as i stated in my earlier posts, all of my data - ALL of it - came directly from CSV dumps of the CDC's all cause mortality data. the CDC gathers all-cause mortality information from all of the states and then aggregates data together to come up with national statistics and statistics broken down by age. i downloaded it from the flu-view site which breaks out data about flu deaths, pneumonia deaths, and all-cause deaths.

now, i was careful to use recent data-dumps. i understood that it probably took some time for information to be gathered, but i also relied on the fact that the CDC published a statistic which explicitly showed the completeness of each week's data. i ONLY included 2020 data which was certified to be either 99% complete or above, and only included historical data which was certified to be 95% complete or above when making averages.

that last point leaves me completely flummoxed where this problem came from, given that i was gathering data to which i was told was complete. and yet, the numbers have been SIGNIFICANTLY revised as of yesterday night.

and i'll eat crow - the newly revised numbers paint a COMPLETELY different picture than the one i'd been giving before. just take a look. here is the latest update...

https://docs.google.com/spreadsheets/d/e/2PACX-1vRf4vLTyjks-agvS9qWW7q1GcBuTEJ0fD5azbqSNhcUs5RUU1kyVmg2MXOv2CnEtrDNFxWDjHe51bBz/pubhtml

side by side with the numbers gathered from sunday night:

https://docs.google.com/spreadsheets/d/e/2PACX-1vSVX3M0yCQWOpGcVPPEGHMEJZ2sW_GqZrFIo2swUdl1ywLkmCzjl9rlq-FYklLMXhXJUfb3Pt9g2zAh/pubhtml

i legit don't get how this happened. i checked MONDAY to correct the bloomberg article and pulled new data for NYC from the CDC at that time. sometime between then and now, all the data changed on the CDC side.

but i meant what i said - i came in without bias and went where the numbers took me. so i guess i'm back on the side of believing that this is real. weird to say online but i'm crushed... using the old CDC numbers i had hope things were getting better. they are not.

(http://www.autoadmit.com/thread.php?thread_id=4534096&forum_id=2#40207780)

all ages nationwide all cause mortality delta to date:

- using new numbers: 40,114

- using old numbers: -3,575 (so this isn't CLOSE)

all ages nationwide "people to death" to date:

- expected: 341.48 people per death

- using new numbers: 333.75 people per death

- using old numbers: 349.13 people per death

>18 year old nationwide all cause mortality data to date:

- using new numbers: -2,471

- using old numbers: -3,575

18-65 year old nationwide all cause mortality data to date:

- using new numbers: 3,330

- using old numbers: -9,482

65+ year old nationwide all cause mortality data to date:

- using new numbers: 46,149 (!!!)

- using old numbers: 15,871 (which i already thought was HUGE)

NYC all ages data:

- new delta: 19,217 (HOLY FUCK)

- old delta: 13,347 (again, already thought it was bad)

- expected people per death: 426.64 people per death

- new PPD: 213.95

- old PPD: 251.96

MI all ages data:

- new delta: 3,767

- old delta: 2,213

- expected PPD: 298.72

- new PPD: 270.33

- old PPD: 282.16

CO data... strangely? is the same

...so is FL. wtf.

(http://www.autoadmit.com/thread.php?thread_id=4534096&forum_id=2#40207879)

and the worst part? the data reported to the NYT, that bloomberg used in that article i debunked? THEY ALL USED THE SAME DATA I DID.

what in the fucking hell is the CDC doing certifying their data as ">99.9% complete" when they add +17,000 deaths three weeks later?

like fuck maybe tracking all-cause mortality is really hard, and it takes about a month for any particular week's number to crystalize. fine. hard to believe but fine. JUST SAY THAT

(http://www.autoadmit.com/thread.php?thread_id=4534096&forum_id=2#40208191)

i can't take the quarantine much more. i felt hope it might be over soon. now i feel utterly hopeless.

(http://www.autoadmit.com/thread.php?thread_id=4534096&forum_id=2#40207882)

lots of people feel the same way. i think it will ease up soon. too many businesses/livelihoods are dying and we've gotten used to a steady level of death. as long as the death rate doesn't accelerate, we're reopening.

(http://www.autoadmit.com/thread.php?thread_id=4534096&forum_id=2#40208009)

i've come around to the view that the IFR is interesting but not nearly as important as the skew toward olds and how policy-makers react to it

i suspect we'll start to see a lot of news articles soon about the outsized impact on nursing homes and how badly we fucked up on protecting olds there, which will hopefully redirect the resources into turning those homes into fortresses with intense lab-grade HVAC and other things that we would have installed if we actually took the threat seriously

(http://www.autoadmit.com/thread.php?thread_id=4534096&forum_id=2#40208036)

very hard to take.

(http://www.autoadmit.com/thread.php?thread_id=4534096&forum_id=2#40208175)

- using new numbers: 3,330

- using old numbers: -9,482

Yep we really need to lock down all 18-65 year olds now.

(http://www.autoadmit.com/thread.php?thread_id=4534096&forum_id=2#40208167)