Data and analysis thread, started 12 November

[NoGoodPunsLeft]

Previous thread here:

Data and Analysis Thread, started Oct 29 www.mumsnet.com/Talk/coronavirus/4064113-Data-and-Analysis-Thread-started-Oct-29

Regular lurker but I frequent poster, didn't want to lose the threads.

The 23rd looks a bit spikey?

23rd was a Monday, typically the peak day over each of the last few weeks. In England mon 16th had 23661 by spec date Monday this week at 15478 with the peak reporting days now complete so likely to top out at 18000 max, and probably nearer the 17000 mark. So likely 6000 or so cases fewer than the previous Monday

Mondays (by specimen date) are always spiky. I would expect the number for 23/11 (England only) to go up to 18,000-19,000 (boys3 might have a more accurate estimate). Last Mon (16/11) was 23,611 so this would give us a 20% drop week on week, or a 3.5 day R-rate of 0.9. This is consistent with Zoe.

There is definitely a case for reviewing North West tiering earlier. GM numbers keep going down very consistently.

Liverpool is dropping to a very low level. The trajectory has been consistent and it started way before the lockdown.

twitter.com/rp131/status/1331959576279781376?s=21

Chart by RP131 of the new tiers. There’s a few outliers -tunbridge wells, Stratford upon Avon, Lancaster with low rates but in tier three, and a few boroughs in tier 2 with higher rates than many tier 3.

I'm not sure of this whole county approach. It's a bit sledge hammer in some places

Luton should be tier 3. They are prob only being saved by sharing a hospital with central beds who has a low rate so it's not heaving. They are a v localised problem... and stays that way

The map here is borked:
coronavirus.data.gov.uk/details/interactive-map

Is it the same for everyone?

@Quarantino yes somethings wrong - showing as “suppressed” or “Data not available” for everywhere I looked

I don't think this govt realise that saying review after 2 weeks means 3 weeks from now. I base this on Sunak talking this morning about the rates coming down for the lockdown for the last 4 weeks and how we can't waste that. It's been 3 weeks I was shouting at him but he couldn't hear.

I think it is a little bit worrying that the man who has to balance the books can't count to 4.

Downing St briefing. Didn't know Boris was due on.

Oh he's giving us a breakdown of what's happening in Italy, France and Germany. That's useful.

It’s interesting to hear France hospitality is closed until Jan! Is that really the case?

If Boris says it's true then you've probably got a 50/50 chance.

Apparently there was a problem with the data and they will be updating the so at about 6pm

This is from the actual software engineer that produces all the data who I happened to come across on Twitter!!
twitter.com/Pouriaaa

That's interesting ceeveebee.

Great!

Does anyone know whether the mass Liverpool tests are included in the pillar 2 data?

Pillar 2 includes mass testing (according to the dashboard). I’d have thought a separate pillar might have helped transparency.

England 7 day moving average falling at a reasonable pace. To Saturday 21st spec date stands at 16816 down from a peak of 22169 on 13th November. 16816 was last seen around 21st October. It is still however just over double the figure at the start of October (8116) and four times higher than the figure on 21st September (4072). And for really fond memories the figure on 1st September was 1297 - a number already causing concern at the time on the thread!

A way to go still.

www.google.com/amp/s/www.nytimes.com/interactive/2020/world/europe/france-coronavirus-cases.amp.html

France cases Map

Tier “evidence” in case not already posted
assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/938964/Coronavirus_England_briefing_26_November.pdf

@ceeveebee That’s a very interesting link. Thank you!

Is anyone else struggling with the case number reported for England today.

Dashboard has 13557 daily and 1367546 cumulative.

however cumulative cases by date reported yesterday shows as 1355272.

So a difference of 12274

When I look at today’s data file and yesterday’s to see the cases added by spec date I see daily additions of

789 for yesterday

9433 for Wednesday

2741 for Tuesday

447 for Monday - taking Monday to 15295 in total so far

Then the next 6 days all show a reduction

114 less for 22nd

101 less for 21st

132 less for 20th

170 less for 19th

476 less for 18th

134 less for 17th

I have yesterday’s file downloaded and so can double check the numbers

Something is not adding up.

Any thoughts appreciated

They've dropped them down the back of the sofa, where they'll find it (with an old sweet wrapper) in 2024?

Yes, I have also been puzzled by the today's numbers. Looks like the daily number might not have included the negative adjustments from the previous week?

Either the negative adjustments are a glitch and the correct cumulative total for today should have been higher or the daily number should in fact have been lower but PHE did not want to "understate" the daily numbers and excluded the negative adjustments from the today's figures.

The ONS report for this week included a comment that there was an issue with a laboratory in Milton Keyes that erroneously overstated positive cases in Pillar 4. Perhaps, this problem also impacted Pillar 2 results and the negative adjustments relate to the correction of the error.