3 ms·
> In the US, we have had a sustained 400-500 dying a day. Are you saying this is still currently happening? If so, I'm curious where you're getting this number
by _gabe_ 4y ago
> In the US, we have had a sustained 400-500 dying a day.
Are you saying this is still currently happening? If so, I'm curious where you're getting this number from. The CDC says the current 21 day rolling average is 358 deaths[0].
Edit: I just wanted to add, if this is the case then the deaths per 100,000 is approximately 358/300,000,000*100,000 = 0.1193 over that 21 day period. If I did those calculations correctly and the statistics don't fail if I try to calculate a deaths per 100K based off a rolling average, then this means you're about 100 times more likely to die from a car crash[1] then covid right now.
[0]: https://www.cdc.gov/coronavirus/2019-ncov/covid-data/covidview/index.html https://www.cdc.gov/coronavirus/2019-ncov/covid-data/covidvi...
[1]: https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813209 https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/...
- jcpst 4y agoSorry, I’m not seeing how you figured that. Ok, so 2566 weekly deaths from covid: https://covid.cdc.gov/covid-data-tracker/#datatracker-home https://covid.cdc.gov/covid-data-tracker/#datatracker-home That’s 366 per day currently. The crash stats you linked to say 36,096 deaths from crashes in 2019. 366 x 365 = 133,798 So what I see is there’s currently 3.7 times as many people dying from covid in US that from crashes.
- wizofaus 4y agoYou can even just ask Google how many road deaths a day there are in the US, and it responds with a sentence from wikipedia including the phrase "an average of 102 per day".