Cont: 2019-nCoV / Corona virus Pt 3

Status
Not open for further replies.
Doghouse Reilly said:
I never said it was right, I was sharing that guy's opinion. I didn't claim the UK would have fewer than 8,000 deaths.
Click to expand...
That's disingenuous. You linked to John Oxford's opinion piece and quoted him saying that he didn't think the COVID-19 total death toll in the UK would exceed 8,000, an opinion that is looking increasing absurd each day.

Without offering an opinion on the article, it's fair to assume that you agreed with it, given that he was offering the same opinion to you, that this COVID-19 is no worse than seasonal flu.

Linking to an article that agrees with your point of view and then shrugging your shoulders and saying "Hey, I never said I agreed with it!", when it's shown to be very wrong, is intellectually dishonest. Don't be that person who trawls Google or social media looking for opinions and articles that support your point of view and just link to anything you find that happens to agree with you, only to disown them when they turn out to be completely wrong.

The death toll in the UK has gone from under 2,000 to over 6,000 in the week since that article was written. He's clearly very, very wrong if he thinks the death toll in the UK won't go up by another 2,000. It will in the next few days, never mind before the pandemic is over.
 
Last edited:
Let’s drop this side line about each other’s predictions and bets. Take it to PM with the member if you want to chat about it. Let’s return this thread to directly discussing the science, current figures etc. topics. Ignoring this mod box will result in suspension without appeal.
Replying to this modbox in thread will be off topic  Posted By: Darat
 
Good news from Germany: The number of people sick from the thing is sinking (at least from Saturday to the end of the data). I found that out because it angered me that certain crappy "news" websites only show the ever-accumulating infections and deaths, but never those recovered. Johns Hopkins Istitute has a (however accurate) number for that so I took their raw data sets and made a fancy plot to show it to some frightened, easily-misled people ("Genesen" = recovered):

NadgvaO.png
 
The Atheist said:
And just to keep everyone on their toes, a predictable side effect of the shutdowns: rats!

https://www.bbc.com/news/world-us-canada-52177587
Click to expand...

We skipped Passover last night, to enforce social distancing. Maybe we could have a special "coronavirus edition", with ten special plagues for the modern era.

We've got the disease itself, rats, locusts (in Africa....shipments of insecticide have been delayed or blocked, leading to locust swarms), daily Trump briefings, I'm sure we could come up with six more plagues.
 
Childlike Empress said:
Good news from Germany: The number of people sick from the thing is sinking (at least from Saturday to the end of the data). I found that out because it angered me that certain crappy "news" websites only show the ever-accumulating infections and deaths, but never those recovered. Johns Hopkins Istitute has a (however accurate) number for that so I took their raw data sets and made a fancy plot to show it to some frightened, easily-misled people ("Genesen" = recovered):

[qimg]https://i.imgur.com/NadgvaO.png[/qimg]
Click to expand...


I've noticed a lot of social engineering in what the media are and are not reporting in the U.S.. I suspect it's from the media and various levels of government self-censoring with concerns about what effects their reporting might have, rather than some shadowy authority enforcing it. But either way, I can't really say I'm against it.

Just about when I was starting to track cases in my state county by county, which information is provided consistently day by day by the state, the state governor advised individual towns not to disclose their own data. (Not every town complied.)

Why? Well, I'm in the southern part of my own county, and it turns out that half the cases in the county were happening in one single town in the northern end. It would have been easy to get a false sense of security from that. People might think, my town doesn't even have any cases, why should I stay home?

Now, it appears the number of newly confirmed cases per day might be leveling off. (I'll know more at 10pm UTC.) So the media's emphasis has shifted to the number of deaths, which of course lags well behind the number of newly confirmed cases and will continue increasing dramatically for at least another week. Again, why? Well, the deaths are arguably more newsworthy. But also, people might think a leveling off of new cases justifies getting careless at a still-crucial time. (Those who have been consistently careless from the start won't be affected either way.)

The total number of confirmed cases and the total number of deaths can only ever increase, of course. The media will use that to stretch out the "it's not over, you know" message for as long as necessary.
 
crescent said:
Okay - big ugly spreadsheet post.

History: On March 20, I started a spreadsheet. I took the total number of deaths in America as of that day (from the Worldmeter site), and multiplied times 1.2, compounding each day (a 20% daily increase). I picked 1.2 more or less out of my butt, because just looking at the numbers for the previous week 1.2 seemed to be a bit lower than the actual rate of growth. Also, it made a nice comparison to 1.1, which seemed to be about the daily rate of growth for the worldwide total deaths (but I didn't put that into a spreadsheet because I was really just playing around).

I ran out the spreadsheet for sixty days (the total for May 19 was 14,368,616 deaths). But the rate of spread was actually a fair bit above 20%, so the numbers got off pretty quickly. Here's the numbers from that first set of calculations compared to actual numbers:



date|projected|actual
date|deaths|deaths
20-Mar|255|255
21-Mar|306|301
22-Mar|367|458
23-Mar|441|555
24-Mar|528|780
25-Mar|634|1027
26-Mar|761|1295
27-Mar|913|1695
28-Mar|1096|2220
29-Mar|1315|2583
30-Mar|1578|3141
31-Mar|1894|4053

By March 31, the two columns were pretty far off. In a bad way - actual deaths by then were more than twice as high as projected. The death total was increasing at well above 20% daily.

So I restarted the column on March 31. I posted that to the forum, but I had started it with the total as of mid-day. I should have picked an end-of-day total instead. So that's what I have below (starting with end of day numbers for March 31, so it is a bit off compared to the numbers I posted last week). This is the calculation of a 20% daily increase in total deaths for the U.S., run out for three weeks of projection with the actual included up to yesterday:

Date|20% growth| Actual
date|projection| Numbers 31-Mar | 4053| 4053 1-Apr | 4864| 5102 2-Apr | 5836| 6076 3-Apr | 7003| 7121 4-Apr | 8404| 8451 5-Apr | 10,085| 9616 6-Apr | 12,102
7-Apr | 14,522
8-Apr | 17,427
9-Apr | 20,913
10-Apr | 25,095
11-Apr | 30,114
12-Apr | 36,137
13-Apr | 43,364
14-Apr | 52,037
15-Apr | 62,445
16-Apr | 74,934
17-Apr | 89,920
18-Apr | 107,904
19-Apr | 129,485
20-Apr | 155,382
21-Apr | 186,459

This is actually "better" so far - things are sticking closer to 20% growth. I mean, 20% daily growth in total deaths is horrible - but the numbers since March 31 suggests that the rate of acceleration of total deaths is slowing down. Most days, more people die than did the previous day - but not by as wide of a margin.

(The table does look very good, I could not figure out how to make a decent looking table in the forum post. I would be happy if someone with better forum-format juju could make it look prettier.)(ETA: Nevermind, I made it look better by looking at the formatting in one of Novaphile's posts. Now it looks better)
Click to expand...


Update on the numbers for the newer table:

Date|20% growth| Actual
date|projection| Numbers 31-Mar | 4053| 4053 1-Apr | 4864| 5102 2-Apr | 5836| 6076 3-Apr | 7003| 7121 4-Apr | 8404| 8451 5-Apr | 10,085| 9616 6-Apr | 12,102| 10571 7-Apr | 14,522 | 12841 8-Apr | 17,427
9-Apr | 20,913
10-Apr | 25,095
11-Apr | 30,114
12-Apr | 36,137
13-Apr | 43,364
14-Apr | 52,037
15-Apr | 62,445
16-Apr | 74,934
17-Apr | 89,920
18-Apr | 107,904
19-Apr | 129,485
20-Apr | 155,382
21-Apr | 186,459

Still increasing, but it is not accelerating as fast as it was in late March. Even yesterday's seemingly big jump in U.S. deaths was 'only' an 18% increase, well below the 25% daily increases we were seeing in March, and the actual numbers are getting further and further below what a 20% daily increase would result in, unlike in March when the actual numbers were much higher than what a 20% daily would result in.

That's edging into something like good news, as this makes the less-than-catastrophic outcomes look a bit more likely. Things are getting worse, but not a quickly as before. The U.S. daily increase is still well above the average rate of increase for the rest of world for whatever that's worth.
 
Last edited:
crescent said:
Update on the numbers for the newer table:

Date|20% growth| Actual
date|projection| Numbers 31-Mar | 4053| 4053 1-Apr | 4864| 5102 2-Apr | 5836| 6076 3-Apr | 7003| 7121 4-Apr | 8404| 8451 5-Apr | 10,085| 9616 6-Apr | 12,102| 10571 7-Apr | 14,522 | 12841 8-Apr | 17,427
9-Apr | 20,913
10-Apr | 25,095
11-Apr | 30,114
12-Apr | 36,137
13-Apr | 43,364
14-Apr | 52,037
15-Apr | 62,445
16-Apr | 74,934
17-Apr | 89,920
18-Apr | 107,904
19-Apr | 129,485
20-Apr | 155,382
21-Apr | 186,459

Still increasing, but it is not accelerating as fast as it was in late March. Even yesterday's seemingly big jump in U.S. deaths was 'only' an 18% increase, well below the 25% daily increases we were seeing in March, and the actual numbers are getting further and further below what a 20% daily increase would result in, unlike in March when the actual numbers were much higher than what a 20% daily would result in.

That's edging into something like good news, as this makes the less-than-catastrophic outcomes look a bit more likely. Things are getting worse, but not a quickly as before. The U.S. daily increase is still well above the average rate of increase for the rest of world for whatever that's worth.
Click to expand...

Since 20% is already looking too high, here are the expected numbers at 15% starting from April 5


05-Apr 1.00 09616
06-Apr 1.15 11058
07-Apr 1.15 12717
08-Apr 1.15 14624
09-Apr 1.15 16818
10-Apr 1.15 19341
11-Apr 1.15 22242
12-Apr 1.15 25578
13-Apr 1.15 29415
14-Apr 1.15 33827
15-Apr 1.15 38902
16-Apr 1.15 44737
17-Apr 1.15 51448
18-Apr 1.15 59165
19-Apr 1.15 68039
20-Apr 1.15 78245
21-Apr 1.15 89982

Here is another version where the increase continues to slow. For simplicity sake I just added a couple inflection points, but an actual curve would be better.

05-Apr 1.15 09616
06-Apr 1.15 11058
07-Apr 1.15 12717
08-Apr 1.15 14624
09-Apr 1.1 16818
10-Apr 1.1 18500
11-Apr 1.1 20350
12-Apr 1.1 22385
13-Apr 1.1 24623
14-Apr 1.1 27086
15-Apr 1.05 29794
16-Apr 1.05 31284
17-Apr 1.05 32848
18-Apr 1.05 34491
19-Apr 1.05 36215
20-Apr 1.05 38026
21-Apr 1.05 39928
 
Some good trend changes!

IHME''s updated projections show a substantial drop in deaths from 82k a few days back to 60k with their 95% range for 31k to 127k.

Caveat: These projections are only for the first wave (The Hammer) and assume full social distance control through the end of May. The Dance part from loosening controls selectively will add to these numbers. How much will depend on the tradeoffs made between the economy and health.

Also, they have now extended their model to EU countries which provides nice visibility into what's happening there.

https://covid19.healthdata.org/united-states-of-america
 
Researchers at Harvard University have proposed that long-term exposure to high levels of particulate pollution is a significant risk factor for fatal cases of covid-19. This seems intuitive, but I'm a little surprised that they were able to get results so quickly if they were normalizing for population density, poverty rates, smoking rates, etc.

https://www.hsph.harvard.edu/news/h...tion-linked-with-higher-covid-19-death-rates/
 
Last edited:
TellyKNeasuss said:
Researchers at Harvard University have proposed that long-term exposure to high levels of particulate pollution is a significant risk factor for fatal cases of covid-19. This seems intuitive, but I'm a little surprised that they were able to get results so quickly if they were normalizing for population density, poverty rates, smoking rates, etc.

https://www.hsph.harvard.edu/news/h...tion-linked-with-higher-covid-19-death-rates/
Click to expand...


Well, maybe the looked at a map of air pollution (NO2) in Europe in February 2020 and found Lombardy...

R7ZISyt.png
 
sir drinks-a-lot said:
Looking forward to higher death numbers. Keep it classy.
Click to expand...
Do you believe that epidemiologists predicting x deaths in y country look forward to it? :rolleyes:

EHocking said:
Whether China has learned its lesson regarding covering up infection numbers after today and report any resurgence of infections is yet to be seen.
Click to expand...
Unlikely. Messengers are always shot in totalitarian countries. And every government regardless of system will have monstrous incentive to downplay numbers in every way possible. Only question is which countries lie more.
 
crescent's and lomiller' tables merged

date|20% daily|15% daily|decreasing|actual
31-Mar| 1.2 4053| | | 4053
1-Apr| 1.2 4863 | | | 5102
2-Apr| 1.2 5836 | | | 6076
3-Apr| 1.2.... 7003 | | |7121
4-Apr| 1.2.... 8404| | |8451
RESET
5-Apr| 1.2.... 9616| 1.15.... 9616| 1.15.... 9616|9616
6-Apr| 1.2.... 11539| 1.15.... 11058| 1.15.... 11058|10871
7-Apr| 1.2.... 13847| 1.15.... 12717| 1.15.... 12717|12841
8-Apr| 1.2.... 16616| 1.15.... 14624| 1.15.... 14624|
9-Apr| 1.2.... 19939| 1.15.... 16818| 1.15.... 16818|
10-Apr| 1.2.... 23927| 1.15.... 19341| 1.1..... 18500|
11-Apr| 1.2.... 28713| 1.15.... 22242| 1.1..... 20350|
12-Apr| 1.2.... 34456| 1.15.... 25578| 1.1..... 22385|
13-Apr| 1.2.... 41347| 1.15.... 29415| 1.1..... 24623|
14-Apr| 1.2.... 49616| 1.15.... 33827| 1.1..... 27086|
15-Apr| 1.2.... 59639| 1.15.... 38902| 1.05.... 29794|
16-Apr| 1.2.... 71447| 1.15.... 44737| 1.05.... 31284|
17-Apr| 1.2.... 85737| 1.15.... 51448| 1.05.... 32848|
18-Apr| 1.2.... 102884| 1.15.... 59165| 1.05.... 34491|
19-Apr| 1.2.... 123462| 1.15.... 68039| 1.05.... 36215|
20-Apr| 1.2.... 148153| 1.15.... 78245| 1.05.... 38026|
21-Apr| 1.2.... 177785| 1.15.... 89982| 1.05.... 39928|

I don't see these as actually having any predictive value, but I find them useful as frames of reference. What happens if we keep up the growth rate we saw earlier? What happens if it drops a bit. Drops a lot?

None of that tells us which table columns are more or less likely, but coupled with news of actions being taken, we get an idea of how different courses of action can lead to very different death tolls, even if the change is not immediately apparent.
 
Last edited:
Belz... said:
Why do you say that? I haven't seen any numbers to that effect.
Click to expand...

The growth in Russia is just starting to hit top gear, and Vlad didn't take measures early enough to stop a fairly severe outbreak.

Keep watching.

Childlike Empress said:
Good news from Germany
Click to expand...

Yep. They're definitely over the hump - mind you, that was obvious when they started taking other patients and lending equipment.

What - if anything - is the government saying about what happens next?

I'd say you're still several weeks away from any kind of return to normality but I haven't seen many governments saying how they're going to come out.

Meadmaker said:
We skipped Passover last night, to enforce social distancing. Maybe we could have a special "coronavirus edition", with ten special plagues for the modern era.
Click to expand...

Do they have to be plagues? The western Pacific islands have just been wasted by a Cat 5 cyclone that would have been big news in normal times.

TellyKNeasuss said:
Researchers at Harvard University have proposed that long-term exposure to high levels of particulate pollution is a significant risk factor for fatal cases of covid-19. This seems intuitive, but I'm a little surprised that they were able to get results so quickly if they were normalizing for population density, poverty rates, smoking rates, etc.
Click to expand...

Very good link, thanks.

I like this bit:

“The study results underscore the importance of continuing to enforce existing air pollution regulations to protect human health both during and after the COVID-19 crisis.”
Click to expand...

That's something that needs to be heavily emphasised once the disease starts to die down. It might well encourage people to tele-commute rather than sit and breathe exhaust fumes.
 
JesseCuster said:
That's disingenuous. You linked to John Oxford's opinion piece and quoted him saying that he didn't think the COVID-19 total death toll in the UK would exceed 8,000, an opinion that is looking increasing absurd each day.

Without offering an opinion on the article, it's fair to assume that you agreed with it, given that he was offering the same opinion to you, that this COVID-19 is no worse than seasonal flu.

Linking to an article that agrees with your point of view and then shrugging your shoulders and saying "Hey, I never said I agreed with it!", when it's shown to be very wrong, is intellectually dishonest. Don't be that person who trawls Google or social media looking for opinions and articles that support your point of view and just link to anything you find that happens to agree with you, only to disown them when they turn out to be completely wrong.

The death toll in the UK has gone from under 2,000 to over 6,000 in the week since that article was written. He's clearly very, very wrong if he thinks the death toll in the UK won't go up by another 2,000. It will in the next few days, never mind before the pandemic is over.
Click to expand...

It's not disingenuous at all, as I agree with the general sentiments expressed in the article. That doesn't mean I agree with/endorse every exact figure he put forth. I agree with him on the general basis that the deaths in the UK will be far, far below those predicted by the Imperial College.

I think it IS disingenuous to suppose that just because someone posts an article with opposing views that that means the poster agrees with every last detail therein.
 
The Atheist said:
Yep. They're definitely over the hump - mind you, that was obvious when they started taking other patients and lending equipment.

What - if anything - is the government saying about what happens next?

I'd say you're still several weeks away from any kind of return to normality but I haven't seen many governments saying how they're going to come out.
Click to expand...


There likely won't be any word from Merkel before early next week when she will meet with the federal governments. These measures are almost completely in the hands of the latter, including education. Since when they were closed in mid-March the plan is to open the schools again on April 20, where in many states the regular Easter school break would have ended as well.

Merkel wants a unified line but it could well be that different federal states choose different approaches to get out of that. Which could be helpful to collect data on how it goes from there.

It's been like that also during the last weeks. I'm in NRW under relatively light quarantine while f.e. the folks in Bavaria or Thuringa are slightly more restricted.
 
The Atheist said:
That's something that needs to be heavily emphasised once the disease starts to die down. It might well encourage people to tele-commute rather than sit and breathe exhaust fumes.
Click to expand...
Already talk of converting commercial to apartments in New Zealand.

Ann Gibson in podcast I can't find from Herald. This occured to me weeks ago.
 
Lennart Hyland said:
Currently my brother wifes grand father is being treated for Covid19, he is in critical condition. He is 80 years old and suffering from a quite bad diabetes.
Click to expand...

The sad thing about it is that when it started to spread in Sweden he called his children and begged them not to come and visit in order to not get infected. Then the care service did their regular check up and spread the virus to him.

Now some week later he died, alone with no family next to his side. Like so many other people infected with COVID-19.

It strange because I spend some time staring at these graphs over the spread but this turned into something else.
 
TellyKNeasuss said:
Researchers at Harvard University have proposed that long-term exposure to high levels of particulate pollution is a significant risk factor for fatal cases of covid-19. This seems intuitive, but I'm a little surprised that they were able to get results so quickly if they were normalizing for population density, poverty rates, smoking rates, etc.

https://www.hsph.harvard.edu/news/h...tion-linked-with-higher-covid-19-death-rates/
Click to expand...

The CNN article on this had a little more information. From what I could gather via a hasty reading, the study pertained to particulates under 2.5 micrometers. As I understand it, they looked at the average concentration in counties from 2000 to 2016, and found that above an average concentration of 13 micrograms/cubic meter, every additional 1 microgram/cubic meter was correlated with a 15 percent increase in the per capita fatalities in that county.

https://www.cnn.com/2020/04/07/health/covid-19-air-pollution-risks-wellness/index.html
 
Samson said:
Already talk of converting commercial to apartments in New Zealand.
Click to expand...

Christ, if you can show me a way to short that plan I'd put it in the "Even better than TVIX" category.

It really shouldn't surprise me, but people are just not thinking this through.

The NZ housing crisis is over; dead, buried & cremated, and we didn't even need to kill all the old ladies to do it.

1 AirBNB will bring thousands of properties onto the market. No tourists = no need for AirBNB.
2 Tens of thousands of twenty-somethings will move back in with their parents, because they have no job.
3 Tens of thousands of student visa holders will leave the country as soon as they can, because they absolutely will not find 20 hours work a week in the local cafe any more. They don't qualify for the dole and will have zero income.
4 There will be a massive number of properties looking for a fast or forced sale at cheap prices.

Current apartments will be a financial millstone and thinking you'll do conversion of commercial because the commercial property market is down 60% is stupid as hell.

My guess is, banks will see through it and not lend, so there won't be any. Bank economists are usually pretty pragmatic types and while money's cheap, it still needs to be paid back.
 
Lennart Hyland said:
Now some week later he died, alone with no family next to his side. Like so many other people infected with COVID-19.

It strange because I spend some time staring at these graphs over the spread but this turned into something else.
Click to expand...

I said well back that every person on the planet will be touched by this disease.

Sorry for your loss. Making it personal is what viruses do.
 
crescent said:
crescent's and lomiller' tables merged
I don't see these as actually having any predictive value, but I find them useful as frames of reference. What happens if we keep up the growth rate we saw earlier? What happens if it drops a bit. Drops a lot?

None of that tells us which table columns are more or less likely, but coupled with news of actions being taken, we get an idea of how different courses of action can lead to very different death tolls, even if the change is not immediately apparent.
Click to expand...

IK grabbed USA historical data from the Johns Hopkins GitHub depo and applied my 7 day straddle to it producing this:

Date|Deaths/day|Ratio|R
20-Mar|49|
21-Mar|60||1.28|4.4
22-Mar|109|1.33|5.63
23-Mar|125|1.25|3.84
24-Mar|154|1.32|5.36
25-Mar|236|1.29|4.63
26-Mar|267|1.19|2.9
27-Mar|372|1.19|2.83
28-Mar|445|1.31|4.95
29-Mar|441|1.2|3.06
30-Mar|511|1.24|3.57
31-Mar|895|1.17|2.53
1-Apr|884|1.16|2.42
2-Apr|1169|1.12|1.98
3-Apr|1161|1.09|1.69
4-Apr|1320|1.13|2.12

This shows a trend toward lower daily increases and associated R's.


https://covid19.healthdata.org/united-states-of-america
I also took data from IHME's (above) model for future projections for the latter part of April. Given that deaths tend to have a sharper up ramp than down ramp after infection and different regions will have different curves, these numbers look pretty reasonable.


Date|Deaths|Deaths/day|Ratio|R
7-Apr|12704|||
8-Apr|14607|1903||
9-Apr|16644|2037|1.07|1.5
10-Apr|18748|2104|1.03|1.21
11-Apr|20899|2151|1.02|1.14
12-Apr|23112|2213|1.03|1.19
13-Apr|25248|2136|0.97|0.81
14-Apr|27404|2156|1.01|1.06
15-Apr|29506|2102|0.97|0.86
16-Apr|31589|2083|0.99|0.95
17-Apr|33591|2002|0.96|0.79
18-Apr|35553|1962|0.98|0.89
19-Apr|37411|1858|0.95|0.72
20-Apr|39200|1789|0.96|0.8
21-Apr|40899|1699|0.95|0.73
22-Apr|42514|1615|0.95|0.74
23-Apr|44027|1513|0.94|0.68
24-Apr|45445|1418|0.94|0.68
25-Apr|46774|1329|0.94|0.68
26-Apr|47997|1223|0.92|0.61
27-Apr|49107|1110|0.91|0.56

Deaths continue to decline to approx. 60k with 95% range of 31k to 127k. This seems about right given that most parts of the USA implemented significant social distancing/shutdowns on March 17-22 and there is about a 3 week delay between infection and those that die.
 
Last edited:
Doghouse Reilly said:
It's not disingenuous at all, as I agree with the general sentiments expressed in the article. That doesn't mean I agree with/endorse every exact figure he put forth. I agree with him on the general basis that the deaths in the UK will be far, far below those predicted by the Imperial College.

I think it IS disingenuous to suppose that just because someone posts an article with opposing views that that means the poster agrees with every last detail therein.
Click to expand...
The only part of the article I took issue with was the exact part of the article you quoted. The death toll in the UK is currently over 7,000 and in 1 or 2 days will overtake the 8,000 figure he says he feels it won't pass. He's simply way, way off here. This article is just one of many opinions I've come across online determined to equate the COVID-19 pandemic to something like seasonal flu. Even if it's nowhere near as bad as the worst case scenarios being thrown around, he (and others) are grossly downplaying the seriousness of it.

I never had a go at you for agreeing with anything else in the article, never mind "every last detail". I simply corrected the part of the article you went to the trouble of quoting.
 
Last edited:
Status
Not open for further replies.

Back
Top Bottom