I think a mistake people often have is assuming there is a political agenda behind all things.

Theres corruption in the world, but sadly most things are much more mundane!

I know you mentioned that we’re at an impasse but can I just clarify that I’m not struggling to understand the point you’re making; it’s really not that difficult to understand.

Where we differ is that you think I want scenarios where no action is to be taken “non-event” whereas removing restrictions is absolutely an action and therefore it is important to include such data. If the government specifically requested not to include any data from SA and insisted that omicron be treated as fatal as delta for a specific query then I’m fine with that as long as these numbers aren’t pumped through the media and used to prep people for more restrictions.

Like this for example which I know BMCL will hate:

https://amp.theguardian.com/world/2021/dec/18/uk-scientists-curbs-covid-infections-omicron-deaths-restrictions-sage


Medley says that these aren’t predictions, that they are illustrations of possibilities and uncertainty. So why not include the SA data?


Anecdotally (yes I know that doesn’t hold much weight but perhaps others have encountered the same), I known a fair few people now to have caught covid recently and tested negative within 7 days which I had never heard before.

comment by Black Hawk (U16342)
posted 24 minutes ago
comment by bmcl1987 (U14177)
posted 1 hour, 34 minutes ago
comment by Sat Nav (U18243)
posted 36 minutes ago
comment by Black Hawk (U16342)
posted 5 minutes ago
And thats not to say SAGE or these scientists are right, or have been accurate with their findings.

A lot of it is just so incredibly hard to predict. But its still useful look at scenarios.
----------------------------------------------------------------------
But mate, these models aren’t just being looked at, are they?! They are very much determining the government’s course of action which is why I find their involvement in deciding what SAGE model to be utterly disgraceful.
----------------------------------------------------------------------
Latest update from SPI-M-O;

“There currently remains no strong evidence that Omicron infections are either more or less severe than Delta infections.”

That was on 15 December. As and when strong evidence emerges around the nature of omicron impact on hospitalisations then this will be reflected within scenario modelling.

Shout-out to BH, agree with all your posts today on this, and articulating my frustration with the approach from Nelson and his ilk better than I ever could
----------------------------------------------------------------------
Thanks! It's been a headache to try convey my point!

And yeah, the discussion between Nelson and Medley is interesting, as it shows the differences between rhetorical skills in discussion.

Nelson is obviously every good at communicating his points, and asking tricky questions. Medley on the other hand, shows the rhetoric skill of a scientist!

Often though people mistake rhetorical devices for salient points. Rhetorical devices are much flashier!!
----------------------------------------------------------------------

This is what people who can't debate say

Also - would note, report 49 which I believe BMCL is referring to puts vaccine efficacy between 55 and 80 per cent.

This is my issue with relying on such models - the difference between 55 and 80 per cent is the difference between needing a three month lockdown and virtually no measures at all

https://twitter.com/Simmons__/status/1472533727309963265

This is encouraging data from South Africa

And no - I am saying it directly predicts what happens in UK.

Older, fatter population in winter - I get it.

I am just genuinely happy South Africa is coping better than anticipated.

https://twitter.com/Sacha_Lord/status/1472575522525822976

Hard to disagree with this.

The worst of both worlds situation where we have an economy behaving as if its in lockdown, but case numbers as if the markets have re-opened needs to be fixed.

If the government wants lockdown - it should do it now, give businesses detailed support, and bear the brunt of the criticism.

The fact we are instead staggering into an inevitable lockdown over lots of umming and ahhing and fears of cancelling Christmas twice is quite embarassing.

Obviously I oppose lockdown and would have not scare everyone senseless over Omicron, but now that we have - and as we also already know SAGE will eventually get their way - let's actually plan a lockdown properly..

The most consistently accurate figures have most often been their lowest figures. Again, this isn’t their fault but christ…

The numbers their modelling pumped out in July, the closest to reality was on the basis of a small and gradual drop in precautionary behaviour over a 3 month period (always with default vaccine efficiency)

Their small change within a one month period had us with around 500 daily deaths in October


So with this latest one if the guardian for example wrote 200 deaths instead of 6000 what do we think the reaction/actions would be?

Has any of their worst case modelling come to pass? Maybe January 2021’s numbers? Maybe they were predicting 5,000 daily deaths for that period, can’t say I remember.

comment by NPE - Finding Timo (U22712)
posted 8 minutes ago
https://twitter.com/Sacha_Lord/status/1472575522525822976

Hard to disagree with this.

The worst of both worlds situation where we have an economy behaving as if its in lockdown, but case numbers as if the markets have re-opened needs to be fixed.

If the government wants lockdown - it should do it now, give businesses detailed support, and bear the brunt of the criticism.

The fact we are instead staggering into an inevitable lockdown over lots of umming and ahhing and fears of cancelling Christmas twice is quite embarassing.

Obviously I oppose lockdown and would have not scare everyone senseless over Omicron, but now that we have - and as we also already know SAGE will eventually get their way - let's actually plan a lockdown properly..
----------------------------------------------------------------------
They’ve broken you 😂

NPE rote that while reeching 4 the tin foil in the kitchen draw<rfl>

comment by Sat Nav (U18243)
posted 49 seconds ago
The most consistently accurate figures have most often been their lowest figures. Again, this isn’t their fault but christ…

The numbers their modelling pumped out in July, the closest to reality was on the basis of a small and gradual drop in precautionary behaviour over a 3 month period (always with default vaccine efficiency)

Their small change within a one month period had us with around 500 daily deaths in October


So with this latest one if the guardian for example wrote 200 deaths instead of 6000 what do we think the reaction/actions would be?

Has any of their worst case modelling come to pass? Maybe January 2021’s numbers? Maybe they were predicting 5,000 daily deaths for that period, can’t say I remember.
----------------------------------------------------------------------
Most of the predictions on deaths in January were fairly close only Cambridge was way off with 4000 daily deaths.

comment by 4zA - I'm down here most of the time, down under the world in a crawlspace (U22472)
posted 2 minutes ago
NPE rote that while reeching 4 the tin foil in the kitchen draw<rfl>
----------------------------------------------------------------------

Can't even spell the emoji properly you sloppy weirdo

comment by Sat Nav (U18243)
posted 1 minute ago
comment by Sat Nav (U18243)
posted 49 seconds ago
The most consistently accurate figures have most often been their lowest figures. Again, this isn’t their fault but christ…

The numbers their modelling pumped out in July, the closest to reality was on the basis of a small and gradual drop in precautionary behaviour over a 3 month period (always with default vaccine efficiency)

Their small change within a one month period had us with around 500 daily deaths in October


So with this latest one if the guardian for example wrote 200 deaths instead of 6000 what do we think the reaction/actions would be?

Has any of their worst case modelling come to pass? Maybe January 2021’s numbers? Maybe they were predicting 5,000 daily deaths for that period, can’t say I remember.
----------------------------------------------------------------------
Most of the predictions on deaths in January were fairly close only Cambridge was way off with 4000 daily deaths.
----------------------------------------------------------------------
Maybe the default vaccine efficiency is where they’ve been going wrong? I dunno.

I’m just sick of these reports then x, y and z headline “6000 deaths” and WFH becomes, no Christmas again, becomes lockdown again.

We had around 16000 hospitalisations this time last year, we’re at 7500 currently

Last year it go to 38500 hospitalisations so it more than doubled. If we more than double again then we’ll be at 17000? Does the NHS break down at that point? Can’t say I know

It sounds as if, like me, your problem is with the headline writers satters 😃

comment by bmcl1987 (U14177)
posted 39 seconds ago
It sounds as if, like me, your problem is with the headline writers satters 😃
----------------------------------------------------------------------
I am but I don’t appreciate the constant high figures that sage pump out along with the medium ones which are also generally way too high. I don’t think the govt should be involved at all other than saying “when is your latest paper coming out?” and I don’t like all the ‘these aren’t predictions’ - I’m sorry but that’s a pssyhole cop out too.

If these aren’t their predictions, what are their predictions? They always write “these aren’t predictions”.

Well you’re the experts, I want some fking predictions beeeeches

And now my Secretary has tested positive too

comment by Sat Nav (U18243)
posted 5 minutes ago
And now my Secretary has tested positive too
----------------------------------------------------------------------

You have a secretary?

comment by NPE - Finding Timo (U22712)
posted 4 minutes ago
comment by Sat Nav (U18243)
posted 5 minutes ago
And now my Secretary has tested positive too
----------------------------------------------------------------------

You have a secretary?
----------------------------------------------------------------------
Yeah. She’s tested negative and interestingly it only showed negative after 30 minutes. Last time I did a lateral flow I only waited like 15 mins max

*positive

https://www.spectator.co.uk/article/another-lockdown-will-only-fuel-the-cancer-crisis

This is an interesting rebuttal to the claim oft-repeated that cancer treatments won't be cancelled because of lockdown, and that lockdowns are what keep other services in then NHS running.

comment by Sat Nav (U18243)
posted 1 hour, 29 minutes ago
I know you mentioned that we’re at an impasse but can I just clarify that I’m not struggling to understand the point you’re making; it’s really not that difficult to understand.

Where we differ is that you think I want scenarios where no action is to be taken “non-event” whereas removing restrictions is absolutely an action and therefore it is important to include such data. If the government specifically requested not to include any data from SA and insisted that omicron be treated as fatal as delta for a specific query then I’m fine with that as long as these numbers aren’t pumped through the media and used to prep people for more restrictions.

Like this for example which I know BMCL will hate:

https://amp.theguardian.com/world/2021/dec/18/uk-scientists-curbs-covid-infections-omicron-deaths-restrictions-sage


Medley says that these aren’t predictions, that they are illustrations of possibilities and uncertainty. So why not include the SA data?


Anecdotally (yes I know that doesn’t hold much weight but perhaps others have encountered the same), I known a fair few people now to have caught covid recently and tested negative within 7 days which I had never heard before.
----------------------------------------------------------------------
So your problem is that they did not include a model which includes removing restrictions? For the sake of balance and fairness?

Of course removing restrictions are actions, but we are talking about decisions based on 'mitigating' high severity risks, and the respective models for them.

And removing measures is a very different kind analysis. It's not about mitigating number of cases, but about trying to reestablish normalcy in society.

I am not sure why though it would need to be included right now. And thats why I feel we are at an impasse.

I don't think it's persuasive to say that it adds useful information right now to the government. It would miss the point completely.

I started a 12 hour row

comment by Black Hawk (U16342)
posted 7 minutes ago
comment by Sat Nav (U18243)
posted 1 hour, 29 minutes ago
I know you mentioned that we’re at an impasse but can I just clarify that I’m not struggling to understand the point you’re making; it’s really not that difficult to understand.

Where we differ is that you think I want scenarios where no action is to be taken “non-event” whereas removing restrictions is absolutely an action and therefore it is important to include such data. If the government specifically requested not to include any data from SA and insisted that omicron be treated as fatal as delta for a specific query then I’m fine with that as long as these numbers aren’t pumped through the media and used to prep people for more restrictions.

Like this for example which I know BMCL will hate:

https://amp.theguardian.com/world/2021/dec/18/uk-scientists-curbs-covid-infections-omicron-deaths-restrictions-sage


Medley says that these aren’t predictions, that they are illustrations of possibilities and uncertainty. So why not include the SA data?


Anecdotally (yes I know that doesn’t hold much weight but perhaps others have encountered the same), I known a fair few people now to have caught covid recently and tested negative within 7 days which I had never heard before.
----------------------------------------------------------------------
So your problem is that they did not include a model which includes removing restrictions? For the sake of balance and fairness? ~~ No my point is that they should have included the data available from SA into their modelling as a separate line across the graphs.

Of course removing restrictions are actions, but we are talking about decisions based on 'mitigating' high severity risks, and the respective models for them. ~~why are not also including data into these models to show lower risks? Why always focus on the highest risk levels, it’s not just blind curiosity, these numbers are used for something

And removing measures is a very different kind analysis. It's not about mitigating number of cases, but about trying to reestablish normalcy in society.
~~ justifying the action of removing restrictions is absolutely based on the numbers

I am not sure why though it would need to be included right now. And thats why I feel we are at an impasse. ~~ because omicron is clearly the new kid on block on the available data should be included. There is already a huge amount of guesswork involved hence the regularly disclaimers they graffiti all over their documents so why not pump the omicron data through their models also? It’s still not a prediction, right?

I don't think it's persuasive to say that it adds useful information right now to the government. It would miss the point completely. ~~ miss what point? The point of mainly focusing on worst case, worse case and bad case only? And refusing to use the current data on omicron? Sorry but that makes no sense.
----------------------------------------------------------------------

Replies above mate

comment by NPE - Finding Timo (U22712)
posted 11 minutes ago
I started a 12 hour row
----------------------------------------------------------------------
We’re having a nice discussion

The omicron data on cases is in the latest iterations though. The specific issue you have on omicron is (and please correct if I’ve misinterpreted) is that you believe there was strong evidence at the time of the latest iteration of the model regarding hospitalisation rates (from SA) that you think should have been included in the model? And the latest SPI-o-m update included a statement that this info wasn’t strong enough at this point to include. The implication being from the statement that when evidence is more robust, assumptions will be adjusted. That is reasonable no?

It's ok - I still don't understand why the modelling didn't include scenarios where Omicron was milder.

I also don't grasp why if you are asked to build a model on Omicron severity you don't factor in 95 per cent of the country have antibodies alongside T cell & memory cell resistance.

It is also clear that if the models weren't mean as forecasts, SAGE should have been much less determined to push lockdowns on people and that politicians on both sides of the argument who have been present models as predictions should have been publicly corrected.