Thread by @Nrg8000

Today the Epoch Times released a report claiming that real internal figures for #CVD19 cases in Shandong province were far higher than official figures. epochtimes.com/b5/20/2/24/n11…
The Epoch Times is not a reliable source but this data looks legitimate for these reasons:
THREAD:

Basically, the raw documents are very detailed and intricate, far more than they would need to be for the reporting, they were sent from the 'official mail of Shandong Province' & the data matches conclusions I'd made completely independently to this release. 2/

First I should make clear that this is not a verification. It's a reasonable assessment that is very interesting.
The Epoch Times isn't always a reliable outlet - I doubt they would fully fabricate a document but it wouldn't be hard to dupe them into publishing something fake. 3/

Luckily (perhaps knowing their reputation) the ET has shared links to ALL the data across 3 files, which crucially is far more detailed than there reporting.
The data includes 16 xlsx files of data every day's release from Feb 9-24 & a pdf of the spreadsheet from Feb 21st.

A (machine) translated version of the latest spreadsheet plus the original text is available here: drive.google.com/file/d/14XY0Jb…
5/

There are 3 things that stuck out to me from the spreadsheets:
the detail;
that it is a working document;
that it reflects officially acknowledged data.
6/

Firstly the amount of detail they have, while the reporting just focuses on daily totals and cumulative numbers, the spreadsheets not only break those figures into specific hospitals but also provide info on the number of tests ran each day, cumulative & total capacity. 7/

It also has details for each CDC and how many close contacts and suspected cases they have followed up, plus a 'type' of hospital with 3 classes & how many days hospitals can test for without resupply of reagents. 8/

The documents are also clearly meant to be working documents, for example the 'Total Qingdao' row doesn't have numbers but "=SUM(E23:E27)".
9/

It also reflects real information which has been officially confirmed. On the Feb 21st, it was reported that there was an outbreak in a prison/coal mine in Jining, that's reflected in the data with 200 cases reported across 4 hospitals in the district.
scmp.com/news/china/soc…
10/

The data also shows that another 200 cases in Jining were confirmed on Feb 24th, saying "备注：当日新增200阳性均为之前核酸检测阳性的任城监狱犯人样本", or: 'Remarks: on that day the newly added 200 positive nucleic acid detection cases were samples from Rencheng prison'. 11/

Translation thanks to @abeleung

@abeleung The reporting also provides a screenshot of the email that sent this info: [e-mail].
Testing this email address on online tools shows through 3 different tools that it is a real and valid email address. 12/

@abeleung Beyond the fact that we know the email address is valid, you can trace the mail server back to mail.shandong.cn which is on the internet archive as the "Official Mail of Shandong Province".
The website's manual probably has more info: web.archive.org/web/2019071514… 13/

@abeleung The tech-helpline here is a number for @inspurCorp and the secondary email server is "eYou Email defender" http://139.162.32.54/gw/user/ - which is referenced in the 'official mail of Shandong province's user manual (linked above). 14/

@abeleung @InspurCorp Other than looking at any attempt at technical verification, there's also what the data itself shows. I've long said (since the 16th) that in my opinion COVID cases outside of Hubei declined at an unrealistic rate since the 14th (see this thread:

https://twitter.com/Nrg8000/status/1232825664652238849

) 15/

@abeleung @InspurCorp This same pattern is brought out in the data, prior to the 14th, the reported figure was larger than the official one by an average factor of 3.62, following the 14th (removing outliers - Feb 18th, 20th & 23rd) the official figure was on average 16.2x larger. 16/

@abeleung @InspurCorp The 18th was removed because there was only 1 officially confirmed case which blows the whole thing out, the 20th & 23rd both had prison cases. 17/

@abeleung @InspurCorp All in all, it is impossible to rule out completely that this document is fake, but if it is, it is a breathtakingly intricate fake, far more than it would need to be. 18/

@abeleung @InspurCorp I have an idea on how to basically confirm it as 100% real, but that will take some time, I'll get back to this thread when I've done it! 19/19

@abeleung @InspurCorp CC @Comparativist @WilliamYang120 @stegersaurus @joshkellerjosh @mgerrydoyle @YuanfenYang @goldencaskcap @muyixiao @melissakchan @EmilyZFeng FYI

@abeleung @InspurCorp I think I messed up the link in the first tweet.
Try this: epochtimes.com/b5/20/2/24/n11…

It was also published 2 days ago not today! I'm just clearly not as on top of Epoch Times reporting as I could be😝

@abeleung @InspurCorp It's been mentioned (thanks @goldencaskcap!) that "当日检测标本数" simply refers to the number of tests positive, not necessarily new cases. Which is important because patients being released will often be tested multiple times before being relesed.

@abeleung @InspurCorp @goldencaskcap I would guess that given the total numbers (eg 黄岛区 CDC has tested 457 samples for 6 positive resuls) that the majority of tests are not being performed on previously confirmed cases.

@abeleung @InspurCorp @goldencaskcap It also couldn't fully explain the numbers here in recent days being 10-270x larger than official figues per day. But it should fully be noted that youre not directly comparing apples to apples here like the Epoch Times implies, but rather positive test samples not cases!