COVID-19

You expected many parts of the South and Midlands to reach very low second wave peaks and then drop away again?

Not just East Europeans. Indian subcontinent, too. But they were just the obvious ones seeding in the UK. There would be others. There's been a surge of people bringing it back from Greece and Spain, too. Is it spreading if half of England is falling - more than half if you adjust for the increasing testing?

You expect 100% accuracy with this lot in charge? That's the thing about statistics. They are almost always wrong. To understand them, you have to work out why they might be wrong and how far out they might be from what you are looking for.

 

How long do those statistics take to show the full picture? There is an obvious lag unless there were only 4 cases in London yesterday?

No, quite the opposite. I expect the statistics to take time to take shape. This is exactly what I'm saying, the downturn on the graphs only happens if you take the last 7 days as gospel but we know that they haven't counted all tests yet. Exactly what I'm saying.

Yes. Some people brought it back from overseas but the point was that it was always spreading in the community too.

 

The data is starting to speak for itself. The trends are good. I just need ot work out if there's anything systematically wrong with it.


Hospital admissions trends for Southeast and Southwest are following the new cases trends down in those areas.

SE Cases:

https://coronavirus.data.gov.uk/cases?areaType=ltla&areaName=Waverley

https://coronavirus.data.gov.uk/cases?areaType=region&areaName=South East

SW cases:

Https://coronavirus.data.gov.uk/cases?areaType=ltla&areaName=Torbay

Https://coronavirus.data.gov.uk/cases?areaType=region&areaName=South West

SE Admissions:

https://coronavirus.data.gov.uk/healthcare?areaType=nhsregion&areaName=South East

(7-day average has dropped from 11.4 to 10.0.)

SW Admissions

https://coronavirus.data.gov.uk/healthcare?areaType=nhsregion&areaName=South West

(7-day average has dropped from 3.7 to 3.0.)



Note that cases and hospitalisations are still rising in parts of the North and Scotland but the turn in the South suggests these will follow. Maybe some parts of the North and Scotland had a few more areas that did not get the full hit of the first wave.

https://coronavirus.data.gov.uk/cases?areaType=region&areaName=North%20West

https://coronavirus.data.gov.uk/healthcare?areaType=nhsregion&areaName=North%20West


https://coronavirus.data.gov.uk/cases?areaType=nation&areaName=Scotland

https://coronavirus.data.gov.uk/healthcare?areaType=nation&areaName=Scotland

 

You're avoiding the question though.

How long does it take for that 'specimen date' data to be a true reflection? It looks to me like it takes 7-10 days?

So any analysis of the info within that period is intrinsically flawed and paints a picture which is more positive than the reality.

 

I don't know but I think we're past that already. South is turning down after a very modest wave 2 and some early places like Swindon peaked weeks ago. North has not got there yet but it looks like it will.

 

Why not just answer what you think the delay is then?

Even the explanation on the data you're citing says the last "few" spots on any curve should be ignored but what is a "few"? Its only turning down if you incorrectly include these before the full picture is known.

I have no issue with a positive spin being put on things but let's analyse the data honestly. It the only way to make informed decisions about where we go. You're trying to find specific places that show what you want it to and then including data that is incomplete. It's not a honest appraisal.

 

So Whitty the doom and gloom is going to be talking to us all today.

No Boris..

BORIS JOHNSON IS A SPOON!

 

Would you rather hear from Boris or the actual experts?

I know my preference!

 

I would rather him do his job and lead the country!

Get his own failure of a team in order and like Handcock. Sack people. Get this right. Not baffle about like an idiot.

 

Look at larger regions:

East peaked two weeks ago

https://coronavirus.data.gov.uk/cases?areaType=region&areaName=East%20of%20England

London peaked two weeks ago

https://coronavirus.data.gov.uk/cases?areaType=region&areaName=London

East Midlands peaked two weeks ago

https://coronavirus.data.gov.uk/cases?areaType=region&areaName=East%20Midlands

Southeast peaked over two weeks ago

https://coronavirus.data.gov.uk/cases?areaType=region&areaName=South%20East

Southwest peaked over two weeks ago

https://coronavirus.data.gov.uk/cases?areaType=region&areaName=South%20West


All these regions peaked two weeks ago or more despite testing increasing 30% since the end of August. The last few data points don't make any difference. Hospitalisations in the Southeast and Southwest have started falling about a week ago.. It ties in. The South Looks to have finished a wave 2 that was much smaller than wave 1 and it looks to be working northwards.

 

The peak is incorrectly calculated if it takes 2 weeks for the data to catch up.

That is my whole point but one you seem to be ignoring.

 

It looks to have started rolling over in the last two days. Last 7 days increase: 5057,6008,6582,7517,8262,7942, 7584. The South might have peaked two weeks ago but parts of Midlands and North have yet to do so while Scotland and Northern Ireland are still going up, and maybe Wales. I'd take hospital admissions ahead of this data, though. I never said the whole country had peaked. The peak and fall of wave 2 is spreading from the South.

Https://covid.joinzoe.com/data

 

I'm not ignoring it. The Southeast and Southwest peaked on 3rd/4th September and the other regions quoted around the 7th. That's 17/18 days ago and 14 days ago. Some local authorities peaked in August. It's normally only the last 3 or 4 days that will be significantly topped up. The latest delayed test results I've heard of have been 6-8 days. I doubt the data will alter significantly past that timeframe.

 

6-8 days is precisely the sort of time scales I was thinking.

Look at the South East in that case - the reason the line trends down at the end is because at the 8th day back it falls off a cliff.

Now we could look at that and say it peaked on the 10th - but we know, or at least suspect, that the last 8 days data will be topped up over the coming days. Will any of them beat the 8th or the 10th, who knows, we'll have to see, but analysing it now gives obviously false results.

19-09-2020 3
18-09-2020 42
17-09-2020 77
16-09-2020 112
15-09-2020 136
14-09-2020 150
13-09-2020 66
12-09-2020 137
11-09-2020 165
10-09-2020 211
09-09-2020 170
08-09-2020 192

But if we're looking at a line that shows a 7 day average - and we're saying that there is a 6-8 day potential lag then all the curves should go down towards the end but that's simply because the full picture hasn't been revealed yet. If the lag is 8-10 days then it's even more severe.

 

With all these stats that @Aleman says points to the south of the country already peaking I wonder why The two that are about to speak to the nation are not agreeing? Is it that they have more data to look at or are better at analysis it?

 

19-09-2020 3
18-09-2020 42
17-09-2020 77
16-09-2020 112
15-09-2020 136
14-09-2020 150
13-09-2020 66
12-09-2020 137
11-09-2020 165
10-09-2020 211
09-09-2020 170
08-09-2020 192
07-09-2020 204
06-09-2020 180
05-09-2020 160
04-09-2020 232
03-09-2020 226
02-09-2020 209

You seem to have left a few days off there which I've added. The peak was September 4th. That's over 2 weeks ago - as I posted earlier.

https://coronavirus.data.gov.uk/cases?areaType=region&areaName=South East

And hospital admissions peaked at 17 on the 9th of September, with the 7-day average peaking on the 11th at 11.4. That ties in well and I've read of no issues with delayed data for hospital admissions. It's peaked over a week ago and not really very high for the peak of wave 2 for the whole Southeast really is it, in the scheme of things? Wave 1 peaked at 350 per day. It's less than 10 per day now and looks to be falling, and in line with national admission figures that look to have flattened still way below wave 1 levels.

https://coronavirus.data.gov.uk/healthcare?areaType=nhsregion&areaName=South East

 

None of that changes what I said. If the delay is 6-8 day then those latest days could still end up anywhere and any curve which includes those dates will be wrong.

 

What a waste of time that press conference was