KCRag Forum

kboish wrote: ↑Fri Apr 17, 2020 12:51 pmThis recently conducted study says about 2-4% of people tested had antibodies for Coranovirus in Santa Clara County. While this suggests a much higher infection rate than currently known in the area, it also shows you that we are NOWHERE CLOSE to so called "herd immunity". This means if we went right back to normal, cases could easily spike and overrun health systems. Herd immunity is not a strategy.

I know this isn't relevant to the point you're making here, but the Santa Clara study is junk science. The methodology is bad and their conclusions are contradicted by the fact that the infection mortality rates that they predict are lower than the percentage of the *total population* that has died in hard-hit areas. Their study had a 1.5% positive rate (which they adjust upwards after giving more weight to underrepresented demographics), and a 1.5% positive rate could have been attained exclusively through false positives in a population with 0 true cases. (The test they used is advertised to have a .5% rate of false positives, but it could be anywhere from .1% to 1.7% based on the limited testing done to validate it.) Their study did not prove anything about the lower bound, and their upper bound is impossible save for the possibility that deaths in Santa Clara have been massively underreported. In short, their results are meaningless. And that's even without touching upon the fact that their study participants were not randomly selected.

Truman has three testing sites this week:

So no prescription needed?

grovester wrote: ↑Sat Apr 18, 2020 9:37 pm So no prescription needed?

I believe the CARES Act will cover costs: https://www.hhs.gov/provider-relief/index.html

Vermont will start loosening restrictions starting Monday for workplaces. They are doing it more cautiously than Texas. VT requires masks in public for workers statewide but apparently not mandated yet for all, just suggested. Not a requirement in TX either statewide. They need to add that.
https://www.wcax.com/content/news/Gov-S ... 21071.html

Vermont currently ranks 17th in deaths per million and TX ranks 41st.

Oil prices are crashing again today. I am reading it is possible that oil may begin trading at a negative price. This essentially means that an oil producer will pay you to take their oil because the producer has no where to store. If someone doesn't take the oil, then the oil fields may shutdown, which is incredibly costly and bears the risk that they may never start back up again.

This has soooo many implications. We have not idea the kinds of knock on effects this oil issue in particular will have, not to mention everything else going on.

kboish wrote: ↑Mon Apr 20, 2020 12:40 pm Oil prices are crashing again today. I am reading it is possible that oil may begin trading at a negative price. This essentially means that an oil producer will pay you to take their oil because the producer has no where to store. If someone doesn't take the oil, then the oil fields may shutdown, which is incredibly costly and bears the risk that they may never start back up again.

This has soooo many implications. We have not idea the kinds of knock on effects this oil issue in particular will have, not to mention everything else going on.

Some May oil-contracts now trading at sub-zero prices

Some Canadian Oil is at -4.68 a barrel right now

According to ABC News:

More than 41,575 known COVID-related deaths.

Hospitalizations and death rates in New York are going down.

New York City has cancelled all parades and public events through the end of June.

California not seeing downward trend yet. Large-scale antibody study shows actual number of COVID infections could be 55 times higher than already known.

More than 700 workers at a Smithfield meat plant in S. Dakota have tested positive. Meat plants have been shut down in at least seven states. Family members of Smithfield employees are getting infected through secondary infections. South Dakota is one of the states where the governor didn't order stay-at-home.

Growing data that COVID complications also affect heart and might cause blood clots. Might cause damage to heart muscle.

In Denmark, some students are back in school. Businesses like hair salons are back open. It was reported that Denmark is able to do this because they imposed severe measures quickly, and their citizenry followed the guidelines.

If your expecting a v shaped automatic recovery (consumer orgy debt bubble) you may want to dial back your expectations a few notches. Maybe thinking longer about what we are doing from here on out is the way forward?

Steve Rattner said on MSNBC this morning that it's going to take a long time for the economy to recover.

https://www.youtube.com/watch?v=Rro1W55v4dA

https://www.taxpolicycenter.org/taxvox ... job-market

Yeah a long time seems more likely. Consumer confidence will highly likely be low for a while so the downward trends shown below are more likely. And the trends shown are more optimistic than the downward trend from the 2008 recession. But which of the 3 scenario starting points we'll be leaning toward by end of year is a mystery. Rattner apparently doesn't think Scenario 1 will happen but that is what Congress, Fed and White House are shooting for.

Another 4M+ jobless claims expected to be reported this week for a total of 26M+ in 5 weeks. And it's not just hospitality/travel/entertainment industries, it's hitting more industries every week that goes by. The longer the lockdown, the deeper and broader the hole.

According to ABC News:

US COVID known death toll: 43,921.

418 died in New York in the last 24 hours.

Massachusetts death toll: More than 1,500. Contact tracers will fan out around the state to track COVID cases.

Illinois expects to hit peak in mid-May.

Senate passes $484 Billion stimulus package for small business, as well as hospitals and testing.

Countries lead by women are showing the best results in combating the pandemic.

The head of CDC, Robert Redfield, believes a second wave in the fall could be much worse than the first wave as COVID and flu cases hit at the same time.

Today the US hit a new daily high in reported deaths, per Worldometer's recordkeeping.

However, if you look under the hood, things are better than they appear.

Some people apparently don't work weekends, and the reported tallies on Sunday and Monday are always depressed, followed by a spike on Tuesdays. This week the Saturday totals were also a bit depressed.

Digging into the state-level data helps figure out where these reporting lags are coming from, and Michigan is a culprit of the current three-day lag:





Today they reported 232 deaths, an alarming figure which exceeds their previous peak by a considerable margin. However, the average over the past four days is only ~120, which is lower than each of the days preceding this four day period. If you mentally adjust the graph to smooth out the Sunday-Tuesday blips, you can see a pretty smooth curve which is now trending downwards nicely (and which meets up with the projection).

We should continue to see downward movement from the states that have been generating most of the deaths. The thing to watch right now is whether the numbers also start going down in the states where people maybe didn't have as much reason to take the lockdown seriously. If they don't hit a peak by now then there's probably not any peak coming for them, and that's going to open a whole new can of worms. This model I've been looking at (link) is medium confident that most states (individually) are past the peak, but it might be a little overconfident due to the weekend effect. We'll see. (btw the actual trend line that the model predicts is not reliable because it's only capable of making symmetrical curves, so the only real value is to see whether it's pointing up or down -- the back side of the curve will almost certainly have a long tail)

edit: apparently some of the recent totals reflect old deaths being reclassified as covid-19 cases, but I'm not sure on all the details or how Worldometer is sorting this

FangKC wrote: ↑Tue Apr 21, 2020 6:28 pm

Illinois expects to hit peak in mid-May.

Illinois also said peak would happen mid April. https://wrex.com/2020/04/04/pritzker-ex ... ate-april/

These epidemiologist's are like meteorologist's they are experts in their field of a particular science but they can still get the modeling/projections wrong like weathermen.

They aren't even doctors actually, no PhD is required, all u need is a degree in Public Health (which many think is a useless major more along the lines of social sciences), with maybe some stats course and u can be an epidemiologist. The ones working for major governments probably have a MPH or DrPH.

And the governor of Illinois might change the expected peak date to even later DEPENDING on many factors, including how residents of his state behave. We are just learning now that there is a fairly sizeable percentage of people who have infection of COVID, who show no symptoms; and who are likely have been going around spreading it to others. Perhaps one of the reasons he extended the peak date is because he didn't anticipate having such difficulty getting materials to do adequate testing in his state, and a federal administration that has openly stated it isn't their responsibility to hasten this process.

^The questionable study that claims that the 'real' infection rate could be 50X+ higher has another study that indicates the same. Even if overstated, it wouldn't be surprising if 20X higher in hotspot regions like NYC, Chicago, Detroit, etc. Every few weeks that go by the factor may increase at least in hotspot areas. As long as there is no vaccine, which some think won't even come next year, some think we ultimately indeed may end up herd immunity. Though even that may not be possible if re-infections can happen, which hasn't been concluded either way yet.

Need that anti-bodies test released broadly.

brewcrew1000 wrote: ↑Tue Apr 21, 2020 8:20 pm

FangKC wrote: ↑Tue Apr 21, 2020 6:28 pm

Illinois expects to hit peak in mid-May.

Illinois also said peak would happen mid April. https://wrex.com/2020/04/04/pritzker-ex ... ate-april/

These epidemiologist's are like meteorologist's they are experts in their field of a particular science but they can still get the modeling/projections wrong like weathermen.

They aren't even doctors actually, no PhD is required, all u need is a degree in Public Health (which many think is a useless major more along the lines of social sciences), with maybe some stats course and u can be an epidemiologist. The ones working for major governments probably have a MPH or DrPH.

The peak in daily deaths will occur roughly 3.5 weeks after R drops below 1, and the purpose of the stay-at-home orders was to get R below 1.

The only reason for this type of projection to be wrong is if people don't adequately obey the rules.

Models really only have two inputs: What the virus will do and what the people will do. The virus is predictable. People are not. The purpose of models is to figure out what you need to get people to do in order to get the desired results. Given how little testing is available and how long it takes for patients to end up in hospitals, there's really no way to measure the effects of behavior changes until ~2 weeks after they take place, so under some circumstances the models won't be useful for making realistic forecasts.

Looking at what the governor of Illinois actually said, it doesn't make much sense. I think he may have just been misusing the terminology. He said that the stay-at-home order was successful which pushed the peak out a month later... which is bizarre. What he said about the numbers (decreasing rate of hospital admissions/stabilization of total hospitalized patients) is indicative of peaking, and the daily death toll also shows at minimum a plateau if not the start of the decline.