3 ms·

Explain to me how this comparison makes sense.

londons_explore 7y ago
Tomorrow, my chance of dying from COVID-19 is lower than my chance of dying in a car crash. In 3 weeks time, those probabilities might have changed. My behaviour today should reflect probabilities today, and my behaviour in 3 weeks should reflect probabilities then. My behaviour today has negligible impact on my chances in 3 weeks too. Collective behaviour might effect collective probabilities in 3 weeks, but even that is relatively minimal considering the nature of exponentials.
- jjeaff 7y ago
  So, since -my- risk of catching measles is very low... I shouldn't get immunized, right?
- geofft 7y ago
  Your behavior today has serious and direct impact on not only your chances in three weeks but on the lives of other people around you. A cluster of sick people in South Korea is traced to one person who attended church two weeks ago. So, no, this isn't about collective probabilities - this is about what you personally do. In any case, Elon Musk doesn't speak for individual behavior, he controls a company of thousands of employees. That's already collective behavior.
- spiderfarmer 7y ago
  Now, with this many asterisks, is it really a good and valid comparison?
- riversflow 7y ago
  >considering the nature of exponentials. That’s the thing that makes this analysis nonsensical. You are looking at the first 2 months of an exponential and treating it like a fixed rate. You should treat it as an exponential and extrapolate another 10 months with number of cases doubling every 4-6 days and then use that number as the deaths per year. I’m honestly blown away that Musk is saying this.
  - smaccona 7y ago
    To put this in perspective, especially since it’s difficult for humans to intuitively appreciate exponential growth, and assuming the conservative 6 days doubling rate, and starting with today’s US death count of 61 (2020-MAR-15): - on 8 Apr US will be just under 1000 deaths. Scary, but that’s almost a month away - on 2 May, deaths will be ~15,000 - on 20 May, deaths will be ~125,000 - on 7 June, deaths will be ~1,000,000 - on 1 July, deaths will be ~16,000,000 - on 19 July, deaths will be ~128,000,000 - by end of 2020, deaths would be over 17,000 trillion if the US population was infinite and mitigating factors such as herd immunity didn’t kick in, but you get the general idea This is all presuming the existing political/societal response and herd immunity not kicking in (which it won’t in the US until around the 200MM mark). Social distancing will lessen this massively, which is why it’s so important.

Tomorrow, my chance of dying from COVID-19 is lower than my chance of dying in a car crash. In 3 weeks time, those probabilities might have changed. My behaviour today should reflect probabilities today, and my behaviour in 3 weeks should reflect probabilities then. My behaviour today has negligible impact on my chances in 3 weeks too. Collective behaviour might effect collective probabilities in 3 weeks, but even that is relatively minimal considering the nature of exponentials.

jjeaff 7y ago
So, since -my- risk of catching measles is very low... I shouldn't get immunized, right?
geofft 7y ago
Your behavior today has serious and direct impact on not only your chances in three weeks but on the lives of other people around you. A cluster of sick people in South Korea is traced to one person who attended church two weeks ago. So, no, this isn't about collective probabilities - this is about what you personally do. In any case, Elon Musk doesn't speak for individual behavior, he controls a company of thousands of employees. That's already collective behavior.
spiderfarmer 7y ago
Now, with this many asterisks, is it really a good and valid comparison?
riversflow 7y ago
>considering the nature of exponentials. That’s the thing that makes this analysis nonsensical. You are looking at the first 2 months of an exponential and treating it like a fixed rate. You should treat it as an exponential and extrapolate another 10 months with number of cases doubling every 4-6 days and then use that number as the deaths per year. I’m honestly blown away that Musk is saying this.
- smaccona 7y ago
  To put this in perspective, especially since it’s difficult for humans to intuitively appreciate exponential growth, and assuming the conservative 6 days doubling rate, and starting with today’s US death count of 61 (2020-MAR-15): - on 8 Apr US will be just under 1000 deaths. Scary, but that’s almost a month away - on 2 May, deaths will be ~15,000 - on 20 May, deaths will be ~125,000 - on 7 June, deaths will be ~1,000,000 - on 1 July, deaths will be ~16,000,000 - on 19 July, deaths will be ~128,000,000 - by end of 2020, deaths would be over 17,000 trillion if the US population was infinite and mitigating factors such as herd immunity didn’t kick in, but you get the general idea This is all presuming the existing political/societal response and herd immunity not kicking in (which it won’t in the US until around the 200MM mark). Social distancing will lessen this massively, which is why it’s so important.