Enevhar Aldarion
Hero
Maybe and maybe not? I look at Virginia's numbers as of today and only 10% of tests find positive cases. So either a lot of unnecessary testing or defective test kits.
D&D and the rising pandemic
Maybe and maybe not? I look at Virginia's numbers as of today and only 10% of tests find positive cases. So either a lot of unnecessary testing or defective test kits.
FrogReaver
As long as i get to be the frog
A week or 2 ago it was at 6.7% of tests finding positive cases.
However, Virginia is barely testing at all. To date they've performed 17,589 tests.
Enevhar Aldarion
Hero
Yes, but with the more focused testing being done here, I would expect a much higher percentage to be positive.
Once they can start doing the antibodies check, it will be interesting to see how many people had it, never knew it, and got better.
I still expect the US to have a death rate no higher than 1%, but that is still 300k, if 30m get sick.
Well, note there's also the heuristics around who they choose to test that will impact this. Like, say you are testing only first responders and health care providers, but among that group, you test for even an occasional cough, you'll see many tests coming back negative.
FrogReaver
As long as i get to be the frog
I'm more interested in death rate compared to confirmed cases at this time. That's going to be much higher than 1% IMO.
Overall deathrate is near meaningless when we have no idea the actual numbers of people that have it. I mean I suppose it gives us a maximum number of possible deaths (1% of 330million) = 3.3 million. However, for while we are going through it all we really know is how many test positive and how many of those test positive die.
It isn't bungling.
It is something that doubles every 3 days and has a 20 day delay between measures and results.
So when you see 2 deaths in a day, you have already locked in 250 deaths/day in 20 days, and it then starts to grow slower. If you wait 3 days after 2 deaths in a day, you have 500/day locked in. Wait a week? 1000/day.
Look, I have absolutely no excuse to not understand exponential curves, and it took me far too long to get how horrible this is. And after I did, I still missed stuff like the running out of ICU beds.
What more, the error bars involved where ridiculously huge early on. "Tiny" changes in spread rate, symptomatic rate, etc resulted in orders of magnitude (like, more than 1) difference in impact.
Getting non-epidemiologists to get how dangerous this is is hard. And the epidemiologists -- well, the naive answer remains "This is an upper repiratory infection. Pandemics of these are all the same. Everyone gets sick." - the fact that we have managed to contain and extinguish it is something new. The consensus in the UK was to let it burn, until the Imperial college paper pointed out how hot the fire would be, and that there was a plausible way to avoid the fire. Then they changed course.
But, TL;DR, don't assume Italy bungled it.
There are 4 countries on the planet who are not currently on track to duplicate what happened to Italy. If you think Italy bungled it, you will have false confidence that your government seems competent, so it won't.
Look here:
There is a main sequence. Almost everyone is on it or near it. Japan, AUS, Switzerland are marginal. SK, Singapore, Taiwan and China are off it. Iran is lying.
(This is a double-log graph of total diagnosed vs new diagnoses in the last week. At the far left, you have noise. Once you are past SK, there is a main sequence that USA is currently at the top of. Falling under this sequence a bit is modest controls, delaying disaster. SK and China dropped of it like a cliff when they contained their epidemics.)
Per day.
647k / 365 is 1770 per day.
Cancer is 1600 per day.
Today, Covid-19 was the 3rd largest killer of Americans. And unlike everything else in the top 100, it is growing exponentially. It will be hard for it not to surpass Cancer/Heart disease by Wednesday next week.
Now you could say I'm cheating. That I'm cherry picking a day. But we have 8 months left, and Covid-19 has a lot of runway.
Zardnaar
Legend
It isn't bungling.
It is something that doubles every 3 days and has a 20 day delay between measures and results.
So when you see 2 deaths in a day, you have already locked in 250 deaths/day in 20 days, and it then starts to grow slower. If you wait 3 days after 2 deaths in a day, you have 500/day locked in. Wait a week? 1000/day.
Look, I have absolutely no excuse to not understand exponential curves, and it took me far too long to get how horrible this is. And after I did, I still missed stuff like the running out of ICU beds.
What more, the error bars involved where ridiculously huge early on. "Tiny" changes in spread rate, symptomatic rate, etc resulted in orders of magnitude (like, more than 1) difference in impact.
Getting non-epidemiologists to get how dangerous this is is hard. And the epidemiologists -- well, the naive answer remains "This is an upper repiratory infection. Pandemics of these are all the same. Everyone gets sick." - the fact that we have managed to contain and extinguish it is something new. The consensus in the UK was to let it burn, until the Imperial college paper pointed out how hot the fire would be, and that there was a plausible way to avoid the fire. Then they changed course.
But, TL;DR, don't assume Italy bungled it.
There are 4 countries on the planet who are not currently on track to duplicate what happened to Italy. If you think Italy bungled it, you will have false confidence that your government seems competent, so it won't.
Look here:
There is a main sequence. Almost everyone is on it or near it. Japan, AUS, Switzerland are marginal. SK, Singapore, Taiwan and China are off it. Iran is lying.
(This is a double-log graph of total diagnosed vs new diagnoses in the last week. At the far left, you have noise. Once you are past SK, there is a main sequence that USA is currently at the top of. Falling under this sequence a bit is modest controls, delaying disaster. SK and China dropped of it like a cliff when they contained their epidemics.)
Our numbers aren't doubling every day. It's been 50-80 for the last week.
They've increased testing, 1 death, 1 in ICU they released 5.
So yes the numbers are going up but it's a lot slower.
They closed access to China Feb 2 iirc, closed the border 2 weeks ago, self isolation 10-11 days ago, lockdown 8 days ago.
2 cases of community transmission, 2 days later lockdown.
FrogReaver
As long as i get to be the frog
IMO we aren't testing when people get this on day 1. Most people have it 1-2 weeks before getting testing and then maybe live an additional 7 or so on average after getting the test results (if they are going to die). So I think what you are saying here isn't quite true of the reality of our situation.
Also, my best forecasting (in the U.S.) is coming from setting the timespan of the average confirmed case to death to about 8 days (when death occurs).
Theo R Cwithin
I cast "Baconstorm!"
Ah, gotcha. I see where you're coming from now.
It is a telling statistic; that's why I wanted to understand it. Thanks for the clarification!
