

[tharne]

Not irrational, since, like pandemics, terrorism involves a lot of "unknown unknowns". You're thinking past terrorism attacks == future terrorism attacks, which is the wrong way to look at it. It's not a linear concept like auto accidents. Deaths from terrorism can jump around wildly from year to year. You can have zero deaths from terrorism one year and one million the next. That's how certain phenomena work. Others are linear and it's important not to confuse the two.

[jerf]

Indeed, while I haven't seen anyone even theorize that COVID-19 is a terrorist attack in the darkest rumors I've seen, it's the sort of thing you need to worry about if you let a terrorist organization ever got far enough to produce one of these things, which is a completely realistic concern.

To my eyes, COVID-19 looks like a single source that is spreading like wildfire, but, imagine this was a deliberate attack. Things spread fast enough on their own; consider how fast this would be spread if there was an organization doing it on purpose. What if this basically appeared in every major US city simultaneously?