I don't mean to downplay; clearly and surely this should be investigated. But actually, I typo'd quite badly: the ratio is 1 in 700000, not 1 in 15000 (how did

I don't mean to downplay; clearly and surely this should be investigated. But actually, I typo'd quite badly: the ratio is 1 in 700000, not 1 in 15000 (how did I make that mistake?) Anyways, I hope it's clear that a 1 in 700000 event is hard to detect.