2 ms·

Setting aside all the other evidence for and against the lab-leak hypothesis, I gotta argue with your numbers. Humans are not randomly distributed over the land

Setting aside all the other evidence for and against the lab-leak hypothesis, I gotta argue with your numbers. Humans are not randomly distributed over the land-surface of the earth. Neither are virology labs. The first known starting point of a disease outbreak will almost definitely be in a densely populated area, since that's basically a requirement for an outbreak. So taking that into account we have to correlate densely populated areas with BSL4 labs. For a back-of-the-envelope estimate, let's use the assumption: > "About 90% of the earth’s people live on only 10% of the land." And let's assume that all of those 59 labs are in that 10% as well. I'm sure it's actually smaller than that - since not all of that 10% of the land has enough density to be the epicenter of an outbreak, but... let's roll with that 10% That'd put you off by a factor of 10. 14^2 * pi * 59 = 36,329 km2 148,326,000 * 10% = 14,832,600 36,329 / 14,832,600 = 0.00245 aka 0.2% On top of that there have been multiple tries at this - in recent memory we've seen SARS classic, Covid-19, MERS-CoV, and Swine Flu. That's four chances at a target at least as big as 0.2%. My probabilities are rusty, but I believe that's 99.8% * 99.8% * 99.8% * 99.8% = 99.2% chance of none of those appearing near a lab or an 0.8% chance it was near a lab. If we take that into account, it starts to seem a bit more likely that we'd see something like this... And let's not even touch the > 1300 BLS3 labs that are out there.

mandmandam 5y ago
With all respect, how do you think this proves your point at all? Do you not find it absurd in the extreme that the novel Coronavirus lab a few km from the outbreak of a novel coronavirus was barred from investigation? Is it not a giant raging blinking neon clue that when Australia wanted an investigation China threatened them with economic apocalypse? And in this context, you think a .8% chance that it would appear near a lab is high enough to discount the possibility??? My head is exploding.
- kennywinker 5y ago
  > With all respect, how do you think this proves your point at all? I wasn’t trying to prove a point, I was trying to dis-prove a point: that you can’t prove the lab leaks based on geography alone. 0.02% chance is VERY unlikely - it’s not proof but it sure does suggest. A 0.2% chance or a 0.8% chance is… well it’s still rare, but not shockingly rare. Again I suspect the geographic odds are even higher, if you properly correlate outbreak-size population density and labs. Bsl3 labs brings the odds closer to 5% (or 19% if you count the four other novel disease incidents and I’m doing my probability math right). Again, none of this disproves a lab leak. But it sure does tell us that we can’t prove it with geography, as the post I was responding to suggested.

mandmandam 5y ago

With all respect, how do you think this proves your point at all? Do you not find it absurd in the extreme that the novel Coronavirus lab a few km from the outbreak of a novel coronavirus was barred from investigation? Is it not a giant raging blinking neon clue that when Australia wanted an investigation China threatened them with economic apocalypse? And in this context, you think a .8% chance that it would appear near a lab is high enough to discount the possibility??? My head is exploding.

kennywinker 5y ago
> With all respect, how do you think this proves your point at all? I wasn’t trying to prove a point, I was trying to dis-prove a point: that you can’t prove the lab leaks based on geography alone. 0.02% chance is VERY unlikely - it’s not proof but it sure does suggest. A 0.2% chance or a 0.8% chance is… well it’s still rare, but not shockingly rare. Again I suspect the geographic odds are even higher, if you properly correlate outbreak-size population density and labs. Bsl3 labs brings the odds closer to 5% (or 19% if you count the four other novel disease incidents and I’m doing my probability math right). Again, none of this disproves a lab leak. But it sure does tell us that we can’t prove it with geography, as the post I was responding to suggested.