Read on Twitter
THREAD
@CMOIreland had a tough press conference there. @GeorgeLeeRTE confused over numbers.
I understand it now, but I think I've found another problem...
1/13
@CMOIreland had a tough press conference there. @GeorgeLeeRTE confused over numbers.
I understand it now, but I think I've found another problem...
1/13
The @HSELive reports at the press conference every day at 5-9pm. Sometimes it's just a press release.
It's the most up to date number.
2/
It's the most up to date number.
2/
The total has risen from 6,574 to 8,089 in one day, an increase of 1,515.
@CMOIreland says if he's asked, *this* is the number of cases he will say exists in Ireland *right now*.
3/
@CMOIreland says if he's asked, *this* is the number of cases he will say exists in Ireland *right now*.
3/
The problem is that two numbers were given tonight.
7,054 and 8,089
The difference between the two is the German lab results.
4/
https://www.gov.ie/en/news/7e0924-latest-updates-on-covid-19-coronavirus/
7,054 and 8,089
The difference between the two is the German lab results.
4/
https://www.gov.ie/en/news/7e0924-latest-updates-on-covid-19-coronavirus/
So if I illustrate this, you can see that 1,035 tests from German labs comprise most of the increase.
Nothing untoward here by @CMOIreland, they're just being upfront that lab results skew the numbers, but increase the total.
But that's not the end of it, however.
5/
Nothing untoward here by @CMOIreland, they're just being upfront that lab results skew the numbers, but increase the total.
But that's not the end of it, however.
5/
There's a SECOND set of data released by @hpscireland. Why? Because it gives more detail (county, age etc.)
Problem is, their data is "as at midnight" each night, and it's reported with a lag, because it takes time to verify more detailed data.
6/
Problem is, their data is "as at midnight" each night, and it's reported with a lag, because it takes time to verify more detailed data.
6/
This is the @hpscireland number. You can see we're still waiting on April 9th and 10th data because of the lag.
It shows 7,071 cases (as at midnight) on April 8th.
7/
It shows 7,071 cases (as at midnight) on April 8th.
7/
So the different data points are:
7,071 - April 08 (midnight, HPSC)
6,574 - April 09 (press conference, HSE)
7,054 - April 10 (press conference, no German lab)
8,089 - April 10 (press conference, *including* German lab)
8/
7,071 - April 08 (midnight, HPSC)
6,574 - April 09 (press conference, HSE)
7,054 - April 10 (press conference, no German lab)
8,089 - April 10 (press conference, *including* German lab)
8/
But wait.....
Now it looks like the number went *down* yesterday (from 7,071 to 6,574) and then back up again today (to 7,054 AND 8,089)
AGH, I know, confusing!
9/
Now it looks like the number went *down* yesterday (from 7,071 to 6,574) and then back up again today (to 7,054 AND 8,089)
AGH, I know, confusing!
9/
The data from March 22nd to April 5th (Sunday) has been consistent.
Gradual increase between the 5-9pm press conferences and the midnight reporting.
10/
Gradual increase between the 5-9pm press conferences and the midnight reporting.
10/
But look at the past 5 days.
Numbers going up and...... *down*.
I could be wrong, but it looks like @HSELive and @hpscireland have had inconsistent numbers and NOBODY NOTICED
11/
Numbers going up and...... *down*.
I could be wrong, but it looks like @HSELive and @hpscireland have had inconsistent numbers and NOBODY NOTICED
11/
Look at it zoomed in.
It shouldn't be like that at all.
I can't believe I missed it.
The whole country missed it?
Again, I could be wrong. Please tell me if I am!
12/
It shouldn't be like that at all.
I can't believe I missed it.
The whole country missed it?
Again, I could be wrong. Please tell me if I am!
12/
I've no doubt this is an innocent mistake, unintentional human error. But it's confusing alright.
Anyway, tomorrow is another day, which means another press conference, so I'm sure the media can ask about it then!!
13/13
Anyway, tomorrow is another day, which means another press conference, so I'm sure the media can ask about it then!!
13/13
Thanks to @gavreilly and @SineadOCarroll for clarifying that the HPSC numbers *include* the German lab data spread over several days, but the HSE data only adds it in *today* (hence the difference).
So that explains it.
A bit woeful from @HSELive that they let this go on
14/13
So that explains it.
A bit woeful from @HSELive that they let this go on
14/13
Here's what the Press Conference numbers would look like if the German lab data was included. I've *estimated* these numbers, making sure yesterday's change was the 427 @GeorgeLeeRTE mentioned.
That fixes it!
15/13
That fixes it!
15/13
