Miss Ann Thrope

Right, this argument about models is going nowhere. On the larger point, I struggle to believe that you could reasonably be arguing that social controls were not necessary given where we are. I agree that not everything that has been done everywhere can be justified but, if anything, the evidence suggests the UK acted too late and has further compromised the effectiveness of the announced controls by almost a week of mixed messaging. Daily covid deaths in the UK are approaching annual deaths from car accidents. The NHS is buckling (you can't get to see a doctor now for anything unless you are already an emergency case). What alternative do you think was available in the absence of reliable testing and tracing?

On a few points of information:
- I think you need to get updated on how the situation in Japan is developing.
- Singapore and Sweden (and to an extent Hong Kong) acted much earlier on the curve and therefore did not require such severe restrictions. They also have populations who are much more geared to comply with public health guidelines.
- I agree deaths are not being recorded everywhere systematically and consistently but that's a question of organisation and protocols that can easily be corrected (as has just happened in France). This is quite unlike infection rates (and hence fatality rates) which can only be accurately determined by mass testing which is, as we know, not available in most places.
- R0 is the measure that is most commonly used in epidemiology e.g. as mentioned here and in any number of reports on this epidemic. I don't know what Rt is and havent found it in a quick search so I will be happy to be enlightened.

scrubbedexpat141

I think you lot need to get acquainted with a different type of model. They've got names and tits and arses, depending on your preference.

FriendlyExpat

You'll enjoy reading this article - even IHMES have massively revised the death count downwards https://nymag.com/intelligencer/2020...ronavirus.html

csdf

Dragging this back to data - here's how well the US has managed to control the outbreak so far. Days on labels are doubling rates (i.e. how long it takes to double the number of cases). FYI UAE is about 5.4 days.

Maxima

"Cases" of outbreak sound scary, but in reality are just a function of testing.

If country A does 10,000 tests and finds 500 infected, it sounds so much scarier than country B which does 500 tests and finds 100 infected, it can be painted as "Country A has 5 times as many sick people as Country B"

What is more important is how many people are hospitalized, require Ventilators, and the eventual fatality rate

Millhouse

The best way to contain this is to stop testing.

csdf

OK, here are cumulative fatalities, following the 100th cumulative death. US still looking pretty special. As is the UK. I've added Ireland, which is as close as an analogue to the UK as you can get (even has the same number of ICU beds per capita). The difference between Ireland and the UK is stark.

Maxima

The US is interesting, up until 2 days back, NY counted for almost half the deaths, rest of US is a different ballgame....

If you take it as 2 separate places (NY and outside NY), you would probably see ex-NY being better than the larger European countries and UK.

California took measures early, Texas is still going along nicely; Seattle was feared as a hotspot but somehow controlled it without an official lockdown. New Orleans doing bad (probably because of Mardi Gras)

NYC is the only city in the country with 1) so many apartment blocks, 2) so much dependence on mass transit

Millhouse

i love the time and effort you have put into your spreadsheet. My only request is to format the numbers with commas every thousand so the hard of thinking like me can see the big numbers.

have you also done a log chart?

csdf

Aware that you are almost certainly taking the piss, here you go:


None of them start exactly at 100 because, obviously, none of them had exactly 100 cumulative deaths on any given day.

Millhouse

half piss taking, half serious actually. I always complain of the same to our analysts. I’m very format focused!

a log chart makes things look better than they actually are but does show it slowing... except for USA, they are winning.

scrubbedexpat141

Fingers crossed. Got told last night we're all being lined up this week. Might get canned, might be safe. Lockdown just cranked the anxiety to 14.

Millhouse

Sorry to hear that. One of our investee companies is a bit bloated (before the crisis) and they are cash poor, but we have decided to give them 60days grace before making any decisions.

Another recruiter told me he got a 25% pay cut.

scrubbedexpat141

It is what it is. Got a few complications around what it would mean depending on the scenario so can't really plan too much until I know.

My role is EMEA wide so the fallout in Europe is hurting, the MENA business is doing well in the part I service but the panic for cash is strong.

The equation is a bit like this:

('we must protect our people' + 'marketing content showing desks at home' + 'videos from group leadership') x ('can only actually think short term' + 'where is the $15bn+ of cash we made last year' + 'we're all way out of our depth') = uh oh.

Miss Ann Thrope

It's a good article thanks and confirms that the case for social controls is irrefutable. As it says, far more cogently than I could manage (my emphasis):

"Nevertheless, the gap between what experts projected a few weeks ago and what they are projecting now is absolutely astounding, and it is primarily a reflection of just how much has been done, and how quickly, to defend against and respond to the coronavirus. In many cases, we have achieved so much more than modelers even imagined possible that the range of outcomes we are now looking at did not even appear at the very low end of initial forecasts. The models weren’t “wrong,” exactly, they seemingly just underestimated how widespread, thorough, and steadily maintained social-distancing measures could be."