NWR new zero tolerance covid thread

ONS bulletin

What's in the bulletin?​


  • In England, the percentage of people testing positive for coronavirus (COVID-19) continued to increase in the week ending 31 December 2021; we estimate that 3,270,800 people in England had COVID-19 (95% credible interval: 3,163,500 to 3,377,500), equating to around 1 in 15 people.

2 findings:
Number of cases published daily is related to number of tests,
Number of people having COVID much higher (3.27 million in England only) and still increasing.
For the time being, no real decrease in casualties (one would expect as weakest people must have died by now)
Too early to call an end to Covid-19 but, obviously, growing number of people creating antibodies will bring future decrease, so there is hope.
Shortage of tests will bring decrease in number of cases anyway
 
This is two weeks old Antoine. The latest stats show a clear decline in cases and plateauing of hospitalisations. In London there is a pretty steep decline in cases and hospitalisations. Sadly deaths will keep going up for a further few weeks but will then it seems they will decline as rapidly as the hospitalisations. Although obviously not unmitigated good news it appears the UK is in much ruder health compared to most of our friends across the straights of Dover and beyond. We test more and have better vaccination rates so hopefully we can ride out this storm fairly quickly. It does feel like quite a turning point with the Government likely to introduce legislation about living with Covid and moving on from an emergency footing in the next month.
 
These are ONS latest statistics, they do take time to create.
...And numbers were still growing.
Indeed, numbers are expected to be going down in London ...
...but Number of deaths been over 300 for a few days now...(you'll tell me there is a lag time...).
Of course, you are right, Covid have always been by wave and this will be no different. Numbers will go down.
Still a lasting peak over 300 deaths a day is quite a lot with most people at risk vaccinated... and weakest people already gone...

Shortage of test kits do indeed lead to less recorded cases. I hope this is not the only cause.
 
Last edited:
Still a lasting peak over 300 deaths a day is quite a lot with most people at risk vaccinated... and weakest people already gone...
There have not been over 300 deaths a day since Feb 2020 Antoine. Please see previous posts about the rather important difference between the date reported and the date of the actual death. The date reported is a really misleading figure.
 
Gareth! Come on, Of course not,
I am talking about the peaks, not for a year! We are having a peak at the moment. Peak early 2021 went up to 1500 a day with people unvaccinated. We now have a new peak at well over 300 a day. Look at the ONS websites, you will realise COVID comes by waves which last a few weeks, not years. And waves mean numbers go down, so they will.

300 a day with people vaccinated and 28 days threshold is not very reassuring.
 
There has been no days with deaths over 300 in one day since February 2020. The last month is below. Antoine you need to look at the date of the actual death not when reported. I mean yes they are around 200 a day on average but not over 300.

It does seem the amount of people this confuses is very high. I really wish the Government would remove it!
Screenshot 2022-01-13 at 18.21.25.png
 
Well...if 1 in 15 people has covid, which is what we are being told, then I would expect 1 in 15 deaths to have covid on the death certificate, whether covid kills them or not.

Based on 2019 (pre-covid) numbers published by Gov.uk, I would expect on average 96.563 people to die each day (using the British decimal point and not the Euopean comma) with covid on the death certificate, and not be killed by covid (if the non-covid death rates in 2022 are comparable with total deaths in 2019, where covid didn't exist. Leaving the difference which are real covid deaths not imaginary ones.

Does that make sense?
 
Is this true? we were rather low down the rankings a few weeks ago, in spite of repeated claims to the contrary, but perhaps the booster programme has improved things.
The times and others continue to quote the number with 1 shot which skews the statistics towards those vaccinating children.
 
Gareth,

There are at least 3 sets of data in UK so ... let me clarify my numbers

- I took the Guardian number for deaths and it relates to UK: 393 yesterday and 335 today (and was over 300 the day before)... they indeed are a counting for death within 28 days of testing positive for the UK (and new medicines allow a lot of people with Covid to survive later these days...as you know)
- The ONS data on number of people having Covid was about England, hence the confusion.
- Your graph illustrates very clearly that we are peaking (we went from 100 to 200 daily deaths within a few weeks) and obviously, the grey bars do show incomplete data. (not sure what your source is).

2 or 300 deaths a day with full vaccination is very very high indeed by any standards. Especially after 2 years (having culled the most fragile and after vaccination and booster for many)

This was basically my message. What is yours? Pandemic over?
 
Gareth,

There are at least 3 sets of data in UK so ... let me clarify my numbers

- I took the Guardian number for deaths and it relates to UK: 393 yesterday and 335 today (and was over 300 the day before)... they indeed are a counting for death within 28 days of testing positive for the UK (and new medicines allow a lot of people with Covid to survive later these days...as you know)
- The ONS data on number of people having Covid was about England, hence the confusion.
- Your graph illustrates very clearly that we are peaking (we went from 100 to 200 daily deaths within a few weeks) and obviously, the grey bars do show incomplete data. (not sure what your source is).

2 or 300 deaths a day with full vaccination is very very high indeed by any standards. Especially after 2 years (having culled the most fragile and after vaccination and booster for many)

This was basically my message. What is yours? Pandemic over?
Antoine it is simply that you don't seem to understand the data. There were not 393 deaths yesterday, nor 335 today. I and others have explained the difference between deaths by date and deaths by date reported, the latter of which you use. It is an absolutely useless statistic. As previously put in a post above by your count Covid takes time off for the weekend, often on Mondays and rarely works a bank holiday. You are misrepresenting the data by saying there were 393 deaths yesterday. There weren't.

As for my message you don't need to put it as a binary. There is not one option between deaths being very very high and the pandemic being over. Perhaps there is a bit more nuance. I would say the deaths are low for this time of year and in comparison to the previous year. I think that is a fair comparison. They are also very low compared to what SAGE was predicting for this time.

In fact if you compare the modelling from SAGE assuming a 50% decline in death rates for Omnicron vs Delta (which most people thought was a good basis to go on including myself) then you'll see they were modelling 2,276 deaths today, whereas once the final figures are in it will likely be less than 10% of that.
Screenshot 2022-01-13 at 20.34.44.png

In fact if you put in deaths at 10% of Delta you see we are still tracking at approx 50% (475 predicted for 13th Jan). This is really good news. It means Omnicron is very likely far less severe than Delta. And if the numbers for cases overall continue to go down then we should hit the peak soon and will not even approach the 10% peak of 591 predicted by Sage.
Screenshot 2022-01-13 at 20.37.40.png
Antoine if you click here you can see the deaths by date as reported every day by HMRC. The last four days are incomplete but you can see accurate data. The Guardian et al simply use whatever stat fits their narrative regardless of how accurate it is.
 
Last edited:
I think what Antoine is saying is that even if it's 'only' 200 that is still a lot, which seems a perfectly reasonable position to me.
It is curious that models which predict the possibility of an even larger number of deaths should be so criticised. It is responsible to consider and if possible ameliorate that possible outcome; and if a gamble that it wouldn't happen happily turns out to be successful it remains just that, a gamble.
 
I think what Antoine is saying is that even if it's 'only' 200 that is still a lot, which seems a perfectly reasonable position to me.
It is curious that models which predict the possibility of an even larger number of deaths should be so criticised. It is responsible to consider and if possible ameliorate that possible outcome; and if a gamble that it wouldn't happen happily turns out to be successful it remains just that, a gamble.
I agree Tom over 200 is a lot and a great loss. But it is not nearly twice that (at 393). Nor is it very very high. We are in a much better position than last year. As for modelling I realise I am super critical. But when the data used to help HMG decide what restrictions to impose, lift, keep et al it is imperative that the modelling at least bears some resemblance to the reality that emerges otherwise it brings into the question the accuracy of their data and method they are using as inputs. Of course it will never be spot on but to be out on every count by a minimum of more than 50% is worrying.
 
I don't think it's worrying at all-surely the important thing is to act on the worst likely scenario, not the best. What happens next time if a variant just as contagious as Omicron emerges but which is ten times more deadly? there needs to be thought given to this very real possibility.
 
I don't think it's worrying at all-surely the important thing is to act on the worst likely scenario, not the best. What happens next time if a variant just as contagious as Omicron emerges but which is ten times more deadly? there needs to be thought given to this very real possibility.
Surely the worst case scenario needs to be realistic. Otherwise it is not a scenario. Otherwise the worst case scenario would be the universe ceasing to exist. But obviously that is not included as it is not realistic.

And it should surely have been that the best case was much lower than where we are now. I mean ideally the modelling would show a middle option which they hope with their data projected forward is the most likely and then a worst and best case. I mean that is what I studied when I did stats and modelling for sociological qual and quant research back in the day. But it was a long time ago and I it was entirely out of my field.
 
Gareth,

Your 200 is average England only, my 300 for 3 days in a row is UK so both numbers are high and show an increase on previous weeks. This we agree.

I know and understand your point about reporting making irregular daily reports. But this does not change the picture.

The modeling was right: Max Forecast by SAGE indicated assuming Omicron as potent as Delta (it was clearly indicated and they expected this to be a worst case scenario as they hoped it would be less potent, it is unfair to accuse them of misleading, they indicated premise and result). It has been confirmed as less potent (thankfully) and more contagious as expected. It is relatively good news and consistent with what they said but media wanted the worst to make headlines..
 
Gareth,

Your 200 is average England only, my 300 for 3 days in a row is UK so both numbers are high and show an increase on previous weeks. This we agree.

I know and understand your point about reporting making irregular daily reports. But this does not change the picture.

The modeling was right: Max Forecast by SAGE indicated assuming Omicron as potent as Delta (it was clearly indicated and they expected this to be a worst case scenario as they hoped it would be less potent, it is unfair to accuse them of misleading, they indicated premise and result). It has been confirmed as less potent (thankfully) and more contagious as expected. It is relatively good news and consistent with what they said but media wanted the worst to make headlines..
I guess it depends on what you want modelling to do. As I said before I would want the modelling to offer a best guess at the most likely scenario and then a higher and lower from that. What we have ended up with across most of the SAGE modelling is a very high top end and a typically too high bottom end. That is not good modelling. As it means ministers will have taken the best case presented as optimistic and the middle case a perhaps the baseline.

I'm obviously not a modeller and I'd appreciate any insights people can give on this. But to me the modelling is simply wildly inaccurate and has been for a lot of the last twelve months. This has created unrealistic scenarios being presented as plausible when in fact they were likely not plausible at all.
 
To make an analogy if you were investing a significant sum in an investment portfolio and you received a brochure that modelled growth at 1000% to 300% annually as a range and then invested and at the end of the year the fund had actually returned 2% you would obviously be upset. Particularly if the fund had previously never preformed in that range and only averaged 1-6% return on average. You would of course say the modelling was terrible and wildly inaccurate.

How is this different? I'm not trying to be flippant. Just interested as essentially its just data inputs and outputs regardless of what that data is.
 
or vaccine technology continues to advance to the extent that currently debilitating illnesses can be vaccinated against.
This almost deserves its own thread. Not conclusive but - very important news the potential cause of MS


And...moderna is already working on a vaccine.


I have been, perhaps fairly, accused of excessive optimism in these parts. But truly - isn't science wonderful.
 
Top