I agree. They can all quarantine together while they work. Maybe the President can give them hamburgers for a job well done.
1 Like
Hambergers.
1 Like
The University of Washington model that has been used most by the media and Trump is in complete shambles.
The model is updated weekly and it is currently almost exactly six days behind the actual trend. They will update their model tomorrow but clearly they donât have a handle on this. Now with premature re-openings and armed fuckwits gathering everywhere we can look forward to a long, hot, deadly summer. I expect close to 100,000 dead by June.
During the month of April, the US recorded a death from COVID-19 every 44 seconds. For a month!
Remember this stat when some idiot tries to compare this thing to anything other than a nuclear bomb.
One article I read said that the University of Texasâs model wasnât predicting things very well either. IIRC it stated earlier this week that we had reached the peak deaths/day⌠2-3 days before it peaked again.
I mean, you have to imagine that itâs difficult for a computer to really comprehend human stupidity.
3 Likes
I fear the algorithm that taps into this wellspring.
SkyNet
2 Likes
How can you model something when the parameters get changed arbitrarily and arenât consistent across the entire population - and that population is mobile?
Also, models require data. We havenât come close to testing enough people to have any idea of the infection rate.
And then there are random events like the Michigan and Huntington Beach protests that will spike numbers randomly.
Honestly, thereâs no way to model this thing at all. Itâs less certain than predicting the weather (which is also a factor).
Like I said before you have to pay attention to the error bars. The models are doing what they are supposed to do (try to manage resources) and have made useful predictions, but as Limey said the data is very nonuniform and messy and honestly we donât understand the disease very well.
This is a quote from a recent Vox article on the IHME model:
One analysis of the IHME model found that its next-day death predictions for each state were outside its 95 percent confidence interval 70 percent of the time â meaning the actual death numbers fell outside the range it projected 70 percent of the time.
Hereâs a link to the article: https://www.vox.com/future-perfect/2020/5/2/21241261/coronavirus-modeling-us-deaths-ihme-pandemic
Iâm not out to trash models as they are just tools, some more useful than others and often used for the wrong task, but Iâm also not going to discard criticism of them either.
Weâre now seeing places like meat-packing plants, prisons and nursing homes - where there are hundreds or thousands of people jammed in together - being 100% tested, and coming up with infection rates of 50% to 80%. So these will trash the model in an instant.
Rather than looking a the model and trying to figure out where to put capacity, we know from reading the news that these three types of places are responsible for the worst outbreaks in the country right now - the nursing homes being the most deadly, the plants and prisons being the most volume of cases - which then spread into the small towns where the workers live.
We should be surging testing and resources into these three facilities countrywide. Fuck the model.
Hereâs How Texasâ Reopening Is Going: 1,200 New Cases
Texas reported 1,293 new coronavirus cases on Saturday, the third consecutive day that metric topped 1,000âa level not seen in three weeks. In the same three-day period, at least 115 coronavirus-related deaths were reported, according to state data.
Iâm generally uneasy about the use of facial recognition technology, but this carelessness has me pissed off enough to overcome that. Record mass gatherings, and if someone shows up at a hospital and theyâre found to have attended one of those, they go to the back of the line for equipment, etc. You want the freedom to run around like an idiot? You get to deal with the consequences, too. (Unfortunately, so do the rest of us.)
Our neighbors behind us had a party last night with a bunch of folks over, for crying out loud. Iâm sure weâll see all kinds of idiocy for Cinqo de Mayo tomorrow.
The irony of that, of course, would be that they will complain about not having the ârightâ to health care access.
1 Like
They should call their legislator then. After all, Iâve been told that healthcare is a privilege, not a right.
1 Like
On the plus side, they would finally get to see what a death panel really looks like.
1 Like
Still trying to work out why relaxing mitigation efforts is a good idea when they were barely working to start with.
Look at those new case numbers (lighter pink) spiking just as the relaxation went into effect. Those are new infections from before the relaxation. That means thereâs more people walking around with this thing than ever, and weâre letting people go to restaurants and hair salons!
In areas not struck hard by COVID, it seems pretty apparent that society is moving towards an increase in cases. The question to me is whether this increase leads to the public stepping back or saying âfuck it, letâs go to herd immunity.â
I donât blame the politicians as much as I blame the public. You hear and see stories like Austroâs, you see the idiotic demonstrations where very few masks are worn, and you see pictures from Georgia where very few people wear masks and are just walking right near each other like in the past. It wonât be long before theyâll be at bars and restaurants right next to each other for long periods. My impression is that much of the public considers swearing a mask to be a sign of weakness.
This might eventually change, but not before it gets much worse. If it does change, it would be nice to have the N 95 masks easily available, but I know thatâs not going to happen.
This is a big issue. Lots of people are saying âthey said lockdown would flatten the curve, and weâd all get over it. But cases keep going up, even in lockdown, and now weâre just broke too. Fuck itâŚâ