Orders of magnitude really don't gloss over huge confidence intervals though, that's just attempting to game the stats.

I get what you're going for and anecdotally that number doesn't even seem crazy if you're going back through our entire history of coal. At the end of the day that number simply didn't exist though, and any time someone uses a start like that to bank up an argument it's just propaganda. Having a bunch that number should be accurate is very different from being able to use it as part of a larger discussion of what the biggest problems are and how to improve it.