3 ms·

What "1000 person test case"? Every thing about what is actually happening in many places is the reverse if what you're saying.

whb07 7y ago
Where’s many places ? Let me guess you’ll say Italy and err Iran? You know there’s like 200 other countries. Furthermore, there are 2 billion+ people living an hour or two away from China. Why aren’t they saying it’s end of days. Please don’t just respond with a “simple they aren’t testing”. You don’t need a test for a massive spike in deaths.
- bpeebles 7y ago
  The specific question I asked was about the test case you mentioned since I couldn't find it with a quick search. Also, I haven't really seen anyone saying it's the end of days, but more to use the US as an example, we're trying to reduce the likely death toll of millions (330M). It's heartening that it looks there's evidence that's some natural maximum coming out of China, but some US hospitals are already close to their capacity so we're trying to avoid a medical system breakdown here. From another comment, you said why "aren’t all dead" in existing regions, when literally no one has ever said they'd all die. The most pessimistic studies I see are like, 50% of population gets it and ~5% die after the hospitals run out of capacity (and another 10% have a really, really scary few weeks), so 2.5% of the population
  - whb07 7y ago
    The Diamond Princess cruise. They were a major cluster, and coincidentally got forced quarantined and turns out to be a natural test case. Notice how we only focus on the bank run on US hospitals. Are hospitals in poorer countries immune to these type of sudden rushes? Why aren’t we seeing millions of people in neighboring countries to China clogging up hospitals? Is it slightly possible this is just a tad over done? What would you say are the chances of that?
    - scottlocklin 7y ago
      People are slowly noticing the use of the WHO CFR as an actual FR is science fiction[1] at least. Well, they're not really noticing that, especially here on HN, because people are statistical imbeciles, but at least they're noticing that the denominator in the FR is a lot larger than they previously thought. That was abundantly obvious at the beginning; if you aren't widely administering a costly and rare test[2], you're enormously left censored in your estimator. You're also radically screwing up your R0 estimate. I suppose there are arguments to be made that more hot and humid climates prevent the spread of respiratory viruses in general, which is why you don't see problems in the more humid and hot regions of China and Vietnam.[3] The real question is why some countries look like Germany, and some look like Italy. The paranoid answer is Germany will soon look like Italy (and this might be true), except the claimed origin of the disease is .... Germany[4]. If it does turn out to be overblown in some sense, it's probably a valuable exercise in autarky. [1] https://www.nytimes.com/2020/03/16/health/coronavirus-statistics-undetected.html https://www.nytimes.com/2020/03/16/health/coronavirus-statis... [2] And FWIIW none of the tests are well calibrated; in fact there are numerous tests in use with radically different biases. [3] https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3551767 https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3551767 [4] https://www.reuters.com/article/us-health-coronavirus-italy-scientists/coronavirus-may-have-reached-italy-from-germany-scientists-say-idUSKBN20Y35B https://www.reuters.com/article/us-health-coronavirus-italy-... (I mean, it seems unlikely, but it's a common assertion)
  - whb07 7y ago
    Also 2% of 2 billion is 40 million dead. We are 3 months+ since this outbreak started. Everyone keeps talking about how exponentials are hard to grasp, but whats harder to grasp is: fear + exponentials + too_many_unknowns == impossible to plot A feature that everyone talks about is how startups grow exponentially, fine, i get it, but most end up failing and going to zero. What if this was such a case? We ascribe to it an exponential growth path and yet it fails to take off for many reasons like perhaps a dog walking service or WeWork isn't the cat's pajamas?

whb07 7y ago

Where’s many places ? Let me guess you’ll say Italy and err Iran? You know there’s like 200 other countries. Furthermore, there are 2 billion+ people living an hour or two away from China. Why aren’t they saying it’s end of days. Please don’t just respond with a “simple they aren’t testing”. You don’t need a test for a massive spike in deaths.

bpeebles 7y ago
The specific question I asked was about the test case you mentioned since I couldn't find it with a quick search. Also, I haven't really seen anyone saying it's the end of days, but more to use the US as an example, we're trying to reduce the likely death toll of millions (330M). It's heartening that it looks there's evidence that's some natural maximum coming out of China, but some US hospitals are already close to their capacity so we're trying to avoid a medical system breakdown here. From another comment, you said why "aren’t all dead" in existing regions, when literally no one has ever said they'd all die. The most pessimistic studies I see are like, 50% of population gets it and ~5% die after the hospitals run out of capacity (and another 10% have a really, really scary few weeks), so 2.5% of the population
- whb07 7y ago
  The Diamond Princess cruise. They were a major cluster, and coincidentally got forced quarantined and turns out to be a natural test case. Notice how we only focus on the bank run on US hospitals. Are hospitals in poorer countries immune to these type of sudden rushes? Why aren’t we seeing millions of people in neighboring countries to China clogging up hospitals? Is it slightly possible this is just a tad over done? What would you say are the chances of that?
  - scottlocklin 7y ago
    People are slowly noticing the use of the WHO CFR as an actual FR is science fiction[1] at least. Well, they're not really noticing that, especially here on HN, because people are statistical imbeciles, but at least they're noticing that the denominator in the FR is a lot larger than they previously thought. That was abundantly obvious at the beginning; if you aren't widely administering a costly and rare test[2], you're enormously left censored in your estimator. You're also radically screwing up your R0 estimate. I suppose there are arguments to be made that more hot and humid climates prevent the spread of respiratory viruses in general, which is why you don't see problems in the more humid and hot regions of China and Vietnam.[3] The real question is why some countries look like Germany, and some look like Italy. The paranoid answer is Germany will soon look like Italy (and this might be true), except the claimed origin of the disease is .... Germany[4]. If it does turn out to be overblown in some sense, it's probably a valuable exercise in autarky. [1] https://www.nytimes.com/2020/03/16/health/coronavirus-statistics-undetected.html https://www.nytimes.com/2020/03/16/health/coronavirus-statis... [2] And FWIIW none of the tests are well calibrated; in fact there are numerous tests in use with radically different biases. [3] https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3551767 https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3551767 [4] https://www.reuters.com/article/us-health-coronavirus-italy-scientists/coronavirus-may-have-reached-italy-from-germany-scientists-say-idUSKBN20Y35B https://www.reuters.com/article/us-health-coronavirus-italy-... (I mean, it seems unlikely, but it's a common assertion)
- whb07 7y ago
  Also 2% of 2 billion is 40 million dead. We are 3 months+ since this outbreak started. Everyone keeps talking about how exponentials are hard to grasp, but whats harder to grasp is: fear + exponentials + too_many_unknowns == impossible to plot A feature that everyone talks about is how startups grow exponentially, fine, i get it, but most end up failing and going to zero. What if this was such a case? We ascribe to it an exponential growth path and yet it fails to take off for many reasons like perhaps a dog walking service or WeWork isn't the cat's pajamas?