301
General Discussion / Re: Untamed Virus Containment Thread: COVID-19: Super Coronavirus Party Edition
« on: May 27, 2020, 05:43:42 pm »
Assuming the worst case scenario, doing 15 million tests and finding 1.7 million infections is terrifying because it would mean we actually have that sort of ratio of infections/total population.
The good news, of a sort, is that what it really means is that we're testing way too many people who obviously are infected.
That downside of that good news is those numbers are then useless for getting an accurate sense of how far it has spread through the population.
Test everybody, 330 million~ people, find 1.7 million cases, you found everyone and it's under control by definition.
Test 1 in 10, 33 million~ people, find 1.7 million cases, assuming you randomly sampled the population you would want to assume you have 17 million cases floating around, worst case scenario, it's in no way close to controlled.
Test 1 in 20, 16.5 million~ people, find 1.7 million cases, again assuming randomly sampled scenario with no weighting, there might be 34 million cases, which is pretty fucking bad, but we can't even make that sort of guess because we're weighting the testing towards those with obvious symptoms, and these results would only be useful if we were actually being proactive and doing contact tracing to determine possible infections and spread, but we aren't doing that so it's like quickly turning the lights off in a dance club and taking a flash photo of the people around you so you can try to figure out what song is playing.
The good news, of a sort, is that what it really means is that we're testing way too many people who obviously are infected.
That downside of that good news is those numbers are then useless for getting an accurate sense of how far it has spread through the population.
Test everybody, 330 million~ people, find 1.7 million cases, you found everyone and it's under control by definition.
Test 1 in 10, 33 million~ people, find 1.7 million cases, assuming you randomly sampled the population you would want to assume you have 17 million cases floating around, worst case scenario, it's in no way close to controlled.
Test 1 in 20, 16.5 million~ people, find 1.7 million cases, again assuming randomly sampled scenario with no weighting, there might be 34 million cases, which is pretty fucking bad, but we can't even make that sort of guess because we're weighting the testing towards those with obvious symptoms, and these results would only be useful if we were actually being proactive and doing contact tracing to determine possible infections and spread, but we aren't doing that so it's like quickly turning the lights off in a dance club and taking a flash photo of the people around you so you can try to figure out what song is playing.