4 ms·

I think that would be a statistical mistake to cherry-pick the cutoff like that. What would be the argument for ignoring the X months/years beforehand in which

by slg 2y ago

I think that would be a statistical mistake to cherry-pick the cutoff like that. What would be the argument for ignoring the X months/years beforehand in which no one died? If I asked you for the odds of the next three coin tosses being heads, you don't start counting on the first heads. The first toss being tails is a possible outcome that can't be ignored.

rootusrootus 2y ago
That is an excellent point. Maybe the math has to be using the average length of time the whistleblowers have been 'out' as it were. That could plausibly end up being several years, which pushes the 1:32 number a lot closer to 'certain' and the 2:32 number way up into entirely plausible.