On the other hand, IIRC nuclear has a reverse bell curve, where it's most dangerous at the very start and end of its' lifetime. And if you take nuclear plants t

On the other hand, IIRC nuclear has a reverse bell curve, where it's most dangerous at the very start and end of its' lifetime. And if you take nuclear plants that are not quite at the end of their lifetime, then average the deaths over their last 40 years, and compare that average to solar (the majority of which are 3 years old or less - at the start of their lifetime), then you can make nuclear appear an order of magnitude safer than they actually are. People don't handle statistics well, but statistics are also easily manipulated and therefore potentially completely unreliable (Mark Twain quote here). It's not quite as simple as "people need to listen to the statistics". Also, people go to the beach way less often than they go near roads. If most people only go to the beach 1 day a year and go on roads every day, then it's more like "you're 50x more likely to die by car than by shark". That's a far cry from 17000x. That said, I've always thought that stat could be abused to push car safety - "Imagine you're attacked by a shark - now, cars are like that but TIMES FIFTY! Drive safe, look both ways, etc".