I suspect former politics student Tom Harwood of GB News will not be
presenting a video pointing out death rates continue to rise to levels
seen only during the lockdown.

I wonder why not. Doesn't Tom and his sidekick have any more graphs to
show how the media, the NHS, Public Health England, the statistics office
and government have got their data "180 degrees" wrong?

Hang your head in shame, Tom, for encouraging gullible viewers to let
their guard down and risk potentially deadly consequences.

Pamela wrote:

> I suspect former politics student Tom Harwood of GB News will not be
> presenting a video pointing out death rates continue to rise to levels
> seen only during the lockdown.

Quite right because the case rate, admission rate and death rate have all fallen
in the past 7 days

<https://coronavirus.data.gov.uk>

What of the 200,000+ cases/day that were predicted in mid-July?

Pamela <pamela.private.mailbox@gmail.com> wrote:
> I suspect former politics student Tom Harwood of GB News will not be
> presenting a video pointing out death rates continue to rise to levels
> seen only during the lockdown.
>
That is not correct if I read the govt data correctly. In January/February
deaths peaked at about 1300 daily. At the moment at less than 200 daily and
*declining*.

https://coronavirus.data.gov.uk/details/deaths?areaType=overview&areaName=United%20Kingdom

Don’t get me wrong, I’m no supporter of GB news.

In article <smjrgr$99d$1@dont-email.me>,
Tweed <usenet.tweed@gmail.com> wrote:

> Don't get me wrong, I'm no supporter of GB news.

Why not, they do attempt to show both sides of arguments which is
unprecedented on the main stream. The number of cock ups gets less as
time goes by and I do like Colin Brazier in particular, he often
shows me another view of things I hadn't considered before. Very good.

Bob.

On 11/11/2021 20:00, Bob Latham wrote:
>
> In article <smjrgr$99d$1@dont-email.me>,
> Tweed <usenet.tweed@gmail.com> wrote:
>
>> Don't get me wrong, I'm no supporter of GB news.
>
> Why not, they do attempt to show both sides of arguments

Did they report today that, contrary to their previous slagging off of
other media, the case rate went back over 40k? Of course the daily
total varies quite a bit, but during the fall of the last two weeks
there wasn't a day with a case rate significantly higher than the trend,
until today. Did they report that? Or just gloss over it?

--

Fake news kills!

I may be contacted via the contact address given on my website:
www.macfh.co.uk

Java Jive wrote:

> Did they report today that, contrary to their previous slagging off of other
> media, the case rate went back over 40k?

It's fluctuated up and down since the middle of July, and no doubt will continue
to do so ... probably half-term related in recent 7-10 days.

On 19:21 11 Nov 2021, Andy Burns said:
> Pamela wrote:
>>
>> I suspect former politics student Tom Harwood of GB News will not
>> be presenting a video pointing out death rates continue to rise to
>> levels seen only during the lockdown.
>
> Quite right because the case rate, admission rate and death rate
> have all fallen in the past 7 days
>
> <https://coronavirus.data.gov.uk>
>
> What of the 200,000+ cases/day that were predicted in mid-July?

The current death rate is 192 per day. Perhaps your page is showing
raw data without 7-day smoothing? I'm using the chart called "UK:
coronavirus deaths per day" from the Guardian here:

https://tinyurl.com/yd7v4zr8 which is captioned:

"People who have died within 28 days of their first positive test
for coronavirus. Deaths assigned to the date they are reported.
Line shows daily average in a given week. Data: data.gov.uk,
updated 11 November, 2021"

The BBC is reporting 195 deaths today, Thursday:
https://www.bbc.co.uk/news/uk-51768274

In his notorious video, Christopher Swanson tells Tom Harwood the
death rate is only 150. In fact the death rate is markedly higher than
that.

https://www.youtube.com/embed/F_p4nvqZLQg

Harwood's prime message is the media is using misleading data. He then
points to some minor variation between data sets and uses these to
draw the false conclusion that the pandemic is receding. Oops!

Harwood even had the cheek to criticise the BBC directly over
something which a phone caller, not the BBC, had said to Jonathan
Van-Tam about th enumber of school cases.

Pamela wrote:

> On 19:21 11 Nov 2021, Andy Burns said:
>
>> Pamela wrote:
>>>
>>> I suspect former politics student Tom Harwood of GB News will not
>>> be presenting a video pointing out death rates continue to rise to
>>> levels seen only during the lockdown.
>>
>> Quite right because the case rate, admission rate and death rate
>> have all fallen in the past 7 days
>>
>> <https://coronavirus.data.gov.uk>
>>
>> What of the 200,000+ cases/day that were predicted in mid-July?
>
> The current death rate is 192 per day. Perhaps your page is showing
> raw data without 7-day smoothing? I'm using the chart called "UK:
> coronavirus deaths per day" from the Guardian here:

I'm using the .gov.uk page I linked above, do you even look at it?

It shows each graph with a red highlight when things have been worse in the past
7 days and green when it's been better, all three graphs are green at the moment.

> https://tinyurl.com/yd7v4zr8 which is captioned:
>
> "People who have died within 28 days of their first positive test
> for coronavirus. Deaths assigned to the date they are reported.
> Line shows daily average in a given week. Data: data.gov.uk,
> updated 11 November, 2021"
>
> The BBC is reporting 195 deaths today, Thursday:
> https://www.bbc.co.uk/news/uk-51768274

Looking at one day compared the the previous day is generally a nonsense, it
depends whether you choose date of death or date of reporting, there's such a
5weekday/2weekend pattern it only makes sense to look at weekly figures.

Additionally there have been a few reporting issues recently, where data has not
been reported on time, so one days spike on a graph may include two days deaths

> In his notorious video, Christopher Swanson tells Tom Harwood the
> death rate is only 150. In fact the death rate is markedly higher than
> that.
>
> https://www.youtube.com/embed/F_p4nvqZLQg

I've been listening to that in the background while I type, in fact I've just
heard that report refer to the same data reporting issues

And they also refer to the importance of using the rolling average, I really do
think the media just big scary bad numbers ...

GB News is calling it right AFAICT

> Harwood's prime message is the media is using misleading data. He then
> points to some minor variation between data sets and uses these to
> draw the false conclusion that the pandemic is receding. Oops!

It's receded to the point we can live with it, vaccines are working, masks are
not required in everyday settings, even the BBC a week ago said (asked?) if
covid has peaked for this year, it's not over, it's here to stay but people need
to be able to get on with life ...

Pamela wrote:

> The current death rate is 192 per day.

I don't know where you get the number 192 from, the named graph you refer to
doesn't show it, it clearly shows the same falling average as the graph I
referred to, with a "spot" number of 195 on the final day, that number is not an
average, it is not a "rate", it isn't even a "count", it's skewed by the
reporting process, if you want to know how many people literally died yesterday,
come back in at least a fortnight.

> Perhaps your page is showing raw data without 7-day smoothing?
the uk.gov graphs do use smoothing, if you look at the graph by date of death,
the average is 152/day and falling; if you look at the graph by date of
reporting it is 162/day and falling.

> The BBC is reporting 195 deaths today, Thursday:
> https://www.bbc.co.uk/news/uk-51768274

Yes, yesterday 195 people were *reported* to have died, that is different from
saying 195 people actually *died* yesterday, even disregarding the skew of
whatever late reporting happens.

> In his notorious video, Christopher Swanson tells Tom Harwood the
> death rate is only 150. In fact the death rate is markedly higher than
> that.

Granted he's rounded down, maybe by 2, maybe by 12 depending on which graph he's
referring to.

I actually feel that apart from death rates the actual figures on infections
are not very accurate, since many people now get Covid and show no symptoms
at all, and so no test. Also in the middle of the last run ins with the
virus, there were far fewer tests, far more were ill and the age ranges
differed.
It is true it seems that in the main, apart from those with depressed
immune systems, its the younger end of the age range, who are probably not
vaccinated who are getting it.

I do agree on one thing though, In crowds, public transport and other places
it is sensible to wear masks, as it is not to protect you, but others, since
any one of us who are vaccinated could be spreading it without knowing.
Also of course most viruses need to spread not kill their hosts, so I
imagine that this pressure will eventually make it less aggressive but more
infectious, and able to persist at low levels even in the vaccinated.

As to the responsibility of news gatherers, well one thing is certain, if
the experts cannot agree here, then how come you might expect the media to.
Perhaps we need a new scientist or Nature magazine channel as they do seem
to often expose the pros and cons and it explains a lot why Governments
often get it wrong.
Brian

--

This newsgroup posting comes to you directly from...
The Sofa of Brian Gaff...
briang1@blueyonder.co.uk
Blind user, so no pictures please
Note this Signature is meaningless.!
"Pamela" <pamela.private.mailbox@gmail.com> wrote in message
news:XnsADDFC32DB164637B93@144.76.35.252...
>I suspect former politics student Tom Harwood of GB News will not be
> presenting a video pointing out death rates continue to rise to levels
> seen only during the lockdown.
>
> I wonder why not. Doesn't Tom and his sidekick have any more graphs to
> show how the media, the NHS, Public Health England, the statistics office
> and government have got their data "180 degrees" wrong?
>
> Hang your head in shame, Tom, for encouraging gullible viewers to let
> their guard down and risk potentially deadly consequences.

On 11/11/2021 20:47, Andy Burns wrote:
>
> Java Jive wrote:
>>
>> Did they report today that, contrary to their previous slagging off of
>> other media, the case rate went back over 40k?
>
> It's fluctuated up and down since the middle of July, and no doubt will
> continue to do so ... probably half-term related in recent 7-10 days.

Yes, as I mentioned myself, the daily case count fluctuates, but that's
not really the point, is it? Bob has been lauding GBNews for what Bob
claims is the accuracy of its reporting, so surely if the case count
jumped fairly dramatically yesterday, as it did, shouldn't they have
reported it honestly, and did they actually do so?

--

Fake news kills!

I may be contacted via the contact address given on my website:
www.macfh.co.uk

Java Jive wrote:

> Bob has been lauding GBNews for what Bob claims is the
> accuracy of its reporting, so surely if the case count jumped fairly
> dramatically yesterday, as it did, shouldn't they have reported it honestly, and
> did they actually do so?

If you're going to report averages, then report averages, any spikes will work
their way into the average in due course, that's the whole point.

To report the "average is XXX" and then add "but today it spiked to ZZZ" is
hyperbole, and seems to be the BBC+Guardian tactic.

On 08:20 12 Nov 2021, Andy Burns said:

> Pamela wrote:
>
>> On 19:21 11 Nov 2021, Andy Burns said:
>>
>>> Pamela wrote:
>>>>
>>>> I suspect former politics student Tom Harwood of GB News will not
>>>> be presenting a video pointing out death rates continue to rise
>>>> to levels seen only during the lockdown.
>>>
>>> Quite right because the case rate, admission rate and death rate
>>> have all fallen in the past 7 days
>>>
>>> <https://coronavirus.data.gov.uk>
>>>
>>> What of the 200,000+ cases/day that were predicted in mid-July?
>>
>> The current death rate is 192 per day. Perhaps your page is showing
>> raw data without 7-day smoothing? I'm using the chart called "UK:
>> coronavirus deaths per day" from the Guardian here:
>
> I'm using the .gov.uk page I linked above, do you even look at it?
>
> It shows each graph with a red highlight when things have been worse
> in the past 7 days and green when it's been better, all three graphs
> are green at the moment.

I saw the disparity with your page. I already mentioned the difference
between that and what the Guardian and BBC are reporting using
essentially the same data source. (The Guardian mentions a slightly
different current week 7-day measure, which I quoted).

The ONS site is clearly doing smoothing, otherwise you would see
weekend/weekday spiking which many other charts show (including the one
in the Guardian link I gave).

>> https://tinyurl.com/yd7v4zr8 which is captioned:
>>
>> "People who have died within 28 days of their first positive
>> test for coronavirus. Deaths assigned to the date they are
>> reported. Line shows daily average in a given week. Data:
>> data.gov.uk, updated 11 November, 2021"
>>
>> The BBC is reporting 195 deaths today, Thursday:
>> https://www.bbc.co.uk/news/uk-51768274
>
> Looking at one day compared the the previous day is generally a
> nonsense, it depends whether you choose date of death or date of
> reporting, there's such a 5weekday/2weekend pattern it only makes
> sense to look at weekly figures.

To address with Christopher Swanson's concern about data from a Monday or
Tuesday, I used Thursday's (7-day smoothed) data.

> Additionally there have been a few reporting issues recently, where
> data has not been reported on time, so one days spike on a graph may
> include two days deaths

In the video, Tom Harwood also makes an animated fuss about the Mirror
not pointing that out subsequently.

>> In his notorious video, Christopher Swanson tells Tom Harwood the
>> death rate is only 150. In fact the death rate is markedly higher
>> than that.
>>
>> https://www.youtube.com/embed/F_p4nvqZLQg
>
> I've been listening to that in the background while I type, in fact
> I've just heard that report refer to the same data reporting issues
>
> And they also refer to the importance of using the rolling average,
> I really do think the media just big scary bad numbers ...
>
> GB News is calling it right AFAICT
>
>> Harwood's prime message is the media is using misleading data. He
>> then points to some minor variation between data sets and uses
>> these to draw the false conclusion that the pandemic is receding.
>> Oops!
>
> It's receded to the point we can live with it, vaccines are working,
> masks are not required in everyday settings, even the BBC a week ago
> said (asked?) if covid has peaked for this year, it's not over, it's
> here to stay but people need to be able to get on with life ...

It is premature to conclude: "It's receded to the point we can live with
it" at the same time deaths continue to increase to levels not seen since
the lockdown earlier in the year. The rate of death and hospitalisation
are set to increase over Christmas.

The BBC link I gave states: "Recent rise in daily deaths. There were 195
deaths within 28 days of a positive test reported on Thursday". That
seems a more balanced summary than Harwood's strident tones suggesting
there's a con being played on the public by all media (except his own).

On 12:32 12 Nov 2021, Andy Burns said:

> Java Jive wrote:
>
>> Bob has been lauding GBNews for what Bob claims is the accuracy of
>> its reporting, so surely if the case count jumped fairly
>> dramatically yesterday, as it did, shouldn't they have reported it
>> honestly, and did they actually do so?
>
> If you're going to report averages, then report averages, any spikes
> will work their way into the average in due course, that's the whole
> point.
>
> To report the "average is XXX" and then add "but today it spiked to
> ZZZ" is hyperbole, and seems to be the BBC+Guardian tactic.

Take another look at the Guardian link I gave you. It shows daily
deaths and also draws 7 day average. It's the chart called "UK:
coronavirus deaths per day". In addition, it currently gives a daily
unsmoothed "spike" figure of 195 deaths.

https://tinyurl.com/yd7v4zr8

The BBC is equally fair by presenting both in its chart "UK Daily
Reported Deaths with Coronoavirus".

https://www.bbc.co.uk/news/uk-51768274

By contrast, the ONS link you gave shows only one set of data and it
is smoothed even though they don't say it. (Click the link "All cases
data" to see the data presented as the Guardian and BBC show).

https://coronavirus.data.gov.uk

Pamela wrote:

> Take another look at the Guardian link I gave you. It shows daily
> deaths and also draws 7 day average. It's the chart called "UK:
> coronavirus deaths per day". In addition, it currently gives a daily
> unsmoothed "spike" figure of 195 deaths.
>
> https://tinyurl.com/yd7v4zr8

Yes, I've seen it, if you want averages, don't quote spikes, wait 3 days and the
spike will have been worked into the average, they never emphasise the dips
which happen every Sunday and Monday ...

Pamela <pamela.private.mailbox@gmail.com> wrote:

>
> It is premature to conclude: "It's receded to the point we can live with
> it" at the same time deaths continue to increase to levels not seen since
> the lockdown earlier in the year. The rate of death and hospitalisation
> are set to increase over Christmas.
>

But is it? Death rates lag case rates by around a month. If you look at the
government data graphs this is apparent. I expect the death rate to be on a
downwards trend soon.

On 11/11/2021 19:21, Andy Burns wrote:
>
> <https://coronavirus.data.gov.uk>
>
> What of the 200,000+ cases/day that were predicted in mid-July?

The case rates in that source always underestimate, as they only record
people who get tested (and if only doing lateral flow, actually reported
a positive test). The figure you should be using is the ONS figure,
which for week before last was around 2% of the population being
infected on any one day.

That measure differs from the dashboard cases, not only in terms of
being more complete, but also in that people could have been infected on
any number of days up to the maximum length of time during which people
test positive.

Assuming 10 days, that is over 100,000 cases a day, and it is even
possible that it came to more than 200,000, if say it turns out that
average time that a positive result would be achieved is 6 days. I'm
not aware that any of the modelling actually models the proportion of
infected people who get and report a test.

(One also has to remember that the models produce a very wide error
band, which is dependent on things like how careful people are in
practice, but sensible governments have to plan for reasonable worst
case scenarios, not gambling on median or better outcomes.)

David Woolley wrote:

> The figure you should be using is the ONS figure

I agree the ONS produce some useful/interesting data, but it is nowhere near as
easy to discover as the dashboard, and not refreshed anywhere near as often.

On 17:05 12 Nov 2021, David Woolley said:

> On 11/11/2021 19:21, Andy Burns wrote:
>>
>> <https://coronavirus.data.gov.uk>
>>
>> What of the 200,000+ cases/day that were predicted in mid-July?
>
> The case rates in that source always underestimate, as they only
> record people who get tested (and if only doing lateral flow,
> actually reported a positive test). The figure you should be using
> is the ONS figure, which for week before last was around 2% of the
> population being infected on any one day.
>
> That measure differs from the dashboard cases, not only in terms of
> being more complete, but also in that people could have been
> infected on any number of days up to the maximum length of time
> during which people test positive.
>
> Assuming 10 days, that is over 100,000 cases a day, and it is even
> possible that it came to more than 200,000, if say it turns out that
> average time that a positive result would be achieved is 6 days.
> I'm not aware that any of the modelling actually models the
> proportion of infected people who get and report a test.
>
> (One also has to remember that the models produce a very wide error
> band, which is dependent on things like how careful people are in
> practice, but sensible governments have to plan for reasonable worst
> case scenarios, not gambling on median or better outcomes.)

I am getting a bit confused. Isn't the ONS survey the one which uses a
randomly seelcted sample in the community (as opposed to self-reported
cases) to determine the infection rates?

Pamela wrote:

> I am getting a bit confused. Isn't the ONS survey the one which uses a
> randomly seelcted sample in the community (as opposed to self-reported
> cases) to determine the infection rates?

I think that's the Imperial College "REACT" study?

Tweed wrote:

> Pamela wrote:
>
>> It is premature to conclude: "It's receded to the point we can live with
>> it" at the same time deaths continue to increase to levels not seen since
>> the lockdown earlier in the year. The rate of death and hospitalisation
>> are set to increase over Christmas.
>
> But is it? Death rates lag case rates by around a month. If you look at the
> government data graphs this is apparent. I expect the death rate to be on a
> downwards trend soon.

It is already (only just).

But the case rate is probably about to tilt upwards again, so maybe more deaths
will follow from that in a couple of weeks time, unless most of the new cases
are school kids and students in case very few of them will die ...

Since Pamela has reminded up of the REACT study, their most recent one-liner
executive summary is

"High and rising prevalence of SARS-CoV-2 infection
in England from end of September 2021 followed by a fall
in late October 2021"

On 15:59 12 Nov 2021, Tweed said:

> Pamela <pamela.private.mailbox@gmail.com> wrote:
>
>>
>> It is premature to conclude: "It's receded to the point we can live
>> with it" at the same time deaths continue to increase to levels not
>> seen since the lockdown earlier in the year. The rate of death and
>> hospitalisation are set to increase over Christmas.
>>
>
> But is it? Death rates lag case rates by around a month. If you look
> at the government data graphs this is apparent. I expect the death
> rate to be on a downwards trend soon.

I hope so but the connection between case rates and death rates has a
lot of confounders: such as non-reporting of cases, susceptibility in a
given demographic, new pharmaceutical interventions, etc.

Interesting to see case rates among schoolchildren are dropping. This
drop will raise the ratio of cases-to-deaths, as almost schoolkids
died from infection.

On 13:10 12 Nov 2021, Andy Burns said:
> Pamela wrote:
>>
>> Take another look at the Guardian link I gave you. It shows daily
>> deaths and also draws 7 day average. It's the chart called "UK:
>> coronavirus deaths per day". In addition, it currently gives a
>> daily unsmoothed "spike" figure of 195 deaths.
>>
>> https://tinyurl.com/yd7v4zr8
>
> Yes, I've seen it, if you want averages, don't quote spikes, wait 3
> days and the spike will have been worked into the average, they
> never emphasise the dips which happen every Sunday and Monday ...

The intriguing thing is the Guardian's chart caption may be suggesting
a "current week" average rather than a rolling 7-day average.

I wonder if that accounts for some of the difference with the ONS.

On 17:51 12 Nov 2021, Pamela said:

> On 15:59 12 Nov 2021, Tweed said:
>
>> Pamela <pamela.private.mailbox@gmail.com> wrote:
>>
>>>
>>> It is premature to conclude: "It's receded to the point we can live
>>> with it" at the same time deaths continue to increase to levels not
>>> seen since the lockdown earlier in the year. The rate of death and
>>> hospitalisation are set to increase over Christmas.
>>>
>>
>> But is it? Death rates lag case rates by around a month. If you look
>> at the government data graphs this is apparent. I expect the death
>> rate to be on a downwards trend soon.
>
> I hope so but the connection between case rates and death rates has a
> lot of confounders: such as non-reporting of cases, susceptibility in a
> given demographic, new pharmaceutical interventions, etc.
>
> Interesting to see case rates among schoolchildren are dropping. This
> drop will raise the ratio of cases-to-deaths, as almost schoolkids
> died from infection.

s/cases-to-deaths/deaths-to-cases/ :)

In article <smm6tf$rbi$1@dont-email.me>,
David Woolley <david@ex.djwhome.demon.invalid> wrote:
> On 11/11/2021 19:21, Andy Burns wrote:
> >
> > <https://coronavirus.data.gov.uk>
> >
> > What of the 200,000+ cases/day that were predicted in mid-July?

> The case rates in that source always underestimate, as they only record
> people who get tested (and if only doing lateral flow, actually reported
> a positive test). The figure you should be using is the ONS figure,
> which for week before last was around 2% of the population being
> infected on any one day.

> That measure differs from the dashboard cases, not only in terms of
> being more complete, but also in that people could have been infected on
> any number of days up to the maximum length of time during which people
> test positive.

> Assuming 10 days,

Having now cleared the 10 day hurdle, I'm told not a get another test for
90 days sice there will be traces left in my system.

--
from KT24 in Surrey, England
"I'd rather die of exhaustion than die of boredom" Thomas Carlyle