[Oldmanmatt]

150k is there figure so about 30k a month. Hence why I think pessimistic bull shit and want to know what the same people said it would be back in March. There were some huge figures thrown about by the doom merchants on here.

You’re not getting it.

The projection is based on the current trend (that is historically over the last x days) and assumes nothing changes, that the trend continues.

However, in reality, behaviour will change, restrictions imposed etc etc.
In 7 days, the projection will be  quite different (hopefully).

If you’d seen the graph back in  late march, say in the first seven days of lockdown, it would have risen steeply to the maximum predicted death toll, because that’s what an unmitigated pandemic would do.

Perhaps focus on the resource  use graph, which is much clearer on the error margin front.
Then mentally apply similar error margins to the other graphs.

There’s little attempt to anticipate changes in behaviour, either mandated or organic in the prediction.
That is to say, some behaviour will be modified by government, some (organic) by people’s personal responses and, well, fear.

[tomtom]

Trying not to be doom mongery gav but I I think it’s getting towards c.45k deaths within 28 days of a test and the other estimates (excess morbidity etc..) put the likely Cv total at 60k or more.

I think the 200k forecast was without any measures - modelled back in March. But might be wrong.
Though, With 45-60k dead now - and 150-200 a day at the moment that puts us on course to have an extra 10k+ by new year even if the death rate per day doesnt grow much (they will as they lag case numbers by 4 weeks).

So 200k - was at the pessimistic end of things but not bonkers. We could be getting closer 100k if things don’t go well.

(Edit - what OMM said!)

What will help is that we know a lot more about how to treat it (eg pcap instead of ventilators) and have new drugs (eg steroids) that can genuinely drop the mortality rate. Which is good 👍👍

[gme]

If I ever comment on here again about this I give you all the right to shoot me.

I’m out of here you have now all pushed me over the edge.

[tomtom]

Sorry Gav - don’t go - we need reeling in!

[spidermonkey09]

Sorry, but GME is talking shite anyway. Being pessimistic has nothing to do with it. Matt explained how that figure was arrived at and how it was never going to look like that in reality because behaviour would change. I suspect, like all of us, he is fed up of the whole situation and wishes it would just piss off, but saying every figure or projection that looks bad is "negative" or "pessimistic" sounds like a knock off version of a brexit ad saying "don't talk Britain down."

[mrjonathanr]

Is there a more civil way to make that point Spidermonkey?

[Offwidth]

Fair enough folks, thanks for setting me straight.

This site is useful, in as much as it updates it’s projections regularly and corrects assumed to actual. Most interesting is that it shows bed availability.

https://covid19.healthdata.org/united-kingdom?view=total-deaths&tab=trend

An interesting site (especially mobile phone data showing average 'social distancing' trends) but I'm not sure how they arrived at their number of deaths. The UK for example isn't the official number nor the ONS number. Excess deaths are likely the best indicator of the overall effects of covid and the FT page shows the UK at 67500 excess deaths and the US at 270,000 and still growing.

https://www.ft.com/content/a2901ce8-5eb7-4633-b89c-cbdf5b386938

The way deaths are counted is important and the UK numbers capture fewer actual covid deaths than most western nations (Belgium seems to be the most honest). The Kings fund have a good article on UK counting but the last update was August

https://www.kingsfund.org.uk/publications/deaths-covid-19

[IanP]

Is there a more civil way to make that point Spidermonkey?

And also GME does have a point in some ways, the use of simplistic projections based on short term trends to produce 'worst case' outcomes can seem to sometimes to be close to scaremongering and I'm not sure that it helps the argument.

e.g. Covid-19: UK could face 50,000 cases a day by October without action - Vallance (21st September), https://www.bbc.co.uk/news/uk-54234084

By 16th October 7 day rolling cases were 16k a day, 21k a day as of 24th - obviously bad but still a way off 50k.   Aware that Vallence did say 'without further action' but it was the 50k that got the headlines and he knew that would be the case.

[Offwidth]

Maybe the population needs a bit of scaremongering given hospitals in the worst hit areas are already close to capacity and we already have a couple of weeks growth built into the older demographics. I do wonder how much more official scaremongering we would be facing if those hospitals were in London instead of in places like Liverpool.

[Oldmanmatt]

Fair enough folks, thanks for setting me straight.

This site is useful, in as much as it updates it’s projections regularly and corrects assumed to actual. Most interesting is that it shows bed availability.

https://covid19.healthdata.org/united-kingdom?view=total-deaths&tab=trend

An interesting site (especially mobile phone data showing average 'social distancing' trends) but I'm not sure how they arrived at their number of deaths. The UK for example isn't the official number nor the ONS number. Excess deaths are likely the best indicator of the overall effects of covid and the FT page shows the UK at 67500 excess deaths and the US at 270,000 and still growing.

https://www.ft.com/content/a2901ce8-5eb7-4633-b89c-cbdf5b386938

The way deaths are counted is important and the UK numbers capture fewer actual covid deaths than most western nations (Belgium seems to be the most honest). The Kings fund have a good article on UK counting but the last update was August

https://www.kingsfund.org.uk/publications/deaths-covid-19

“Where does IHME obtain its data?

These forecasts include data from local and national governments, hospital networks and associations, the World Health Organization, third-party aggregators, and a range of other sources. We use the Johns Hopkins University (JHU) data repository on Github to collate daily COVID-19 cases and deaths. We supplement this dataset as needed to improve the accuracy of our projections. For example, we use data from government websites for Brazil, Canada, France, Germany, Italy, Japan, Mexico, Pakistan, South Africa, Spain, United Kingdom, and the US states of Indiana, Illinois, Maryland, and Washington. For New York, we use data from the New York City Department of Health and Mental Hygiene and the New York Times GitHub repository. We obtain subnational data from government websites. Our models are updated regularly, as new data are available, to provide the most up-to-date planning tool possible.

For testing data, our primary sources for US testing data are compiled by the COVID Tracking Project. For other locations, we rely primarily on data reported by Our World in Data. However, for Brazil, Canada, Cyprus, Dominican Republic, Honduras, India, Italy, Japan, Mexico, Moldova, Pakistan, Philippines, Russia, South Africa, and Spain, we use government data.

We obtain hospital resource data from sources such as government websites, hospital associations, the Organisation for Economic Co-operation and Development, WHO, and published studies. For population density, we use gridded population count estimates for 2020 at the 1 x 1 kilometer (km) level from WorldPop. For masks, we use data from Premise for the US, and from the Facebook Global symptom survey. (This research is based on survey results from University of Maryland Social Data Science Center.) For mobility, we use anonymized, aggregated data from Google, Facebook, and Apple. For the US, we use mobility data from Descartes and SafeGraph.

Our data on mask use come from Premise, Facebook Global Symptom Survey (research based on survey results from the University of Maryland Social Data Science Center), Kaiser Family Foundation (KFF), and YouGov COVID-19 Behaviour Tracker survey.

For a complete list of our supporting organizations, please see our Acknowledgements page.“
http://www.healthdata.org/covid/faqs

Ian, the variation in forecast between the worst case “go back to normal living” scenario, our actual “keep some mandates, wear masks sometimes, locally increase mandates (etc)” and their “full mandate, universal mask use” is immense.
Vallence was pointing out the possibility of the first scenario, if no action was taken.
The reality being, of course, that it’s not just government mandates that influence it, it’s peoples attitudes and I’m sure he would argue that this was his primary  goal; reminding people to be careful.
I think the government’s actions and mandates, so far, have been enough, to keep enough, on their toes enough, to control the spread just enough, to skim along the catastrophe curve...
They could have done more, but not a lot less.
In a couple of weeks, we might find out they haven’t actually done as much as they needed to.

Now, the last paragraph was meant to be all over the place, because I’m trying to convey how fuzzy and inclusive the modelling has to be.

Personally, I think the government have been slightly too cautious, but not as bad as many suggest (leaving aside the corruption aspects etc, merely looking at mitigation policies). I think a huge issue, is actually a personality fault, rather than a policy fault:

Johnson is equivocal in all things, all the time.

He never conveys conviction or decisiveness  in his oration and his audience only ever hear what they want to hear.

[spidermonkey09]

Is there a more civil way to make that point Spidermonkey?

Yes, almost certainly. It wasn't meant to be uncivil but I was getting a bit frustrated with the word "pessimistic." pessimism and optimism are really bad words to use in this scenario, implying that if we all stick our fingers in our ears and should loud enough the problem will just disappear.

It didn't feel like there was an awful lot of engagement with the argument being put forward and an overreliance on "I don't like this projection so you're all scaremongering." I actually agree with IanPs analysis entirely, because clearly the projections are simplistic and headline grabbing. However Gav didn't make that point, he just ranted about "doom merchants". But I do take your point about civility, so my apologies if it came across as me being a wanker.

[mrjonathanr]

I wouldn’t have gone that far Spidermonkey 😀. Just thought the Brexit thing was a bit harsh, that’s all.

[IanP]

Fair enough folks, thanks for setting me straight.

This site is useful, in as much as it updates it’s projections regularly and corrects assumed to actual. Most interesting is that it shows bed availability.

https://covid19.healthdata.org/united-kingdom?view=total-deaths&tab=trend

An interesting site (especially mobile phone data showing average 'social distancing' trends) but I'm not sure how they arrived at their number of deaths. The UK for example isn't the official number nor the ONS number. Excess deaths are likely the best indicator of the overall effects of covid and the FT page shows the UK at 67500 excess deaths and the US at 270,000 and still growing.

https://www.ft.com/content/a2901ce8-5eb7-4633-b89c-cbdf5b386938

The way deaths are counted is important and the UK numbers capture fewer actual covid deaths than most western nations (Belgium seems to be the most honest). The Kings fund have a good article on UK counting but the last update was August

https://www.kingsfund.org.uk/publications/deaths-covid-19

Not wanting to restart an old discussion , but the figures in the FT don't show that the UK capture fewer covid deaths that other western countries.

Comparing deaths on worldometers to excess deaths on FT (accepting that dates aren't in sync but since majority of deaths were earlier in pandemic so hopefully still gives a general idea of numbers) shows UK sits in the middle when compared to other European countries.

Country   FT figures   Gov figures   Gov %
Portugal   7300      2297      31%
Germany   17800      10111      57%
Spain   57600      34752      60%
Netherlands 11300      7019      62%
UK      67500      44745      66%
Italy      50500      37210      74%
Belgium   10700      10737      100%
Sweden   5700      5933      104%
France   26600      34645      130%

You're right about Belgium, France is strange, seemingly having less excess deaths than recorded Covid deaths.

[Offwidth]

Not wanting to restart an old discussion , but the figures in the FT don't show that the UK capture fewer covid deaths that other western countries.

Comparing deaths on worldometers to excess deaths on FT (accepting that dates aren't in sync but since majority of deaths were earlier in pandemic so hopefully still gives a general idea of numbers) shows UK sits in the middle when compared to other European countries.

Country   FT figures   Gov figures   Gov %
Portugal   7300      2297      31%
Germany   17800      10111      57%
Spain   57600      34752      60%
Netherlands 11300      7019      62%
UK      67500      44745      66%
Italy      50500      37210      74%
Belgium   10700      10737      100%
Sweden   5700      5933      104%
France   26600      34645      130%

You're right about Belgium, France is strange, seemingly having less excess deaths than recorded Covid deaths.

They do. You need to compare excess deaths on the FT initial peak with official deaths at the date at the end of that FT peak for a fair comparison of how well the state system recorded covid related deaths (ie look at both sets of data in early June). Back then Belgium was the closest to picking up most covid deaths and Britains 'official' numbers bottom of the countries with a big excess death peak (even though UK government numbers were not time limited to 28 days then). The 'strange' data is due to people who died of covid over the summer and early autumn when excess deaths are not discernable from the normal yearly average; showing that comparing to excess deaths is only practical at times when there are excess deaths.

[IanP]

I knew I shouldn't get involved 

[Offwidth]

I knew I shouldn't get involved 

I'm glad you did it as it illustrates some interesting points (and you were right about Portugal as I hadn't looked at it in my earlier comparisons, as I thought peak excess deaths there were too low compared to the baseline range). I looked at peak deaths below as the excess deaths are least affected by uncertainty in the baseline.

In the UK peak excess deaths were 12,900 a week (John Hopkins...adding data for separate country entries in mid April and subtracting the average baseline, on a baseline data range of 2900) and official UK government weekly deaths (worldometer) were 6600, so only 51% of that. In our official covid numbers we only pick up half the excess deaths (+/-11% on accuracy on excess deaths from the baseline range). The ONS peak weekly death data was 67% of the peak excess deaths (8760 from the Kings Fund link).

The comparative percentage for Portugal of official reported deaths compared  to excess deaths is 46% (it is lower but the actual value might be higher given the difference is well within the baseline range of +/-29%) but that is on peak weekly excess deaths of 470 (with a baseline range of 280 from yearly changes in the baseline data....John Hopkins).  For reference the population of Portugal is about 15% of that of the UK, so our peak weekly excess deaths were about 4 times larger per capita.

[Tony]

It should be noted that excess deaths and COVID19 deaths are very different metrics.

Death for a single cause stats suffer from "competing risk events": we will all die only once and we can only be classified as having died due to one underlying cause (slight simplification).

Excess deaths are about all cause mortality, total deaths that would not "normally" have occurred had COVID19 and its impacts *not* [edit] been around (e.g. acute deaths due to sub-normal medical care/availability, longterm deaths due to disruption to medical services).

Bear in mind, some deaths were probably "avoided" during COVID, e.g. road traffic deaths would have been lower during the (March/April) lockdown.

[IanP]

It should be noted that excess deaths and COVID19 deaths are very different metrics.

Death for a single cause stats suffer from "competing risk events": we will all die only once and we can only be classified as having died due to one underlying cause (slight simplification).

Excess deaths are about all cause mortality, total deaths that would not "normally" have occurred had COVID19 and its impacts *not* [edit] been around (e.g. acute deaths due to sub-normal medical care/availability, longterm deaths due to disruption to medical services).

Bear in mind, some deaths were probably "avoided" during COVID, e.g. road traffic deaths would have been lower during the (March/April) lockdown.

Can't disagree with much of that, obviously excess deaths is measuring something other than specific covid deaths but it has been used as proxy for the impact of covid on overall mortality.  In particular the discrepency between excess deaths officially reported by many countries (including the UK) and the estimates of excess deaths point to a lack of accuracy on those figures - as per my rough ready analysis above it would seem that many western European countries have a pretty significant difference between the 2 numbers.

In the UKs case we actually have better numbers for Covid deaths (though still probably undercounting) from the ONS based on death certificates - as of w/e 9/10 this had approx 58,000 deaths as compared to 43.500 deaths on the daily figures produced by the government. 

[Bradders]

https://www.telegraph.co.uk/travel/news/police-patrolling-english-county-borders-weekend-prevent-non/

Essence of the report; North Yorkshire Police will apparently be patrolling the borders to discourage non-essential travel into the county.

Obviously rules have changed yet again this week, but the way I read them neither travelling for any form of exercise nor leaving a tier 3 area to go to a tier 2 area (e.g. Leeds to Crookrise) are currently illegal. Anyone's guess how this might actually play out if you were pulled over. I imagine they'll be concentrating on larger groups as opposed to lone drivers?

Use of ANPR is mentioned but I don't see how that will work since they can't know why people are on the road? Perhaps just a case of using it to target cars not registered locally for stops?

[spidermonkey09]

 Just a PR exercise from the coppers I think. From memory North Yorks police were the most heavy handed force in the country in terms of handing out fines during lockdown. A cynic might suggest the chief there is trying to make a point of how tough they are to our esteemed home secretary!

I don't think it changes a lot from a climbing perspective for the reasons you say. You aren't doing anything illegal so I'd not take too much notice.

[Bradders]

Yeah, I've heard a few times before all this that NY Police were pretty militant anyway, so imagine that's not changed.

Checked the rules again just to be sure:

https://www.gov.uk/guidance/tier-3-very-high-alert

You can continue to do unlimited exercise alone, or in an outdoor public place in groups up to 6.

Interestingly, exercise is no longer listed as one of the reasons you can travel / leave a tier 3 area:

Travelling into or out of a Tier 3 alert level area

Avoid travelling outside your area, including for overnight stays, other than where necessary, such as:

for work
for education
to access voluntary, charitable or youth services
because of caring responsibilities
for moving home
to visit your support bubble
for a medical appointment or treatment

 

[webbo]

I think what they are trying to discourage is people in Tier 3 areas travelling in to North Yorkshire to go to the pub or for a meal out.

[abarro81]

Reading that part of the link in full, it seems that travel is a "should" not a "must", therefore is frowned upon, but what they gonna do...

[Bradders]

Yeah think you're right, just really can't be arsed having an argument. Was planning to travel through NY rather than stop in it tomorrow anyway but think I'll just go round em. 

[webbo]

So my grandson who lives in NY has been off school because someone connected to a class mate tested positive. So he came to stay with us in Tier 3 because my daughter had to work( Building Society) son in law in army is having to stay in camp because he’s no transport to get too and from camp to home. We will have him again next week then on Friday I will take him back to NY pick daughter up and drive back in to Tier 3 so she can pick up husbands new car. Then she will drive back to another bit of NY to pick him up, then go home.
Are we breaking many rules.