Coronovirus III

[nonobaddog]

Using numbers from May 11, 2020

In Indiana 16.8% of those who were tested came back positive
and 5.7% of those who were positive ended up dead.

In Minnesota 10.2% of those who were tested came back positive
and 5.0% of those who were positive ended up dead.

I think this is pretty obvious - NEVER GET TESTED.

 

[jedi]

99% of the country represents 1/3 of the deaths.

zero chance NY would have shut down if the red spot was in north Dakota

Math is failing me right now and I'm typically good at it.
Why show it by 1/3?
Why not 1/4, 1/8?

I get that the map is trying to show where the concentration of cases is which would be
#1 NYC area
#2 Chicago area
#3 ????
This may be cali but I suspect their infect vs their own state population lowers that number.

Would the map numbers not been better showing death of c-virus vs state population or even usa population?

 

[chipbennett]

Math is failing me right now and I'm typically good at it.
Why show it by 1/3?
Why not 1/4, 1/8?

I get that the map is trying to show where the concentration of cases is which would be
#1 NYC area
#2 Chicago area
#3 ????
This may be cali but I suspect their infect vs their own state population lowers that number.

Would the map numbers not been better showing death of c-virus vs state population or even usa population?

I see it as analogous to the murder-by-county map, that shows something like 50% of all murders taking place in 2% of US counties. As some demand with respect to gun control, a nation-wide pandemic response policy was adopted to deal with a hyper-localized problem.

ETA: another perhaps-useful graphic would be to show nursing home vs all other COVID deaths.

 

[Phase2]

this is a county map showing deaths by thirds

red, green,yellow each make up one third of deaths

I repeat, national quarantine was insane

Does anyone know the source for this map? I'm wondering why New Orleans didn't show up as at least yellow since they had a major outbreak there.

 

[jamil]

Math is failing me right now and I'm typically good at it.
Why show it by 1/3?
Why not 1/4, 1/8?

I get that the map is trying to show where the concentration of cases is which would be
#1 NYC area
#2 Chicago area
#3 ????
This may be cali but I suspect their infect vs their own state population lowers that number.

Would the map numbers not been better showing death of c-virus vs state population or even usa population?

I do think it has some value, as it does show that by far, the worst hotspots are in NY, Chicago, and Detroit, and that most of the nation does not need the same mitigations that those hotspots do. But, that image has a purpose, and it was used for that purpose here. And that is to support the idea that the problem is only in a very few places.

So why 1/3? Well, if you want to tell that story of roses, you arbitrarily pick limits that filter out the other lesser thorns. I think what would be more useful, would be a heatmap that shows the extent to which an area's infrastructure is overburdened. Just saying deaths or cases isn't all that meaningful. Of course denser areas, especially with crowded public transit, would tend to have the highest infect rate, and with the high population count, would result in big hotspots.

I haven't seen that kind of heatmap though, one that shows overburdened infrastructure. But just a straight up heatmap is more informative than the all green map, which doesn't tell the whole story. So the heatmap below still shows that the country does not have a big problem. But it does have some problem spots. It still suggests that it's not appropriate to impose the same mitigations on everyone. Obviously not all states should shut things down. But, there are a lot more areas that have problems than the previous map shows.

 

[chipbennett]

I do think it has some value, as it does show that by far, the worst hotspots are in NY, Chicago, and Detroit, and that most of the nation does not need the same mitigations that those hotspots do. But, that image has a purpose, and it was used for that purpose here. And that is to support the idea that the problem is only in a very few places.

So why 1/3? Well, if you want to tell that story of roses, you arbitrarily pick limits that filter out the other lesser thorns. I think what would be more useful, would be a heatmap that shows the extent to which an area's infrastructure is overburdened. Just saying deaths or cases isn't all that meaningful. Of course denser areas, especially with crowded public transit, would tend to have the highest infect rate, and with the high population count, would result in big hotspots.

There are also only so many meaningful slices you can get out of a set of data, and the conclusions drawn from those data, and slices of data, likely wouldn't change meaningfully. Make it quartiles, or quintiles, or even tenths. You would see a few other, isolated areas pop up in the green. Would it change the analysis? And if it were, say, tenths, the heat map would show the current hotspots even that much hotter. Again: what would change in the analysis or with the conclusions?

In either case, it shows that even state-wide policy - much less, country-wide policy - is overkill to address a hyper-localized issue.

I haven't seen that kind of heatmap though, one that shows overburdened infrastructure. But just a straight up heatmap is more informative than the all green map, which doesn't tell the whole story. So the heatmap below still shows that the country does not have a big problem. But it does have some problem spots. It still suggests that it's not appropriate to impose the same mitigations on everyone. Obviously not all states should shut things down. But, there are a lot more areas that have problems than the previous map shows.

In terms of policy decisions, # of deaths, # ICU beds, # hospitalizations, etc. are probably pretty reasonable proxies for each other, because they are directly correlated. To compare this heatmap to the # deaths heatmap above, it would be useful to know the scale being displayed. (For example, is green "normal burdening of healthcare system", or is green "minor overburdening of healthcare system"?)

 

[T.Lex]

this is a county map showing deaths by thirds

red, green,yellow each make up one third of deaths

I repeat, national quarantine was insane

keyboard test utility

I rarely ever quote pics, but this is important.

THE REASON WE ARE GREEN IS THE QUARANTINE.

Why would we have WANTED to be yellow or red?

And YOU should understand that.

Without the SIP order, we would've joined NY. Does anyone here actually want to be in the same boat as NY? Any boat that NY is in is a boat I don't want to be in.

 

[Phase2]

I rarely ever quote pics, but this is important.

THE REASON WE ARE GREEN IS THE QUARANTINE.

Why would we have WANTED to be yellow or red?

And YOU should understand that.

Without the SIP order, we would've joined NY. Does anyone here actually want to be in the same boat as NY? Any boat that NY is in is a boat I don't want to be in.

Fact Check: False.

The map is based on percentage of infections. Are you claiming that NY and other other high-infection areas haven't announced SIP orders?

NY announced stay at home 3/20. Indiana announced 3/23.

NY has/had serious problems (both in decision-making and existing conditions) with this virus, but SIP is very low on that list.

 

[chipbennett]

I rarely ever quote pics, but this is important.

THE REASON WE ARE GREEN IS THE QUARANTINE.

Why would we have WANTED to be yellow or red?

And YOU should understand that.

Without the SIP order, we would've joined NY. Does anyone here actually want to be in the same boat as NY? Any boat that NY is in is a boat I don't want to be in.

This is, at best, tautological and a self-fulfilling prophecy based on models that were never accurate. There is simply no evidence that this is/must be true. Correlation does not prove causation. There is no evidence that not forcing a nation-wide shut down would have led to more deaths.

 

[T.Lex]

Fact Check: False.

The map is based on percentage of infections. Are you claiming that NY and other other high-infection areas haven't announced SIP orders?

NY announced stay at home 3/20. Indiana announced 3/23.

Ok - and NY STILL saw increases. A certain percentage of people don't listen. Without almost all of us doing the SIP thing, those hotspots would've been cancers that spread. We don't know how much, because the SIP orders were in effect and kept the spread limited.

That's a fact, jack. (Along with all the other measures.)

But hey, if you wanted to follow in NY's footsteps, go ahead Cuomo junior.

 

[GodFearinGunTotin]

I rarely ever quote pics, but this is important.

THE REASON WE ARE GREEN IS THE QUARANTINE.

Why would we have WANTED to be yellow or red?

And YOU should understand that.

Without the SIP order, we would've joined NY. Does anyone here actually want to be in the same boat as NY? Any boat that NY is in is a boat I don't want to be in.

We don’t know what quarantining did or to what extent it did help. We didn’t have control groups to compare them to. We think we know. We also don’t know the long term price we’re going to pay for taking actions based on models we thought were accurate at the time.

Extolling this as a great accomplishment? I think it’s too early to call that.

 

[Ingomike]

I rarely ever quote pics, but this is important.

THE REASON WE ARE GREEN IS THE QUARANTINE.

Why would we have WANTED to be yellow or red?

And YOU should understand that.

Without the SIP order, we would've joined NY. Does anyone here actually want to be in the same boat as NY? Any boat that NY is in is a boat I don't want to be in.

Another reason this is wrong is that the spread is the spread, we just changed the timing a little so as not to overwhelm the healthcare system. It was always going to run its course...

 

[T.Lex]

This is, at best, tautological and a self-fulfilling prophecy based on models that were never accurate. There is simply no evidence that this is/must be true. Correlation does not prove causation. There is no evidence that not forcing a nation-wide shut down would have led to more deaths.

Math is not a magic 8 ball, but before the SIP the fact is that there was exponential growth in deaths.

After the SIP there was not.

There is no evidence - none - that the lack of SIP would've yielded the same result, or better. That's pure speculation because we actually did the SIP.

 

[T.Lex]

Another reason this is wrong is that the spread is the spread, we just changed the timing a little so as not to overwhelm the healthcare system. It was always going to run its course...

This is true in the sense that ALL of that green is more vulnerable (likely) than the yellow and red. But, we have better (relatively) treatment plans than at the beginning of February.

 

[chipbennett]

Ok - and NY STILL saw increases. A certain percentage of people don't listen. Without almost all of us doing the SIP thing, those hotspots would've been cancers that spread. We don't know how much, because the SIP orders were in effect and kept the spread limited.

That's a fact, jack. (Along with all the other measures.)

But hey, if you wanted to follow in NY's footsteps, go ahead Cuomo junior.

There are reasonable alternatives to your assertions. For example: shutting down mass public transport. Closing the NYC subway and DC metro would have a disproportionately large impact on slowing the spread of a a virus as compared to closing a park in Morristown, IN, or a mom-and-pop restaurant in Podunk, IL. Yet the NYC subway remains in operation, while the mom-and-pop restaurant is closed, permanently.

 

[T.Lex]

There are reasonable alternatives to your assertions. For example: shutting down mass public transport. Closing the NYC subway and DC metro would have a disproportionately large impact on slowing the spread of a a virus as compared to closing a park in Morristown, IN, or a mom-and-pop restaurant in Podunk, IL. Yet the NYC subway remains in operation, while the mom-and-pop restaurant is closed, permanently.

Ok, when we open up the multiverse, we can figure out how that would've gone.

Instead, we have this reality.

(Speaking of which, I'm hearing good things about remdemsevir - good job on that!)

 

[Ingomike]

Ok - and NY STILL saw increases. A certain percentage of people don't listen. Without almost all of us doing the SIP thing, those hotspots would've been cancers that spread. We don't know how much, because the SIP orders were in effect and kept the spread limited.

That's a fact, jack. (Along with all the other measures.)

But hey, if you wanted to follow in NY's footsteps, go ahead Cuomo junior.

I want to see the data of NY without nursing homes, which Cuomo forced to take contagious patients back into the general population. There are going to be major lawsuits over that...

 

[chipbennett]

Math is not a magic 8 ball, but before the SIP the fact is that there was exponential growth in deaths.

After the SIP there was not.

There is no evidence - none - that the lack of SIP would've yielded the same result, or better. That's pure speculation because we actually did the SIP.

Math is not a magic 8 ball, but what you're asserting is not now statistical analysis of variance works.

All virus infections follow a curve that includes exponential growth. All viruses naturally burn themselves out, leading to an overall bell curve of overall infection.

It has become abundantly clear that what I speculated weeks/months ago - that the viral spread started not in February, but in December (or earlier) - is true. Thus, all of the models based on Day 0 being in February were wrong. The inflection point was off by months. The change in growth/spread is consistent with a virus that started spreading much earlier than originally assumed.

Further, you are basing "exponential growth" on incomplete data, yielded from intentionally diminished testing. The rate of testing has changed. The ability to test and rate of testing for antibodies has changed. Thus, there is no true picture of rate of growth then vs now, on which to make an ANOVA claim of a statistically significant change in rate of growth before and after SIP orders.

 

[drillsgt]

Yeah... I think this falls into the "just because you can, doesn't mean you should" category.

Ha, i'd like to see him haul around a real M2.

 

[T.Lex]

Math is not a magic 8 ball, but what you're asserting is not now statistical analysis of variance works.

All virus infections follow a curve that includes exponential growth. All viruses naturally burn themselves out, leading to an overall bell curve of overall infection.

It has become abundantly clear that what I speculated weeks/months ago - that the viral spread started not in February, but in December (or earlier) - is true. Thus, all of the models based on Day 0 being in February were wrong. The inflection point was off by months. The change in growth/spread is consistent with a virus that started spreading much earlier than originally assumed.

Further, you are basing "exponential growth" on incomplete data, yielded from intentionally diminished testing. The rate of testing has changed. The ability to test and rate of testing for antibodies has changed. Thus, there is no true picture of rate of growth then vs now, on which to make an ANOVA claim of a statistically significant change in rate of growth before and after SIP orders.

Again, perfect data is an illusion in this reality.

There's no denying the decisions were based on incomplete data, that's part of the problem with a novel virus: we don't know what we don't know.

At the time the decisions were made, the actual deaths appeared to be (and likely were) growing exponentially. What that meant in terms of infections is unknown and likely unknowable. But, the actual math available was compelling.