The World-O-Meter Thread

[Derec]

Over 70,000 daily case count now for the US. Any thoughts on what this will peak at?
Sent from my iPhone using Tapatalk

I do not know what it will peak at - best guess under 100k and about the beginning of August.
If you look at daily cases, they are increasing linearly, not exponentially. That suggest an R that is close to 1. But as more people get infected, they will develop immunity (at least for a while, but memory T and B cells mean likely ongoing immunity in absence of serum antibodies) and thus R will go lower. Since R is close to 1 already, due to a combination of social distancing, mask and other measures and the acquired immunity, it will not take long for it to go below 1.
I think this immunity, let's call it partial herd immunity, is a big reason why states like NY and NJ are enjoying low daily numbers now. Now you may object that the percentage of people who have been infected is not nearly high enough to ensure full herd immunity, and you are right.
However, there are two things that help here
- social distancing, etc.: people are not behaving like we did in February, there is still social distancing going on, reducing R₀ which is reproductive rate of a "virgin" population. R itself gets lower as more people get infected the number of susceptible people decreases. What is needed is an sustained R<1 for the outbreak to die down, and R is calculated by multiplying R₀ by the fraction of population that is susceptible.
- Related to the above, but a bit more in depth: not everybody has the same risk of exposure. People working essential jobs involving contact with the public as well as those who are willing to take more risks going out and being around people for fun (as opposed to for work) are going to constitute a higher fraction of infections, both past and future, than people who avoid possible exposure as much as possible. The raw fraction of people who are susceptible then is not good enough - effective fraction of those susceptible is a better model, where a weighed average based on risk is taken. Therefore the effective fraction susceptible is lower than the raw number.

To reiterate, I think the daily number of cases will probably start dropping around the 1st of August because R is already very close to 1 (no exponential growth evident) and as more people get infected, the fraction of those who are susceptible drops as well, along with R.

 

[Elixir]

To reiterate, I think the daily number of cases will probably start dropping around the 1st of August because R is already very close to 1 (no exponential growth evident)

That's a dumb thing to reiterate. At an "R" of 1 daily case numbers would be exactly flat. You seem to think that "exponential increase" means a continuous squaring (at least) of case numbers. In truth, even an exponent of 1.001 will produce "exponential growth".
This looks VERY much like exponential growth:

 

[Derec]

That's a dumb thing to reiterate.

And yet, it had to be done for the sake of some readers.

At an "R" of 1 daily case numbers would be exactly flat.

In theory, yes.

You seem to think that "exponential increase" means a continuous squaring (at least) of case numbers.

No. Exponential increase means it follows an exponential function. It has nothing to do with squaring - that would be a power function, which is a very different animal.
What it has to do is that equal factors (such as doubling, but any constant factor works) would occur in equal increments of the independent variable - in this case time. For example, a doubling of cases every 5 days is an example of exponential growth. Half a radioactive material decaying every 30 years is an example of exponential decay.

In truth, even an exponent of 1.001 will produce "exponential growth".

You surely mean base, not exponent.
And again, in theory. But in the real world the base (or if you prefer the growth rate with natural base e) is not static, but variable, and unless people start being more reckless it will decrease due to fewer susceptible people.
You can actually rewrite a linear function y=mx as an exponential function with a variable base as y=m(e^(ln(x)/x))^x where e^(ln(x)/x) is the variable base. Or expressed with base e as y=me^rx, linear function would have a growth rate of r=ln(x)/x.

If the R is close enough to 1 and decreasing it can look more like a quadratic (total cases) or linear (daily cases) function, as is the case here.

This looks VERY much like exponential growth:

View attachment 28612

It does not. Daily cases look to be increasing linearly, not exponentially.
One important property of an exponential function is that its derivative is also an exponential function. Thus daily cases should look exponential, with increases by a factor in equal time intervals. We do not see here. The graph since mid-June looks like a straight line would be a very good fit. If anything it seems to go slightly concave down, and certainly not not concave up.

 

[Elixir]

In theory, yes.

And in practice.
In truth (vs practice) it's an unknown - we can't even be certain of the actual rate of spread, number of deaths or number of cases using only the diagnosed or reported cases (thanks to Trump's ban on testing and contact tracing). I don't see many epidemiologists going down happy lane with Trump, whistling "under 100k/day and about the beginning of August". In fact there seems to be a consensus that a peak by early August would be part "B" of wave 1.

It will be interesting to see if Republicans' calculation (which afaics is "if we can infect enough people by September, there won't be many new cases by election day, MAGA") pans out, or if they're as wrong as they have been about every other aspect of the Trump Virus debacle.

 

[bilby]

Over 70,000 daily case count now for the US. Any thoughts on what this will peak at?
Sent from my iPhone using Tapatalk

I do not know what it will peak at - best guess under 100k and about the beginning of August.
If you look at daily cases, they are increasing linearly, not exponentially. That suggest an R that is close to 1. But as more people get infected, they will develop immunity (at least for a while, but memory T and B cells mean likely ongoing immunity in absence of serum antibodies) and thus R will go lower. Since R is close to 1 already, due to a combination of social distancing, mask and other measures and the acquired immunity, it will not take long for it to go below 1.
I think this immunity, let's call it partial herd immunity, is a big reason why states like NY and NJ are enjoying low daily numbers now. Now you may object that the percentage of people who have been infected is not nearly high enough to ensure full herd immunity, and you are right.
However, there are two things that help here
- social distancing, etc.: people are not behaving like we did in February, there is still social distancing going on, reducing R₀ which is reproductive rate of a "virgin" population. R itself gets lower as more people get infected the number of susceptible people decreases. What is needed is an sustained R<1 for the outbreak to die down, and R is calculated by multiplying R₀ by the fraction of population that is susceptible.
- Related to the above, but a bit more in depth: not everybody has the same risk of exposure. People working essential jobs involving contact with the public as well as those who are willing to take more risks going out and being around people for fun (as opposed to for work) are going to constitute a higher fraction of infections, both past and future, than people who avoid possible exposure as much as possible. The raw fraction of people who are susceptible then is not good enough - effective fraction of those susceptible is a better model, where a weighed average based on risk is taken. Therefore the effective fraction susceptible is lower than the raw number.

To reiterate, I think the daily number of cases will probably start dropping around the 1st of August because R is already very close to 1 (no exponential growth evident) and as more people get infected, the fraction of those who are susceptible drops as well, along with R.

This sounds to me just like:

"It’s going to disappear. One day, it’s like a miracle, it will disappear"

Only with more long words.

 

[Derec]

This sounds to me just like:
"It’s going to disappear. One day, it’s like a miracle, it will disappear"

I never said anything about disappearing. But the daily cases will start decreasing soon in my opinion, and likely around the beginning of August. That just means that there will be fewer new cases each day on average, but obviously we will be decreasing from a rather tall peak of more than 70k/d, which is hardly "disappearing".
I think it will take until early to mid-October for new cases per capita in southern hot spots (e.g. Florida, Texas, Georgia, Arizona, California) to start resembling NY and NJ today. And even that is not "disappearing", but it would be very good news indeed.

Only with more long words.

 

[Derec]

And in practice.

No, since it's a rather crude model.

(thanks to Trump's ban on testing and contact tracing).

[Citation needed] that Trump instituted a ban on testing and contact tracing.

I don't see many epidemiologists going down happy lane with Trump, whistling "under 100k/day and about the beginning of August".

What do you mean by "with Trump"? Trump didn't say anything like that either.

In fact there seems to be a consensus that a peak by early August would be part "B" of wave 1.

And nobody known whether, or when, there will be a proper second wave. It is very possible that if all restrictions are lifted by October, with kids going back to school, with people going to events as if everything is back to normal, I can definitely see a significant wave springing up. But if social distancing is maintained, we may avoid it.

It will be interesting to see if Republicans' calculation (which afaics is "if we can infect enough people by September, there won't be many new cases by election day, MAGA") pans out, or if they're as wrong as they have been about every other aspect of the Trump Virus debacle.

First of all, I am not a Republican. Second, I think for example Georgia reopened way to early. I would have waited until June 1st for Phase I and things like allowing restaurants to open for indoor dining is crazy to even contemplate right now. But it is what it is.

However, we are seeing in NY and NJ that the initial severe wave led to significant slowdown in new cases due to partial herd immunity. The same will happen elsewhere, regardless of which party it may help in November.

 

[lpetrich]

EU extends summit to Sunday after deadlock over COVID recovery plan - Reuters

European Union leaders failed to agree on a massive stimulus fund to revive their coronavirus-hammered economies on Saturday after two days of fraught negotiations, but extended their summit for another day to try and overcome their differences.

As the 27 leaders scurried back to their hotels after a late, inconclusive dinner, German Chancellor Angela Merkel and French President Emmanuel Macron stayed behind in the EU’s headquarters in Brussels to haggle with the Dutch-led camp of thrifty countries demanding cuts to the 1.8-trillion-euro package.

“The negotiations were heated,” said Prime Minister Giuseppe Conte of Italy, one of the EU countries most affected by the coronavirus crisis that are seeking generous aid from the bloc. “Europe is under the blackmail of the ‘frugals’.”

“We have to do all what is possible to reach a deal tomorrow. Further delays are not useful to anybody.”

 

[Elixir]

I don't see many epidemiologists going down happy lane with Trump, whistling "under 100k/day and about the beginning of August".

What do you mean by "with Trump"? Trump didn't say anything like that either.

How disingenuous of you, pretending you have never heard any of Trump's Happy Talk. Let me refresh your memory:

“We have it totally under control. It’s one person coming in from China, and we have it under control. It’s going to be just fine.”
“Looks like by April, you know, in theory, when it gets a little warmer, it miraculously goes away.”
"“The Coronavirus is very much under control in the USA"
“It’s going to disappear. One day, it’s like a miracle, it will disappear.”
"...we’re doing a great job with it. And it will go away. Just stay calm. It will go away.”

and then there's this, fitting right in:

"it will peak at under 100k/day and about the beginning of August"

Yes, Derec. You are spouting Happy Talk, just like Trump.
The beginning of August is less than 2 weeks away and we are still posting record new cases almost every day.
I hope you're right about the case counts, and that any recorded decrease in new cases around the beginning of August isn't 100% due to the New and Improved Reporting directly to Trump's toadies instead of the CDC.
Meanwhile, as Bilby correctly points out, you sound just like the orange clown.

 

[Jimmy Higgins]

No, since it's a rather crude model.


[Citation needed] that Trump instituted a ban on testing and contact tracing.

I don't see many epidemiologists going down happy lane with Trump, whistling "under 100k/day and about the beginning of August".

What do you mean by "with Trump"? Trump didn't say anything like that either.

In fact there seems to be a consensus that a peak by early August would be part "B" of wave 1.

And nobody known whether, or when, there will be a proper second wave. It is very possible that if all restrictions are lifted by October, with kids going back to school, with people going to events as if everything is back to normal, I can definitely see a significant wave springing up. But if social distancing is maintained, we may avoid it.

It will be interesting to see if Republicans' calculation (which afaics is "if we can infect enough people by September, there won't be many new cases by election day, MAGA") pans out, or if they're as wrong as they have been about every other aspect of the Trump Virus debacle.

First of all, I am not a Republican. Second, I think for example Georgia reopened way to early. I would have waited until June 1st for Phase I and things like allowing restaurants to open for indoor dining is crazy to even contemplate right now. But it is what it is.

However, we are seeing in NY and NJ that the initial severe wave led to significant slowdown in new cases due to partial herd immunity. The same will happen elsewhere, regardless of which party it may help in November.

There is no partial herd immunity. The rates dropped because NY’ers aren’t a bunch of whining ass pussies, protesting rules that are stopping the spread. They continue to show great restraint.

That partial restraint by others in other states is keeping Georgia and Texas, etc... from exploding in new cases.

The responsible people are doing their best to save us from the dumbass selfish people.

 

[Elixir]

There is no partial herd immunity. The rates dropped because NY’ers aren’t a bunch of whining ass pussies, protesting rules that are stopping the spread. They continue to show great restraint.

That's exactly right. But it doesn't fit with Donny & Derec's Happy Talk.

That partial restraint by others in other states is keeping Georgia and Texas, etc... from exploding in new cases.

They're just skirting the edge, where they're having to call for refrigerator trucks to hold the bodies but people aren't yet dying outside the ER doors in great numbers. Yet.

The responsible people are doing their best to save us from the dumbass selfish people.

I call the latter 'Happy Talkers'.

 

[Derec]

How disingenuous of you, pretending you have never heard any of Trump's Happy Talk. Let me refresh your memory:

Oh, I definitely have, and am not pretending to have not.
But his Insane Talk has nothing to do with my mildly optimistic prediction.

“We have it totally under control. It’s one person coming in from China, and we have it under control. It’s going to be just fine.”
“Looks like by April, you know, in theory, when it gets a little warmer, it miraculously goes away.”
"“The Coronavirus is very much under control in the USA"
“It’s going to disappear. One day, it’s like a miracle, it will disappear.”
"...we’re doing a great job with it. And it will go away. Just stay calm. It will go away.”

I never said anything about disappearing. Or miracles. Or COVID being under control. In Miami they are over capacity for ICUs for fuck's sake! That's not being under control. But unlike you, I see a light at the end of the tunnel, and I don't think it's just a freight train coming my way.

and then there's this, fitting right in:

"it will peak at under 100k/day and about the beginning of August"

I made a prediction about the peak of a curve of new cases (corresponding to the inflection point of cumulative cases), which is very different than Insane Talk about miraculous disappearance.
You seem to treat any even mildly optimistic predictions with disdain. I wonder if you are just emotionally invested in "sky is falling" scenarios.

Yes, Derec. You are spouting Happy Talk, just like Trump.

Mildly Optimistic Talk. Not Miraculous Talk. Big difference.

The beginning of August is less than 2 weeks away and

The new cases 7 day average curve is still rising, but it is bending down. I.e. the rate of change of the rate of change is negative. That is the reason why I see an imminent peak if current trends continue.

we are still posting record new cases almost every day.

Obviously the new case curve is still rising. If it was going down, we'd be past the peak already. Duh!

I hope you're right about the case counts, and that any recorded decrease in new cases around the beginning of August isn't 100% due to the New and Improved Reporting directly to Trump's toadies instead of the CDC.

The states also independently publish their case data.

Meanwhile, as Bilby correctly points out, you sound just like the orange clown.

Since I never used the words "disappear" and "miraculous" I think that's BS.
Even when we reach the new case peak, it will be a long slope downward during which we will still need to wear masks, socially distance, the whole 9 yards (or 6 feet).

 

[Derec]

There is no partial herd immunity.

Oh of course there is! You have no idea what you are talking about! That's the whole basis of epidemiological models such as SIR.
Remember all the talk about "flattening the curve" that we have heard so much about at the start of the pandemic?
That curve is based on partial herd immunity, i.e. decreasing number of susceptible people as people get infected and recover. That's why you have a rise, peak and fall-off.

Think about it this way. Let's say COVID has a "natural" R₀ of 4. That means that, on average, each infected person will infect four others. Those four will infect four more and so on. Exponential growth. But eventually, so many people are infected that there is increasing likelihood that they will meet people who are also infected or have recovered and have immunity. They would still infect 4 others if it was still the beginning of the pandemic, but if that person is 25% likely to find somebody non-susceptible, then the effective R drops to 3. That's why there is talk about 70-75% needed for herd immunity - then effective R drops below one and you have exponential-like decay of new cases on the down-slope. But note, it is not a binary switch where you either have herd immunity or not. Any level of herd immunity slows down the transmission rate to a degree.
Now, flattening of the curve itself is done by reducing R₀. Let's say social distancing, masks etc. lowers R₀ from natural 4 to 2. Effective R is R₀*χ (I will use chi for fraction that are still susceptible) so you only need 50% herd immunity to bring R<1. If you succeed to bring it down to 1.5 you only need 33% - you get my drift. The only caveat is that if you relax social distancing your R₀ goes back up and if there is still too many active cases around you risk a resurgence.

What I wrote so far is regular stuff also told by people like Fauci.
Now my own extension of that is to argue that raw susceptible fraction (let's call it χ₀) should be replaced by effective susceptible fraction (χ_e if you will) because not everybody has the same risk of getting infected. And the groups at highest risk of infecting others are also the same groups most likely to have been infected in the past, i.e. they have lower χ_group.
Let's look at our example again.
So a random person is on average more likely to infect 4 people. But that number is not flat. Some people are in contact with more people, some with less. Person A who may be a grocery store worker or somebody who jumps at the chance to mingle with people for fun as soon as he is able to will interact with far more people than somebody working from home and limiting interactions.
Now, the person interacting with more people is more likely to pass the virus to others but he is also more likely to have been infected and thus become immune. I.e. χ drops fastest in those with highest R₀ and thus χ_e is lower than raw χ₀.
This is not magic, this is not Trumpian Happy Talk, it's simple logic. If you disagree, tell me why in particular instead of invoking Trump as if it was a cogent argument. Thank you.

The rates dropped because NY’ers aren’t a bunch of whining ass pussies, protesting rules that are stopping the spread. They continue to show great restraint.

It is partly that, partly that a significant fraction has already been infected.

That partial restraint by others in other states is keeping Georgia and Texas, etc... from exploding in new cases.

Sure Jan. There isn't really that much interstate travel happening right now for it to make that big a difference.
And I can't speak for Texas, but around here most people I see in grocery stores and even walking around the neighborhoods wear masks. Atlanta metro interstates used to mostly glow orange, red and deep red on Google Maps between 4-7 pm but now, still in July, rush hour is mostly green. So people drive less, which means less interactions. But we are not as far as NY, NJ etc. with regard to dropping our χ_e!

The responsible people are doing their best to save us from the dumbass selfish people.

Count me among the former. I minimize contact with other people, especially indoors! And I wear a mask.

 

[Derec]

That's exactly right. But it doesn't fit with Donny & Derec's Happy Talk.

It's exactly wrong!

I call the latter 'Happy Talkers'.

One can be reasonably optimistic about the future course of the pandemic and still be careful about exposure and spread.

 

[Elixir]

I call the latter 'Happy Talkers'.

One can be reasonably optimistic about the future course of the pandemic and still be careful about exposure and spread.

Whistling in the dark is a well known means of self delusion. Sometimes it helps preserve the sanity of the whistler; good luck with that.
Calling it "reasonable" makes it easier to convince yourself, right?

 

[crazyfingers]

I call the latter 'Happy Talkers'.

One can be reasonably optimistic about the future course of the pandemic and still be careful about exposure and spread.

Whistling in the dark is a well known means of self delusion. Sometimes it helps preserve the sanity of the whistler; good luck with that.
Calling it "reasonable" makes it easier to convince yourself, right?

Whistling in the dark is a good way to make sure that you don't startle a skunk. But doesn't help with a pandemic.

 

[lpetrich]

EU reaches historic deal on pandemic recovery after fractious summit - Reuters

European Union leaders clinched an historic deal on a massive stimulus plan for their coronavirus-throttled economies in the early hours of Tuesday, after a fractious summit lasting almost five days.

The agreement paves the way for the European Commission, the EU’s executive, to raise billions of euros on capital markets on behalf of all 27 states, an unprecedented act of solidarity in almost seven decades of European integration.

Political capital from EU recovery deal dwarfs cash prize - Reuters

The economic recovery package agreed on Tuesday after a marathon EU summit ticks most of the political boxes sketched out by the bloc’s jubilant leaders - but its financial achievements appear far less substantial.

After 90 hours of quarrelling and haggling between richer northern and poorer southern states over the deal’s scope, the big prize for countries worst-hit by the coronavirus pandemic was a package of 312 billion euros ($357 billion) in non-repayable grants.

But this amounts to only some 0.75% of European Union GDP a year over 2021-23, when the funds are meant to be allocated - at a time when its economies are shrinking at more than ten times that rate.

The actual value of disbursals over the next two-and-a-half years, when recovery needs will be greatest, is likely to be lower still.

 

[Derec]

Whistling in the dark is a well known means of self delusion.

You are the deluded one.

Sometimes it helps preserve the sanity of the whistler; good luck with that.
Calling it "reasonable" makes it easier to convince yourself, right?

All you have is gainsaying, without offering any arguments as to why my analysis is wrong.
In any case, the proof of the pudding is in the eating. We shall know soon enough who is right.

 

[ronburgundy]

Oh, I definitely have, and am not pretending to have not.
But his Insane Talk has nothing to do with my mildly optimistic prediction.


I never said anything about disappearing. Or miracles. Or COVID being under control. In Miami they are over capacity for ICUs for fuck's sake! That's not being under control. But unlike you, I see a light at the end of the tunnel, and I don't think it's just a freight train coming my way.

and then there's this, fitting right in:

"it will peak at under 100k/day and about the beginning of August"

I made a prediction about the peak of a curve of new cases (corresponding to the inflection point of cumulative cases), which is very different than Insane Talk about miraculous disappearance.
You seem to treat any even mildly optimistic predictions with disdain. I wonder if you are just emotionally invested in "sky is falling" scenarios.

Yes, Derec. You are spouting Happy Talk, just like Trump.

Mildly Optimistic Talk. Not Miraculous Talk. Big difference.

The beginning of August is less than 2 weeks away and

The new cases 7 day average curve is still rising, but it is bending down. I.e. the rate of change of the rate of change is negative. That is the reason why I see an imminent peak if current trends continue.
View attachment 28635

we are still posting record new cases almost every day.

Obviously the new case curve is still rising. If it was going down, we'd be past the peak already. Duh!

I hope you're right about the case counts, and that any recorded decrease in new cases around the beginning of August isn't 100% due to the New and Improved Reporting directly to Trump's toadies instead of the CDC.

The states also independently publish their case data.

Meanwhile, as Bilby correctly points out, you sound just like the orange clown.

Since I never used the words "disappear" and "miraculous" I think that's BS.
Even when we reach the new case peak, it will be a long slope downward during which we will still need to wear masks, socially distance, the whole 9 yards (or 6 feet).

Given Trumps overt efforts to exert authoritarian control over the reporting of cases and deaths (by eliminating the CDC as the data recipient), and his toady GOP governors efforts to help Trump suppress the numbers (such as Florida who fired the chief data scientist for being too accurate), the actual current spike is likely notably worse than than the available number and isn't likely actually "bending down".

 

[ronburgundy]

Here's some more bad news, the false negative rate is far higher than imagined, with 67% false negatives for those infected in the prior 4 days, 38% false negative for those already symptomatic within the first 5 days, and 20% for those infected 8 days ago and 3 days into symptoms.

Not only does that mean deflated numbers b/c people are getting negatives and some not getting tested again, but it means people are getting what they are taking as a "all clear" sign and going out infected while assuming they are not.

https://www.sciencedaily.com/releases/2020/06/200610094112.htm