[Offwidth]

When COVID-19 was first declared a pandemic 16 days ago, the traditional models were useful in warning us what would happen if (literally) nothing were done to stop it. But scarcely two weeks later, we are (thankfully) a long way from nothing. Let’s go after this disease in the way that does the most good, and stop policing the paths to Pen-y-Fan.

Firstly, if superspreaders were the most important factor the rate of infection  growth would be way faster. Otherwise I agree the models need treating with a pinch of salt. In western nations the population infection rate  is pretty consistent at doubling between 2 and 3 days so any tuned model will meet that. I'd suspect superspreaders are more likely a major problem early in outbreaks and, if after things calm down, in secondary outbreaks.

On Pen-y-Fan the author is ignorant for reasons that have been discussed  time and time again here:  travel to it (and fueling and shop vists), congregations in car parks (potentally all touching the same gates) and the impact on the emergency services if someone breaks an ankle on the top. It also undermines the government message on social distancing, which is the same dumb attitude that many UK libertarians don't seem to get: Quillette is one of their favorite web communities.

[Stu Littlefair]

That quillette article is awful. The biggest problem is that it’s predicated on the straw man that the epidemiologists somehow haven’t noticed that the problem is stochastic and depends on the behaviour of individuals.

Most models used for prediction take this into account somehow. For example, the simple model I linked above uses network graphs where you can make some estimates for how many people are connected to an individual.

More detailed models, like the Imperial flu model that started the Uk lockdown, have a detailed model for transmission that goes way beyond Ro. That model has written into it the different rates of contact people experience at work, at home, when shopping etc, all stratified by age group. It also models Ro not with a single value but a statistical distribution with broad tails - ie it includes superspreaders.

It also includes the effect of people travelling between communities and into/out of the Uk in a stochastic way. Which is one of the reasons the government asked us to reduce unnecessary travel.

There are quite a lot of people publishing blogs and academic papers right now who should leave this kind of communication to people who work in the area. I’d put the author of that quillette article firmly in that category.

[teestub]

That quillette article is awful.

On a broader a point, is there ever anything good off their site? The people I’ve seen posting their articles before are those that think Jordan Peterson has some good ideas, so I’d sort of written of the whole thing as being toward the Alex Jones end of usefulness. 

[TobyD]

That quillette article is awful.

On a broader a point, is there ever anything good off their site? The people I’ve seen posting their articles before are those that think Jordan Peterson has some good ideas, so I’d sort of written of the whole thing as being toward the Alex Jones end of usefulness. 

Everything I've ever read on it has been more or less as you say.

It seems to me that thousands of people (millions!) are pretty bored and pretending to be armchair epidemiologists or statisticians; I really think that now is the time to just listen to some actual experts, do what they say and hope that we all come out of the other side with nothing worse than extreme protracted boredom and weight gain.

[Stu Littlefair]

Agreed, though I realise I’m guilty on that front. At least I’m not publishing my ignorance on a national blog. I mean:

“ ...even small changes in effective R0 can lead to wildly divergent values. That’s why the same British expert who very recently warned us of 500,000 COVID-19 deaths in Britain now says he expects fewer than 20,000.”

Utter bullshit, perhaps it’s because the numbers represent two different scenarios, one in which nothing is done and other in which current social distancing is successful?

[Doylo]

Can anyone explain why China with a population of over 1 billion had 3000ish deaths yet 20k would be seen as a ‘result’ here?

[Oldmanmatt]

Yeah.
Hugely restrictive, rapid, shutdown and aggressive pursuit of infection vectors.
All whilst mobilising the resources of a nation that is actually an entire continent.
Coupled with advantageous geography, that gives large, relatively unoccupied hinterlands between cities.
And a few other things, like draconian powers and restricted civil liberties.

Europe is overcrowded, liberal and (relatively) disorganised.

Anyway, that was round 1. There still exists the possibility of a flare up.

[Johnny Brown]

As Matt said the point is it’s not over in China. They will have to continue to identify and isolate carriers until a vaccine comes along or they’ll have another outbreak. Consensus in the UK is is too late for that sort of containment anyway.

Interesting perspective on Quillette Stu, happy to admit I knew nothing about the site and thought it looked credible.

[Doylo]

Diolch

[Stu Littlefair]

As Matt said the point is it’s not over in China. They will have to continue to identify and isolate carriers until a vaccine comes along or they’ll have another outbreak. Consensus in the UK is is too late for that sort of containment anyway.

Interesting perspective on Quillette Stu, happy to admit I knew nothing about the site and thought it looked credible.

I don’t think I’d have picked up on it, but I’ve been following a lot of epidemiologists pointing out the same flaws in other papers all week.

There’s a paper out today from two cosmologists who think they can learn something by applying a model used for magnetic fields of solids to Covid-19. At the end of pages of waffle they rediscover the Gaussian function. They should get in the sea.

[andy popp]

A bit confusing how different countries takes different stands in this matter and how much wievs of the virus differ.
Im Swedish, we do not have much regulations. More of: stay at home if your feeling ill; do not visit your elders, keep distance and wash your hands.
I was out at a normally quiet boulderingspot today and it was crowded. (Since no one wants to be crowded inside a gym.) 
Dont know whats the right way but guess we will find out.
/Christer

It is the forth time I have watched this movie.

The first was when some Italian students of mine were doing some sentiment analysis from the early days of the outbreak in Lombardy. It was fucking terrifying to learn how relaxed everyone was in Italy since we knew from Wuhan how horrible it was going to turn out.

The second was when France and Spain saw what happened in Italy and did not act swiftly.

The third was when United Kingdom and United States thought they were going to be spared somehow, for very unclear reasons. With the added worry for my father in law who lives in New York who's new wife is on one of those barbarian zero-hour contracts.

The forth is watching Sweden's unbelievably flat response to the treat, knowing full well that my parents could well be infected any day now.

The film has ended in tears the first three times. But I'm sure Sweden will be spared, because...

Sweden certainly seems to be enjoying some powerful magical thinking.

Sorry about the threat to your parents Jonas.

[mrjonathanr]

Surely Toby already has enough exposure with his alt-right freedom to say offensive things Union Pseudo free speech union ?

[teestub]

Wow Toby Young, hard pass from me then! Will be a real shame to miss out on the libertarian stuff, but I guess I’ll stick to my libtard snowflake echo chambers.

[Oldmanmatt]

A bit confusing how different countries takes different stands in this matter and how much wievs of the virus differ.
Im Swedish, we do not have much regulations. More of: stay at home if your feeling ill; do not visit your elders, keep distance and wash your hands.
I was out at a normally quiet boulderingspot today and it was crowded. (Since no one wants to be crowded inside a gym.) 
Dont know whats the right way but guess we will find out.
/Christer

It is the forth time I have watched this movie.

The first was when some Italian students of mine were doing some sentiment analysis from the early days of the outbreak in Lombardy. It was fucking terrifying to learn how relaxed everyone was in Italy since we knew from Wuhan how horrible it was going to turn out.

The second was when France and Spain saw what happened in Italy and did not act swiftly.

The third was when United Kingdom and United States thought they were going to be spared somehow, for very unclear reasons. With the added worry for my father in law who lives in New York who's new wife is on one of those barbarian zero-hour contracts.

The forth is watching Sweden's unbelievably flat response to the treat, knowing full well that my parents could well be infected any day now.

The film has ended in tears the first three times. But I'm sure Sweden will be spared, because...

Sweden certainly seems to be enjoying some powerful magical thinking.

Sorry about the threat to your parents Jonas.

It has an entirely different population geography.

England (in particular in the Isles) is almost one big bloody city.
Massive, crowded, dormitory suburb lined, corridors between huge urban centres. Cheek by jowl.

Surely, Scandinavia (except Denmark) is much better placed to resist pandemics, than Western Europe.

So, maybe it’ll work for them? Or work better than it would have done here, perhaps.
Or not. The FT tracker shows their slope to be on a par with Italy, but ~3 weeks behind.

[andy popp]

JWI, are the graphs you're publishing daily on FB publicly available anywhere?

[andy popp]

Surely, Scandinavia (except Denmark) is much better placed to resist pandemics, than Western Europe.

Many Scandinavians live in high density cities; close to a tenth of the population of Sweden lives in Stockholm.

[jwi]

JWI, are the graphs you're publishing daily on FB publicly available anywhere?

no. Can't been bothered.

[andy popp]

Fair enough.

[Stu Littlefair]

Andy, they publish similar charts in the FT daily. And there’s this site amongst others who plot that kind of data.

https://covid19dashboards.com/

[Oldmanmatt]

JWI, are the graphs you're publishing daily on FB publicly available anywhere?

no. Can't been bothered.

It’s a hard conversation. I almost turned my back on it. Nobody has wondered where our resident Italian expert has been lately for instance (I had a PM, can’t say I disagree with him).

I was genuinely hoping/wishing that the Swedish response might be correct for their situation.

https://github.com/CSSEGISandData/COVID-19

I think that’s what you’re after Andy?

[petejh]

Can anyone explain why China with a population of over 1 billion had 3000ish deaths yet 20k would be seen as a ‘result’ here?

China:
Communist dictatorship.
They can disappear journalists who publish stories that go too far against the party line.
Martial law, enforced with little compassion.
Obedient citizens.
Huge police and military per capita.
Censored media.
Large sparsely-populated areas between population centres, meaning containment more feasible.
Aggressive surveillance of population by state.
Grid-pattern (soulless) cities?? easy to monitor??

UK:
None of the above.

[andy popp]

Andy, they publish similar charts in the FT daily. And there’s this site amongst others who plot that kind of data.

https://covid19dashboards.com/

I was just wondering, I see jwi's graphs on FB.

[Johnny Brown]

The platform is a fairly recent start-up and in my opinion has made one or two questionable hires (Toby Young as UK editor, for example)

Fuck's sake! Sorry guys, will try to be more circumspect in future.

[Oldmanmatt]

Can anyone explain why China with a population of over 1 billion had 3000ish deaths yet 20k would be seen as a ‘result’ here?

China:
Communist dictatorship.
They can disappear journalists who publish stories that go too far against the party line.
Martial law, enforced with little compassion.
Obedient citizens.
Huge police and military per capita.
Censored media.
Large sparsely-populated areas between population centres, meaning containment more feasible.
Aggressive surveillance of population by state.
Grid-pattern (soulless) cities?? easy to monitor??

UK:
None of the above.

There might be limits to some of the above.
This is likely a paradigm shift moment, globally.

https://www.telegraph.co.uk/news/2020/03/28/videos-claim-show-violent-protests-china-amid-coronavirus-fears/

https://www.nytimes.com/2020/03/28/world/asia/coronavirus-china-youth.html

[jwi]

Can anyone explain why China with a population of over 1 billion had 3000ish deaths yet 20k would be seen as a ‘result’ here?

I strongly suspect that the data from China is fictional.

I plotted the total number of dead in Hubei per day (blue dots) versus logistic growth (simplest theoretical model for virus spread)



If this had been data collected by students I would probably fail them for making up data. I would at the very least interrogate them individually, and mark them down severly.