Can someone please explain the math on this? This is taken directly from article:
There were approximately 781,?000 total deaths in the United States from March 1 to May 30, 2020, representing 122?300 more deaths than would typically be expected at that time of year. There were 95?235 reported deaths officially attributed to COVID-19 from March 1 to May 30, 2020.
So, if there were 781,000 deaths between March 1 - May 30, and this is 122,300 more deaths than expected, then the expected deaths during this time period would be 658,700. Let's compare this number to historical figures.
The most recent year I found for US deaths was 2017. During this year 2,813,500 people died. Now I can't fine the 2018/2019 numbers, but the numbers have been slightly steadily increasing every year for several years. So even though 2020 would undoubtedly be predicted to have more deaths than 2017, we will use the 2017 numbers since they are conservative and we have a month by moth breakdown.
Here is a month by month breakdown of 2017 deaths:
As it states, the average daily death count is 7,708. And while you can't tell the exact number, you can tell that May is slightly below the 7,708, April is right on the number, and March is slightly above. It would appear that the average for March-May would be right at the 7,708 number or a little higher, but we will use the 7,708 again to be conservative.
So when we do the math - 7,708(average deaths per day) * 91 (days in March,April,May) = 701,428.
Again this number was using conservative data and likely should be higher.
Now we take 781,000 (actual deaths) - 701,428 (estimated deaths) = 79,572 deaths above predicted.
Compare this to the reported Covid deaths during this period (March 1 - May 30) of 95,235 and you come out with a -15,663.
I'm sure I must be wrong somehow, but would love for someone to prove it wrong.
* I actually found the number for 2018 deaths of 2,839,205. Slightly more than 2017 as expected. So unless there would be a rational reason for expected deaths to have decreased dramatically from 2018 to 2020, I just do see how it adds up.