Model used to evaluate lockdowns was flawed

[from the duuuhhh files~cr]


Research News

In a recent study, researchers from Imperial College London developed a model to assess the effect of different measures used to curb the spread of the coronavirus. However, the model had fundamental shortcomings and cannot be used to draw the published conclusions, claim Swedish researchers from Lund University, and other institutions, in the journal Nature.

WATCH: Three reasons why mathematical models failed to predict the spread of the coronavirus –

The results from Imperial indicated that it was almost exclusively the complete societal lockdown that suppressed the wave of infections in Europe during spring.

The study estimated the effects of different measures such as social distancing, self-isolating, closing schools, banning public events and the lockdown itself.

“As the measures were introduced at roughly the same time over a few weeks in March, the mortality data used simply does not contain enough information to differentiate their individual effects. We have demontrated this by conducting a mathematical analysis. Using this as a basis, we then ran simulations using Imperial College’s original code to illustrate how the model’s sensitivity leads to unreliable results,” explains Kristian Soltesz, associate professor in automatic control at Lund University and first author of the article.

The group’s interest in the Imperial College model was roused by the fact that it explained almost all of the reduction in transmission during the spring via lockdowns in ten of the eleven countries modelled. The exception was Sweden, which never introduced a lockdown.

“In Sweden the model offered an entirely different measure as an explanation to the reduction – a measure that appeared almost ineffective in the other countries. It seemed almost too good to be true that an effective lockdown was introduced in every country except one, while another measure appeared to be unusually effective in this country”, notes Soltesz.

Soltesz is careful to point out that it is entirely plausible that individual measures had an effect, but that the model could not be used to determine how effective they were.

“The various interventions do not appear to work in isolation from one another, but are often dependent upon each other. A change in behaviour as a result of one intervention influences the effect of other interventions. How much and in what way is harder to know, and requires different skills and collaboration”, says Anna Jöud, associate professor in epidemiology at Lund University and co-author of the study.

Analyses of models from Imperial College and others highlight the importance of epidemiological models being reviewed, according to the authors.

“There is a major focus in the debate on sources of data and their reliability, but an almost total lack of systematic review of the sensitivity of different models in terms of parameters and data. This is just as important, especially when governments across the globe are using dynamic models as a basis for decisions”, Soltesz and Jöud point out.

The first step is to carry out a correct analysis of the model’s sensitivities. If they pose too great a problem then more reliable data is needed, often combined with a less complex model structure.

“With a lot at stake, it is wise to be humble when faced with fundamental limitations. Dynamic models are usable as long as they take into account the uncertainty of the assumptions on which they are based and the data they are led by. If this is not the case, the results are on a par with assumptions or guesses”, concludes Soltesz.


From EurekAlert!

4.7 12 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
December 29, 2020 10:12 am

I think UK’s Science as a whole is being devalued daily. They are getting self inflicted wounds that will be hard to heal.

I can say their scientific output level is lower than China’s ATM.

Reply to  Pauleta
December 29, 2020 2:40 pm

If you know anyone who thinks this is accidental, I have a like-new gluten factory I’ll be happy to sell them.

Questing Vole
Reply to  Pauleta
December 29, 2020 3:19 pm

Take a quick look into its future by watching this year’s Royal Institution Christmas Lectures.

Reply to  Pauleta
December 29, 2020 4:21 pm

No. I think that is not fair. In the end the pilot of state is flying almost completely blind, and knows it. They are running lockdown against death case and hospitalization rates. its a massively blunt instrument but it the only control hey have until and unless the vaccination works.

The mutation now is scaring them a lot.the science for what its worth says an accelerating infection rate leads to disaster in 4 weeks time if nothing changes.. They are changing the only thing they can.

The reality is no one knows and they know that. They have to however give the appearance of being masterful and in control, to reduce public panic.
They are not lying to you because there is no problem, but because they are scared of what might happen if you realised how big a problem it actually was.

Swedens much vaunted no lockdown now seems to have been a mistake of course.

Reply to  leo smith
December 29, 2020 7:44 pm

The evening news relayed the finding of another new strain of Covid-19. It seems that this “bug” is resilient enough to change its tactics to find new turf.

Reply to  Sara
December 29, 2020 8:39 pm

doesn’t flu virus change yearly?

Reply to  Sara
December 29, 2020 9:27 pm

Last I looked, there were over 4,000 known variants. What’s one more, to anyone but a brain-dead TV viewer?

Reply to  MarkG
December 30, 2020 2:30 am

A UK study determined there were over 1700 variants of sars-cov-2 displaying changes in the spike protein itself , with thousands of others displaying less significant, random variations in other parts of the genome.

This is normal development of for an RNA virus, which are typically unstable because of their nature.

Viruses do not think and do not have “tactics”. It is natural random variation and evolutionary selection of those which happen to provide a replication advantage.

Old England
Reply to  Sara
December 29, 2020 11:37 pm

More than 20,000 Covid-19 variations found so far – 10,000 of those have been genome sequenced and mapped in the UK – and yet ‘scientists’ fearmongering about a new one with ‘higher’ infection rates are later forced to admit they have No Data to support that claim.

Reply to  leo smith
December 29, 2020 9:26 pm

Again, this disease has a 0.1% death rate in the general population. Almost no-one in the UK under 50 has died from it without other serious health problems. All of this nonsense would have ended months ago if they were actually following the science rather than a political agenda.

Reply to  MarkG
December 29, 2020 10:49 pm

Ended?…..didn’t end in Sweden.
Compare 3 places with similar population 5-6 million
Massachusetts 14000 dead, British Columbia 800 dead, NZ 25 dead
Different places give different results…..oh yes and one of those 3 places ‘ it’s all over’

Old England
Reply to  Duker
December 29, 2020 11:44 pm

Horrifying claims which seem to rely on manipulating the data; BUT when you dive into the official data (in the UK ONS and NHS) data you find that the ‘Deaths’ claimed from Covid-19 are mirrored by the REDUCTION in Deaths from other causes. In other words the average number of excess deaths has changed very little.

The data showing deaths from heart attacks, strokes, etc is dramatically Lower than the average over the last 10 years and the overall Excess Deaths tally is very little changed from the avreage – it is just that these deaths are being blamed on Covid-19.

Politicians and ‘scientists’ have to try and justify draconian policies that are destroying business and society.

Reply to  Old England
December 30, 2020 12:10 am

I would like to see the data for this. It’s fine to make such claims but who is saying it, and where is the data to back it up? And, is the data trustworthy?
I am not suggesting your claims are wrong but I don’t believe anything ANYONE says about covid 19 at the moment because you can find someone saying whatever you want. The vast majority is probably bullshit. It can’t all be right.
I suspect that we will have to wait a decade or more before the real truth comes out – when reputations and agendas are no longer at stake.

Reply to  Old England
December 30, 2020 12:11 am

Exactly. The only metric which counts is Excess Mortality, and currently in the UK it is barely above the average of the last five years.

Reply to  Graemethecat
December 30, 2020 4:11 am

Sorry, Graeme, but you’re wrong! Look at the ONS figures for the “second wave”. From end-October the total number of weekly deaths has exceeded the average by at least 10%. The extent to which deaths exceed the five-year average is the definition of “excess” deaths.

Starting from the same date the number of Covid-related deaths as a % of the total has been 13,16, 20, 21, 24, 23, 22, 23. For every one of those weeks England & Wales deaths would have been below the five-year average but for Covid.

I agree that government (and “scientific”) handling of the response has been questionable — to say the least — but we shouldn’t downplay the seriousness of the virus itself.

Just for fun I ran the figures for three age groups (25-29, 60-64, 80-84) for the weeks ending 6/3, 24/4, 27/11. As a % of total deaths these were:

6/3: 0.35 4.4 16.5
24/4 0.15 4.2 17.1
27/11 0.32 4.8 16.0

It’a only a small and random sample but there is very little evidence to suggest that this last year has had much influence on the death rate of different age groups! Draw what conclusions you like!

Reply to  Old England
December 30, 2020 2:35 am

The official definition of “COVID death” in UK is patient has covid “mentioned” on the death cert. or had a +ve test result in the 28 days before death.

So very clearly this is “death with” or “death shortly after getting better” and not death from covid.

Mark BLR
Reply to  Old England
December 30, 2020 4:17 am

… when you dive into the official data (in the UK ONS …

In other words the average number of excess deaths has changed very little.

Weekly deaths for “England + Wales” are available from the ONS webpage below (up to “Week 51” at the time of posting) :

Data to construct the “Last 5 years / 2015-2019 average” to calculate “Excess” deaths is also available.

My “quick and dirty” calculations show a “Wave 1” from weeks 12 to 24 with 59,345 excess deaths, and an ongoing “Wave 2” since week 33 with another 16,570 (or so) excess deaths so far.
NB : Week 36 had a negative “excess” number, roughly -1440.

75K+ =/= a “very little” number.
It isn’t 510K either.

PS : Those “excess deaths” are currently being assumed to be 100% due to COVID-19. The raw ONS data simply “counts bodies”, it doesn’t assign blame beyond “Deaths where COVID-19 was mentioned on the death certificate” (74,774 cumulative to week 51, “75K-minus” versus “75K-plus” …).

PPS : “… you find that the ‘Deaths’ claimed from Covid-19 are mirrored by the REDUCTION in Deaths from other causes.”

You may “find” that.

I don’t.

Reply to  Mark BLR
December 30, 2020 12:10 pm

Week 36 was a bank holiday. This is a regular reporting artifact.

Reply to  Duker
December 30, 2020 2:38 am

…..oh yes and one of those 3 places ‘ it’s all over’

Really? NZ is not back to normal and now cowering in total isolation with a population almost entirely naive to the virus and it high risk from a single case.

Also comparing a rural population like NZ to the most populous state in NE USA is ill-informed and stupid if it is not disingenuous.

Old England
Reply to  MarkG
December 29, 2020 11:45 pm

In the UK the NHS figures show that just some 350 people under the age of 60 have died from Covid-19 … a fraction of the number who have been murdered or died in car accidents.

Reply to  MarkG
December 30, 2020 4:44 am

item I read had a UK doc saying the total of NO other health issue deaths was a tiny 338 or somesuch
from a pop of many millions thats damned low
and she copped HUGE flack for her factual statement

Vincent Causey
Reply to  leo smith
December 30, 2020 12:39 am

Why is Sweden’s lockdown a mistake if it produce epidemiological results no worse than lockdown? Or are you are you suggesting it needs to be better than lockdown in order to be justified? It sounds like it. But that only makes sense if locking down has no problems attached, such as economic damage, loss of liberty, undiagnosed cancer, depression and suicide. Given the huge damage lockdowns cause, I would say that even if Sweden’s epidemiological outcome was slightly worse, it would still have been shown to be the correct strategy.

Reply to  leo smith
December 30, 2020 2:06 am

it the only control [t]hey have until and unless the vaccination works.

No, there are therapeutics which they refuse to consider.

There has been a massive, organised lobby against the use of HCQ as a prophylactic and as precocious treatment, before the patient is in a critical condition, at which point it is no longer an appropriate treatment for the symptoms of major concern.

All tests including HCQ which were initially included in the DISCOVER program where specifically designed NOT to include the associcated antibiotic and were only to be tested on critically ill patients. ie they were designed to fail w.r.t HCQ.

Further more even these tests were stopped when the blatantly fraudulent Lancet paper was published. This also resulted in a ban on using HCQ for COVID treament in France.

Neither of these decisions were reversed when it was recognised that the paper was fraudulent and it was retracted by The Lancet.

So, if it is the only tool they have, that is by design. You can then ask why they would chose to contrive a sitation where the entire population and the economy is held hostage to a situation where multi-billion pound/dollar/euro contracts are the only way forward.

Prof Raoult has just published a meta study of all papers published about HCQ/COVID and established that 100% of the negative conclusions were published by authors with a conflict of interest by links to Gilliad, the manufacturer of Remdesivir , now proven to be useless against COVID just after EU commision bough 1 billion euros worth ( at a higher price than Belgium managed to negociate for an order of 1/10 the volume of doses ).

Massive instutionalised corruption of the medical bureaucracy and government is certainly a major factor. Though, like climate, there are powerful globalist influences too.

Imagining that half-tested vaccines are the only too available is incredibly naive and unrealistic.

Reply to  leo smith
December 30, 2020 4:10 am

They aren’t flying blind, they’re driving headlong towards the cliff and pushing as hard on the accelerator as they can.

Paul Penrose
Reply to  leo smith
December 30, 2020 9:30 am

They aren’t scared in the way you imply; they don’t really that much about the deaths. What they are scared of is being blamed for the deaths, and thus not getting reelected. The last thing a politician wants to have said about them is that they didn’t “do something” in the wake of a crisis, preceived or real. So they are more than willing to take actions that they know will have negative side effects, even if those actions don’t even address the real problem, just as long as they are seen as “doing something.” Trump stated (correctly) that the “cure” should not be worse than the disease, but most politicians (at least privately) do not agree. Anything that makes them look good and/or helps them avoid blame is perfectly acceptable.

Reply to  leo smith
December 30, 2020 6:40 pm

<blockquote>Swedens much vaunted no lockdown now seems to have been a mistake of course.</blockquote>

No, absolutely not. Accepting the PTB’s metric of “deaths” as a measure of the impact of covid isn’t only mad, it’s bad; utterly immoral, and criminal. All lives are always ‘lost”, no exceptions. The only rational measure of the impact of ‘the virus’ is Quality-Adjusted Life-Years lost. By this measure the “lockdowns” have done at least a thousand times the damage of no “lockdown” and probably hundreds of thousands of times the damage.

The New Zealand economy has been utterly destroyed; it is running on borrowed and printed money; the bill will be coming in for years(and the bill is not just monetary but early death as well. Economic damage causes early death.). Young lives have been destroyed so that a few irresponsible old people can “live” their useless, meaningless, miserable lives a little longer.

PS I am old.

PPS Please send money to Jennifer Marohasy(say one-tenth of what I have. That would be just over USD 1000.)

Pat from kerbob
December 29, 2020 10:21 am

“With a lot at stake, it is wise to be humble when faced with fundamental limitations. Dynamic models are usable as long as they take into account the uncertainty of the assumptions on which they are based and the data they are led by. If this is not the case, the results are on a par with assumptions or guesses”, concludes Soltesz.

Could simply paste this quote again and again on this site

Joel O'Bryan
Reply to  Pat from kerbob
December 29, 2020 10:46 am

At 30 years into the climate scam, it no longer can be ascribed to innocent scientific naiveté about climate models from the CliSci community, but outright intentional deception. A deception that like the Trojan Horse it is carries forth a politcal agenda of anti-capitalism and diminished Western economic power.

Reply to  Joel O'Bryan
December 29, 2020 4:23 pm

The littlle boy cried COVIDWOLF and no one believed him…

Reply to  Pat from kerbob
December 29, 2020 2:15 pm

The same statement jumped out at me as well.

Joel O'Bryan
December 29, 2020 10:40 am

Just as with tuned climate models outputs, it is clear the ICL lockdown simulation results were what the model designers wanted to see. The models are political tools, not scientific tools … by design.

The Left in the US is going to repeat this COVID political deception once again with Dementia Joe and his 100 day COVID emphasis from January 20th. May 1st is day 120, which of course is when last Spring’s COVID surge began to wane. It waned not because of the lockdowns many states pursued, but due to natural forces at work with seasonal respiratory viruses with the arrival of nicer weather and people opening up windows, outside more, more sterilizing sunshine. But the societal costs of the lockdowns have been so severe to people’s economic security and education of our children that the offending politicians, for their credibility, must do everything they can to convince people the lockdowns (and masks) were necessary and worked.

Reply to  Joel O'Bryan
December 29, 2020 2:24 pm

Computers simply execute the instructions that a person programmed them with. To say that the results output by a computer arefabricated by the person programming the computer is a tautology. This is no problem as long as you treat the model results as a hypothesis, or a prediction. But when you start pretending that the model output is data, then you are fabricating data.

Here, that is exactly what they are doing. They are treating the output of a computer as evidence. Not a hypothesis to be tested or measured against real data, but as evidence itself. This is a misuse of computer modeling.

Reply to  Kurt
December 29, 2020 4:22 pm

I’m a programmer. You want a model, I can build it. What do you want it to prove? 😉

John Dueker
Reply to  Joel O'Bryan
December 29, 2020 6:39 pm

That’s exactly the problem scientists have forgotten the scientific method and have adopted the find the politically correct answer method.

Reply to  Joel O'Bryan
December 29, 2020 9:28 pm

And who’s going to destroy their state economy just because the Fake President says so?

Ben Vorlich
December 29, 2020 10:42 am

When I was a reliability engineer trying to identify the cause of a failure the rule was change one thing at a time to identify the issue. In the UK Boris, Whitty et al have been changing things willy nilly sometimes the same thing more than once a day. There’s no way on God’s Earth anyone will be able to know what was effective and what wasn’t.

Phil Rae
Reply to  Ben Vorlich
December 29, 2020 10:54 am


“There’s no way on God’s Earth anyone will be able to know what was effective and what wasn’t“

And that was probably the intention from the start, unfortunately!

Reply to  Phil Rae
December 29, 2020 5:17 pm

Spot on Phil. 55 years after WWII people are still argung over who won the war and how.
Was it the Americans, the Russians, British grit, John Wayne, the Spitfire, The atomic bomb, Radar, cracking enigma and lorenz codes, the batte of El alamein, or strategic bombing and the dam busters.
Or the fact that the allies in the end could mass produce weaponry at a far far greater rate than the axis powers.
In a crisis with imperfect knowledge you just act, And pray. According to the best information you have.

At Sandhurst you are taught to modify the battle plan as soon as new intelligence renders it obsolete, In politics that is called ‘dithering’ and ‘making U turns’ and ‘not giving clear guidelines’

Peole here, for reasons I can fully understand, are advancing hypetheses that the government is overegging the pudding for reasons of profit and politcal control . I would point out that an alternative hypothesis exists, that this time the government are truly scared, in a way they never were about climate change, because their actions have been immediate, drastic and as draconian as anything I have ever seen, I dont think they would risk the political backlash unless they were staining their pants.

Reply to  Phil Rae
December 29, 2020 6:58 pm

“People here, for reasons I can fully understand, are advancing hypotheses that the government is overegging the pudding for reasons of profit and politcal control ………..I don’t think they would risk the political backlash unless they were staining their pants.”
This is the nitty gritty of the problem and solution. Politicians, to me, often seem completely clueless when it comes to solving problems. The political process often works at cross purposes. All too often legislation languishes until some pol figures out an amendment with a solution to another problem, maybe even related in some way, but that is not required. All that is required is that they can point to the bill, ruling, or regulation and say “see, this problem will be solved by this legislation…..” Often the ‘junk’ in a bill is useless but blending the sausage makes it work.

With the pandemic it appears no one has been able to produce a useful explanation. Dynamic models, aka Climate models have to many variables and cannot generate enough detail to be calculable even on modern computers. Dynamic models for diseases seem to be suffering a similar problem. An “epidemic” such as the foxes vs. snowshoe hares that has a long, accurate history has been modeled and continues to give decent results for a fairly simple event. More complicated, wide-spread and much less well-followed seem to be much harder.

The current pandemic event is much more complicated- it appears to be a conglomeration of different epidemics that started at different times and under different circumstances- direct from China to the US, China to Northern Italy, several new epidemics in China, a spread through Russia, possibly direct infection of the UK, and possibly a spread through the Baltic states into Germany-France-Spain and possibly one through the Arab state>Turkey>central Europe.

In their good fortune most all of sub-Saharan Africa and perhaps Australia, and the Far East have been hit much less warm due to the normal high humidity and temperatures. They also have a malaria problem and many, many people were already using anti-malarial drugs such as hydroxychoroquine.

One possible solution is to continue to take great care to keep older, more vulnerable adults, particularly with co-morbidities, separated and as well-screened and treated with anti Covid medicines in addition to vaccination where appropriate. It would be a very sad outcome to save the children who are barely affected by allowing their grandparents to die- although I suspect some political movements would welcome the destruction of historic knowledge.

Then, in the general population start working the vaccine into the population and at the same time quashing all the destructive isolation programs that are destroying the economy, wasting money, and ruining the mental well-being of many adults and children. The idea that “isolating children and adults at home” is more effective than “isolating the infected where they can be treated an separated from healthy people” was LONG avoided. One main reason is that it never worked before the disease became epidemic. Another main reason is that many fewer people are infected than are still healthy. It takes a lot more resources to keep the healthy isolated. Another reason is that children and young people are very often barely affected at all and rarely spread the disease.

While I doubt the likely incoming Administration will be able to handle this epidemic and better, I believe they have many ideas that could make it much worse.

Reply to  Philo
December 30, 2020 1:44 am

It’s unfortunate, to say the least, that politicians are so often clueless at solving problems, because they are generally driven by the desire to be seen to be solving problems. They are, for the most part, falling over themselves to push the green cr*p precisely for that reason.

Reply to  Ben Vorlich
December 29, 2020 11:44 am

Has there been anything effective (other than totalitarian rule)?

Reply to  Scissor
December 29, 2020 4:23 pm

I think it depends on what it’s supposed to be effective at…

Reply to  Scissor
December 29, 2020 5:21 pm

Florida seems to be doing well by not locking down and not mandating masks, yet their population of at risk people is quite high. It’s almost like the disease targets Democrats, but it’s more likely that its the ineffective policy coming from Democratic politicians that makes the disease worse.

Skiing seems to be effective too. The ski resorts are open in California and I know of no case of COVID attributed to the sport. Unfortunately, there’s no bars or restaurants, no apres ski except for tailgating in the parking lot and the mask mandates are a pain and definitely will not be tolerable in the spring.

Reply to  Ben Vorlich
December 29, 2020 4:59 pm

Ben, We are not conducting an experiment here, for your edification. Or to prove the science. Of course our politicians are bumbling idiots, but take your tinfoil hat off. to date excess deaths this year are running at 70,000 and climbing. hospitalization rates are now higher than spring and the new strain is aggressive and more lethal. One thing the data does show is that the more contact there is the more infectious and the more lethal it is. with infection and death rates again increasing exponentially what else can they do but lockdown as hard as the death rates suggest they must? nitpicking over the models doesn’t save lives.

I am sure you would be the first to shout at me that correlation does not imply causation Just because ths pandemic is being modeled on a computer like climate change don’t mean that just because you feel fairly secure you are being lied to with climate change, for reasons of profit and political control, that the same is true of COVID 19.

I am sure the models are not perfect, I am also sure that if the death rate soared to a quarter of a million and the government had done nothing, they wouldn’t last a week.

At least we haven’t been told to eat bananas or inject bleach. One vaccine is being delivered to people, another is in final trial stages. The Army is out testing and vaccinatiing

I dont think that the government could in reality do other than what it has. The biggest mistake is how they have managed the information flow.The minister of health is a colorless apparatchik being groomed for prime minister, he has never managed anything in his life that required that he actually say or do anything definite or anything he wasnt told to.

And the whole media is against the government, because they dared to actually leave the EU. Just as they were against Trump, because he threw two fingers at their liberal agenda.

To use your analogy, you are doing your relaibility engineering in a plane full of passengers that is losing altitude over the Rockies. Never mind changing one thing at a time, you will all be dead. Change everything and anything until it makes a difference and then do more of it.

To date only social distancing/hygeine/ lockdown worked. Jury is out on the new vaccines. And masks. But masks are cheap and dont hurt. Everyone wears one because it’s no big deal, but no one really believes in them. So the politcal and medical balance is that masks are low risk political and financial cost, reassure people and might just do something,

Lockdown carries high political risk, but so too does unrestricted contact. Carp at the science and the models all you want, but be sure what the government is counting are hospital beds, body bags, opinion polls and economic costs…The ‘science’ is what they use to justify what they have to do to balance those

Reply to  leo smith
December 29, 2020 7:17 pm


Lets put your 70K excess into context. Of the 336K with the China virus stated as the COD and if the excess is only 70K, then 266K of them would have passed anyway. 2019 had 2.85M total and this year, it will be about 2.92M. Note that it also increased by about 15K from 2018 to 2019, so perhaps only 55K are actual ‘excess’.

What this doesn’t count are those whose excess COD was self inflicted, the result of domestic violence or the dramatic rise in gun violence in most major cities, especially single party Democratic cities. All of these causes have also increased as a result of these draconian policies.

Most of the excess are the elderly and those with pre-existing conditions, so it’s likely that it will all average out and 2021 and 2022 will probably have a negative excess, although a Biden administration is a wild card.

The real problem is that few of the restrictions put in place by the authoritarians are specific to protecting those most at risk and instead they force significant hardships on the vast majority of the population not at serious risk, if at any risk at all. In fact, Cuomo’s policies specifically harmed those at most risk resulting in a significant fraction of the excess.

December 29, 2020 10:55 am

Give the models more time. They’ll eventually find flaws in resultant data to prove them correct.

Reply to  markl
December 29, 2020 11:12 am

Actually, the models will be used to adjust the data so that there is no disagreement between the models and the data. (Unfortunately, not sarc)

December 29, 2020 11:07 am

You don’t say? Who would ever believe government and science lied? That’s right, I say it wasn’t flawed it was flat-out lying.

Reply to  bluecat57
December 29, 2020 5:18 pm

Is it lying if no one knows what the truth is?

Patrick MJD
Reply to  leo smith
December 29, 2020 9:55 pm

You sound like to me to be an apologist. We have known for sometime the models were wrong, but authorities are unwilling to admit the error and continue to promote practices that have been shown to not work.

Reply to  leo smith
December 30, 2020 5:24 am

Then there is no truth.

December 29, 2020 11:07 am

I know some other high profile models they could look into if they dare.

December 29, 2020 11:09 am

Does this model have any UN connections to defend it and beat back head-in-sand denialists?

December 29, 2020 11:42 am

Pretty sure we need a lockdown of Imperial College London

Ed Zuiderwijk
December 29, 2020 11:52 am

The last paragraph is saying GIGO in the kindest manner possible.

Pariah Dog
December 29, 2020 12:23 pm

Models are not science. They are simply masturbation in code form produced by people who seem to think that looking for real-world evidence for a hypothesis is no longer necessary. Any time i read an article and I see the phrase “models show” or similar, I stop reading and go to the bathroom to wash the jizz out of my eyes.

Reply to  Pariah Dog
December 29, 2020 5:42 pm

Science is all models. Computer ones are no better or worse than a line of differential equations. And a computer model is in the end just a way to take a lot of hypothetical partial differential equations and integrate them using stepwise iteration.

Climate models are perfectly valid, and if they had been correctly interpreted would have shown a decade ago that the original hypothesis, that CO₂ was the primary driver of modern warming, was false.

Instead they were used to obfuscate and to justify a political position that was otherwise untenable.

that is not the fault of the modelling process.

Computer models fail to deliver accurate predictions because either the situation cannot be successfully modeled at all – as in chaotic systems – the data fed in was simply inadequate or wrong , the model was too simple to take all relevant factors into account, or in fact it expresses a correlation that simply doesn’t match reality.

Climate models fail on ALL the above.

Pandemic models probably fail only on point 3.They are simply too simplistic.

Dont throw the baby out with the bathwater. Theoretical Sciece IS models and computerizing them oi do calculations no human could otherwise do, is in many case not the best tool, but the only tool we have to test the hypothetical models with- or in the case of chaos, show that understanding even exactly what the hypothetical model is does not in the end allow you to predict the outcome reliably.

There is nothing wrong with these models. The problem is with the humans interpreting them. They are not being used to validate hypotheses, but to justify political actions.

Rich Lambert
December 29, 2020 12:24 pm

ISO 9000 series standards are widely used in industry. They require validation and verification of computer programs. These standards are even more necessary when computer programs are used by government and should be required.

Reply to  Rich Lambert
December 29, 2020 5:43 pm

Yeah, I’ll try to remember that when I am strugglling to strap on a parachute as the plane erupts in flames about me…

a happy little debunker
December 29, 2020 12:43 pm

Thought I’d try out the attach image function.
If this works – let the meme wars begin….

Reply to  a happy little debunker
December 29, 2020 1:11 pm


Patrick MJD
Reply to  gringojay
December 29, 2020 9:52 pm

That mask is no good. It’s Norton!

December 29, 2020 12:45 pm

Seems the operating rational was an iteration of the USA Speaker of the House Ms. N. Pelosi’s word salad repurposed to become essentially: “We have to pass lockdown orders to find out what’s in it.” Part of the consequential hit on us is that this CCPvirus is so severe that most people have to be tested to find it.

December 29, 2020 1:30 pm

“In Sweden the model offered an entirely different measure as an explanation to the reduction – a measure that appeared almost ineffective in the other countries. It seemed almost too good to be true that an effective lockdown was introduced in every country except one, while another measure appeared to be unusually effective in this country”
Which measure appeared to be effective in the cas of Sweden?

Richard M
Reply to  François
December 29, 2020 4:10 pm

I’m guessing it’s called …. Spring.

Reply to  Richard M
December 29, 2020 5:46 pm

LOL. I think it was just an extension of normal Scandinavian puritanism. Noli mi tangere

December 29, 2020 1:41 pm

Computer models always show what they are programmed to show. We need lockdowns/masks/social distancing? Program a model based on the assumption that lockdowns, masks and social distancing work, and present that as proof. Pitiful.

December 29, 2020 2:00 pm

Oddly enough Imperial College receives significant funding from CCP.

Robert of Ottawa
December 29, 2020 4:05 pm

Ha! “scientific” model disproved by model. Let’s face it, all these computer models are BS. Yes, models have a purpose, to play with hypotheses, but their output is not data, and certainly not information.

Patrick MJD
December 29, 2020 9:51 pm

Sadly but not surprisingly, the ICL, and in particular, and formerly of the ICL, Ferguson, have not had a great track record in their computer model based predictions such as MERS, SARS1, Swine/Bird ‘flu and foot and mouth (Where 6 million beasts were slaughtered needlessly) have been monumentally wrong every single time!

December 30, 2020 12:26 am

In a recent study, researchers from Imperial College London….

So what “study” are the swedes criticising here?

Following the link to Urea Alert, and getting the swedish paper :

the first line begins “Flaxman et al.1 took on the challenge of estimating …”

As always the scientifically illiterate media studies freshmen at Urea Alert have to paraphrase things into obscurity.

The abstract of Flaxman et al says:

Our model calculates backwards from observed deaths to estimate transmission that occurred several weeks previously, allowing for the time lag between infection and death. We use partial pooling of information between countries,

Amidst the ongoing pandemic, we rely on death data that are incomplete, show systematic biases in reporting and are subject to future consolidation.

And that is what we are supposed to use to determine whether we need to continue to destroy our economy, civil rights and entire way of life ? Not convinced.

Already early UK data was an anarchic, inconsistent mess of independent, non coordinated bodies collecting and reporting data in different ways and timescales which were often not even self-consistent over time. Good luck with “pooling” that with other countries which adds further homogeneity to the base data.

When you start out with a mess like that any analysis can produce potentially spurious results, you just chose the ones you personal bias considers “right” and flash it through a “peer review” of colleges with similar biases. Climatology all over again.

Looking at the cross-correlation of deaths and positive test results ( aka “cases” ) to examine the delay between test results and eventual fatal outcome, we find there are significant differences in the lag. France and UK peak at about 14-15 days , Italy is only 10 days. Spanish data has serious structural inconsistencies over time leading to massive swings even after taking 7day means to reduce the effects of week variation in testing and reporting.

Good luck with “pooling” that lot.

Martin A
December 30, 2020 12:44 am

I thought it significant that the Imperial College team, when asked if their model had been validted, replied that it had ben validated because it agreed with other models.

Interesting that this is the method used to ‘validate’ new climate models.

December 30, 2020 3:22 am

An Imperial College computer model is flawed.

Anybody who thinks this is breathtaking news hasn’t been paying attention

December 30, 2020 4:09 am

Did they evaluate the effect of using alcohol-based hand sanitisers and covering any surface that didn’t run away with alcohol? I read an academic paper recently that showed that alcohol had no effect on SARS-COV-1, so it almost certainly couldn’t work on SARS-COV-2.

The Therapeutic Goods Administration lists 134 products that can disinfect Covid-19, but (at a quick glance) none of them are alcohol based. So why does everybody demand that we cover our hands in alcohol every time we go to the doctors, dentist, even the supermarket for Christ’s sake.

December 30, 2020 4:53 am

ok two things I found last night;
ignore the daft header
IN the rather long article is an answer to the question I asked but not answered- of taking a certain vax then the ability to use others/interaction risks etc
seems theres a deal done to pretty much make you ONLY take ONE version/brand recurring
and the costprice deals?
only last for the first immediate hysteria periods NOT ongoing
and then this gem;
oh and this;
The comments were made by WHO chief scientist Soumya Swaminathan during what appears to have been a virtual press conference held Monday.

“I don’t believe we have the evidence on any of the vaccines to be confident that it’s going to prevent people from actually getting the infection and therefore being able to pass it on”, says WHO Chief Scientist @doctorsoumya

— 🚨 (@disclosetv) December 28, 2020

crap sorry re the logo thingy;-(
it wasnt showing in the copy but whammo in paste

Patrick MJD
Reply to  ozspeaksup
December 30, 2020 6:16 am

Like heroin, (Windows 10 and O365) it’s a “subscription”. Once you are “hooked” you are hooked for life. While the vaccine is not required for actual health issues, which there are almost zero issues, it is for “compliance”. You have “Done your bit!”.

Double plus good Comrade!

I still have to get tested, my employer demands it. But what strain? The strain in the person in the building or the new one? I am SOOOO scared! It’s all BS!

December 30, 2020 5:51 am

Why are we debating something when it is obvious “nobody knows “.

The virus behaves as it chooses.

WUWT should leave this subject alone we are adding nothing.

Patrick MJD
Reply to  Jphn
December 30, 2020 6:17 am

A virus can choose?

Reply to  Patrick MJD
December 30, 2020 7:18 am

A pro-choice virus. How about that? 😉

December 31, 2020 7:13 pm

The CoV-19 panic started when the Imperial College London and the University of Washington’s Institute for Health Metrics and Evaluation plugged the data that was available into their models and both predicted 600,000 CoV-19 deaths in the UK and 2,000,000 in the U.S. The first problem is that nobody looked at the past records of predictions from either group. To be polite, their previous results were highly inaccurate and always on the high side. The second is that these are models generated in academia for which there are no consequences if they are wrong. The other problem, IMO, is that these academic models have never been subjected to a Validation and Verification examination. Would our alleged leaders, their minions, their families or those who they have convinced of their expertise knowingly ride in a high speed elevator, drive over a bridge, fly in a passenger aircraft or go on a thrill ride at a theme parkif they had been built with computer models that had never been properly scrutinized?

We can make the same arguments about Mannian Statistics™ or any of the CMIP models that are the foundations upon which the entire Climate Science™ edifice is built.

%d bloggers like this:
Verified by MonsterInsights