Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Coronavirus

This site may earn commission on affiliate links.
You know, pretty much all medical "science" is crap. Humans make lousy test subjects. The docs are clueless about statistics, and have little to no experience (i.e. even a lifetime of practice is next to nothing). Everybody concerned is constrained to do their best for the patients, not for science.

Give me any day a clinical group that has found a protocol which leads to better outcomes, no matter what their process. I'll go for what they say even if it has nothing to back it up beyond that. I am very much anticipating the advent of AI that can integrate data from all over the world over many decades. The diagnostics and treatments that will come out of that will revolutionize medicine. Current approaches are almost useless for producing good data or good analysis.

It will be the AI guys, not the medical guys, that help people most in the end. You'll no doubt enjoy being told to stuff your opinion because you're only a doctor and what would you possibly have to contribute.

So you are saying if a protocol, leads to better outcomes, you will take it? would not everyone?
While in my opinion there is too much pressure to publish, and hence a lot of studies that might be of little benefit, to bluntly state "pretty much all medical science is crap" is hyperbole at its finest. Proof you say? Well just look at life expectancy over the years. Maybe meet someone who has had an organ transplant. Maybe talk to someone who 'beat' cancer. Did not just happen. That is medical science.

Maybe check this out?
Progress in Childhood Cancer: 50 Years of Research Collaboration, A Report from the Children's Oncology Group

Here is a graph, just this one graph, just this one graph, for one disease, negates your blanket condescending statement.


nihms73614f1.jpg
 
As befits a law professor, his article is well researched and well linked, so you can make up your own mind, should you choose to.

I'm sorry, but did we read the same opinion piece?

"We are in the current situation, with deaths and economic devastation worldwide, because China handled this outbreak with its trademark mixture of dishonesty, incompetence and thuggery. Were China a more civilized nation, this outbreak would have been stopped early, and with far less harm, inside and outside of China."

How the rest of the world dealt with outbreak seems to be completely independent of how China handled the information on the virus. We can see from how the different states handled their individual community outbreaks to see that the statement was NOT true. Also, looking at japan, singapore, s. korea, sweden, netherlands, versus italy, spain, uk, and us for further evidence that the author is participating in scapegoating.

Also, the author's issue with China permitting flights out of the country back in Jan looks bad, but gets the details wrong. China had banned their own citizens from flying out of the country a full week before Trump "restricted" inbound flights from China. The travel restriction was too porous anyway, since it only applied to chinese nationals, as if the virus could only attach itself to chinese citizens?!

To me, it looks like a scapegoat article for the mishandling of the outbreak on domestic shores.
 
No, it wouldn't be the first time. But in this case, we have tons of data from other countries about how this goes (look at Singapore - they just blew a circuit breaker and are now not allowing gatherings of any size). It would be prudent to not believe all the current numbers out of China, even. Korea has ongoing low levels of community spread.

Furthermore, we can see with a pretty high degree of certainty what the timing of this outbreak will be in the US, based on the models. It looks like it has to extend through the end of May or so - there's not really any getting around that. And even after that, it's really pretty clear we can't go back to normal until a vaccine is around.

So that's why it seems super irrational to me, and I'm just wondering whether there is something I'm missing here? Is convalescent plasma going to go big and become the mainstream treatment (assuming it has no massive downside yet to be discovered, is this a treatment that can be used on everyone, even those with risk factors?)?

It seems pretty certain to me that all the infrastructure for testing and contact tracing, etc., will have to be put in place - that can definitely be done and it seems like it could be done by the end of May (if we actually start focusing on that, rather than assigning blame).

But even then, things won't exactly be normal. Large portions of the economy can get going again, of course. But things that will have to be different (until a very effective treatment becomes available that works for the majority of patients and eliminates hospitalizations and long-term health impacts) seem like they are large parts of daily life:

1) Sporting events, kids sports
2) Schools & higher education
3) Air travel, vacations
4) Amusement parks
5) Concerts
6) Etc. Anything that involves crowds.

Masks aren't just going to "fix" this. It will help, but can't see it being able to drive R0 below 1 in the context of those types of gatherings.

So, other than irrationality (which is currently my thesis), is there something I'm missing? Is there some cause for optimism that doesn't match my priors which I'm conveniently ignoring?

Can anyone here propose a reasonably probable thesis which involves a quick recovery (say, starting in May?) without significant long term economic impact? What would that look like? I see people here every now and again saying "we'll open up again in May," but I haven't seen anything well supported yet.

I am hoping that Fremont can reopen by May 1 or at least May 15. If the peak in deaths in CA is April 17 and the IHME model is correct then maybe key industries would be allowed to start back up with proper precautions. If Elon could obtain tests and masks for every worker it seems reasonable that it would be allowed.

upload_2020-4-7_13-44-44.png


IHME | COVID-19 Projections
 
Last edited:
I am hoping that Fremont can reopen by May 1 or at least May 15. If the peak in deaths in CA is April 17 and the IHME model is correct

View attachment 530093

IHME | COVID-19 Projections

then maybe key industries would be allowed to start back up with proper precautions. If Elon could obtain tests and masks for every worker it seems reasonable that it would be allowed.

Governor Newsom reiterated today they are expecting the apex to be in May. Second day in a row he’s stated that.
 
Last edited:
The more I look at this model the more obviously absurd it is.
How are they able to predict what will happen a month in the future with higher certainty than what will happen in 10 days?
To get to zero deaths by the middle of next month would require there to be zero new infections today.
Maybe because there's not much variance over longer timer horizons. If a significant portion of a population gets infected, and you have a good idea of the IFR, then it's should be fairly straightforward to estimate the total number of deaths at some point in the distant future. The number of deaths over shorter periods of time on the other hand would be relatively hard to predict.
 
  • Helpful
Reactions: Oil4AsphaultOnly
Maybe because there's not much variance over longer timer horizons. If a significant portion of a population gets infected, and you have a good idea of the IFR, then it's should be fairly straightforward to estimate the total number of deaths at some point in the distant future. The number of deaths over shorter periods of time on the other hand would be relatively hard to predict.
That would be true for total deaths if the assumption was that most of the population would be infected. I'm talking about deaths per day which should be less certain farther in the future.
Their model is predicting that an insignificant percentage of the population gets infected (about 0.5%-1%?) in California.
 
I would hope the governor would have a better view of the numbers than a doctor at a single hospital. His hospital may be running out of ventilators but they can transfer resources between hospitals. Anyway the situation in New York is very dire and Cuomo says that in every one of his press conferences. This just seems like the media trying to create controversy where there is none.
Here is what Cuomo said about the the death numbers:
"But the number of deaths over the past few days has been dropping for the first time. What is the significance of that? It's too early to tell."

P.S. It's Newsom!
I'm skeptical of the IHME model too. They say with 95% certainty there will be ZERO COVID-19 deaths in California on May 23rd...

Could be Gov. Cuomo couldn’t tell from his reports how many hospitals had patients that were in their last hours but this doctor knew better being there on the ground? Sounded to me like this doctor was just tuned into his and other hospitals better given his title and pandemic organization association—doctors talking among doctors. I honestly don’t think the media reported the doctor’s comments for any other reason than the doctor strongly feels that NY isn’t close to peak yet and people shouldn’t have false expectations or let up on their protocols despite the way numbers seem one day.

I have to say while I haven’t agreed or really much liked Newsom (thanks for correcting!) when he was mayor I have been impressed with his knowledge, contingency planning, ability to organizing supplies needed for the situation. He doesn’t belittle people and supports those working along with him without doing a dog and pony show parading them out during his press conferences. He clearly communicates the material. His press conferences are fairly brief and to the point on the subject matter and I find them more watchable than others that have been given by other officials. Not everyone is a good public speaker in command of the facts but I find Newsom is worth watching.
 
Last edited:
Could be Gov. Cuomo couldn’t tell from his reports how many hospitals had patients that were in their last hours but this doctor knew better being there on the ground? Sounded to me like this doctor was just tuned into his and other hospitals better given his title and pandemic organization association—doctors talking among doctors. I honestly don’t think the media reported the doctor’s comments for any other reason than the doctor strongly feels that NY isn’t close to peak yet and people shouldn’t have false expectations or let up on their protocols despite the way numbers seem one day.

Don’t understand your P.S. on It’s Newsome by the way. I have to say while I haven’t agreed or really much liked him when he was mayor I have been impressed with his knowledge, contingency planning, ability to organizing supplies needed for the situation. He doesn’t belittle people and supports those working along with him without doing a dog and pony show parading them out during his press conferences. His press conferences are fairly brief and to the point on the subject matter and I find them more watchable than others that have been given by other officials. Not everyone is a good public speaker in command of the facts but I find Newsome is and worth watching.
His name is spelled "Newsom" not "Newsome" or have I gone crazy? haha.
My only point was that Cuomo is definitely not saying that they are at the peak or that the current situation is in any way sustainable.
Here is the transcript from his press conference yesterday: Video, Audio, Photos & Rush Transcript: Amid Ongoing COVID-19 Pandemic, Governor Cuomo Announces NYS on Pause Functions Extended for Additional Two Weeks
 
China notified the WHO on 12/31/19. They released the genome on January 9. They shut down Wuhan on January 23rd and locked the entire 60m person province down hard within a couple days. Even if every other action China took was pure evil -- torturing doctors, hoarding supplies, denying human-to-human transmission, whatever -- those three facts alone gave us all the info we needed to prepare.

We did nothing except stop arrivals from China. In fact, we did worse than nothing - our government actively forced local labs and regional labs to stop testing, even after one found evidence of community transmission. You can't blame that on China, no matter how hard you twist the facts.
But even then, things won't exactly be normal.
The stock market is not pricing in "normal" or an insta-recovery. It's pricing in an end-of-May return to work for 90% of the economy and free FedBucks to cover the rest.
Can anyone here propose a reasonably probable thesis which involves a quick recovery (say, starting in May?) without significant long term economic impact? What would that look like?
"Long term" is 20+ years for the stock market (despite distractions like daytrading and CNBC giving silly reasons for the market moving 0.2% up or down on a given Thursday). The only new long term issue for the US economy is whether or not we'll have to permanently re-deploy millions or tens of millions of workers in travel, hospitality and entertainment. If it's millions the long term effect is minimal. Tens of millions would cause problems for ~5 years.
 
I can't see how they will be able to keep average selling prices and volumes up if the market lulls.

Agreed. I imagine they're going to keep prices up since they aren't building any cars here in the US. They do continue to build in China, so I guess that's a bright spot (assuming that China keeps things under control).

I already exited my TSLA positions at current price levels due to this uncertainty.

I have taken a small amount off the table, at today's prices, for the same reason. Hopefully won't regret it, but I feel like the upside potential is lower than it was around the turn of the year. I think long term Tesla will be fine, but this will be a rough quarter or two.

If the peak in deaths in CA is April 17 and the IHME model is correct then maybe key industries would be allowed to start back up with proper precautions.

Yeah, pretty sure that model is completely wrong. They should provide access to prior predictions. It doesn't sound like the individual states really agree with the model predictions, either (Washington State plans to be closed through the rest of the school year - why would that be if they're going to be "done" in early May). And specifically in regards to your comments, the tails on these distributions on the decay side completely disagree with the data we have from other countries. It's very tough to drive down the community transmission to the zero level that they predict. And with the levels we are dealing with currently, this will end up being a fairly significant level still.

"Long term" is 20+ years for the stock market

I guess I should have used the term "medium-term" here - I am looking at the 6-9 month horizon here. Things will be fine long-term, of course. The prices will definitely recover as we start to move through this and everything is working well again.

The only new long term issue for the US economy is whether or not we'll have to permanently re-deploy millions or tens of millions of workers in travel, hospitality and entertainment. If it's millions the long term effect is minimal. Tens of millions would cause problems for ~5 years.

I guess in terms of stock prices, in the medium term, they tend to be based on earnings, and I can't see anything other than large hits to earnings for the next couple quarters. Obviously there's a lot of gov't stimulus being injected, which helps prop things up...but is it enough when earnings season hits? Is that really priced in? You don't think the market is delusional?
 
In fact, we did worse than nothing - our government actively forced local labs and regional labs to stop testing, even after one found evidence of community transmission.
That fact could be read as an attempt at suppression but it was not. It was corruption fueled incompetence. A trump related company was sole sourced to be the test developer by the CDC and they screwed it up.
 
The more I look at this model the more obviously absurd it is.
How are they able to predict what will happen a month in the future with higher certainty than what will happen in 10 days?
To get to zero deaths by the middle of next month would require there to be zero new infections today.

All of these are valid gripes.

In addition, note that the model assume the following New Zealand Level 4 restriction are met with full compliance (so in the US we should expect somewhat worse results since we are not doing that - specifically "travel severely restricted"):

"The model includes the effects of social distancing measures implemented at the “first administrative level” (in the US this generally means the state level) and assumes continued social distancing through the end of the modeled period (August 4, 2020). We classified social distancing measures using the New Zealand Government alert system Level 4 and then assume that locations that have instituted fewer than three of these measures will enact the remaining measures within seven days. We also assume that implementation and adherence to these measures is complete. With each model update, the assumption of full implementation of social distancing measures is reset; any delay will be reflected in the number of deaths and burden on hospital systems that the model estimates."
 
As befits a law professor, his article is well researched and well linked, so you can make up your own mind, should you choose to.
LOL. Where is his well researched article on what to do with US for killing a million people in Iraq for no reason ?

However "well researched", xenophobia & racism - look exactly what they look like.

BTW, you know who really like calling this a "Chinese" virus and advocate for banning Chinese goods - right wing folks in India.

ps : This is the reason I've advocated for keeping politics out of this thread.
 
I think you are right it is an AI Neural Network problem similar to FSD, it would be good if a doctor could get an opinion from the AI expert and the doctor can also apply their judgement as we know AI isn't always perfect. In any case medicine is never perfect, there are so many variables it is hard to get it right every time, and sometimes even getting it right doesn't help.

That approach would last somewhere between five minutes and a year or two. First, the AI will learn very quickly, developing judgment better than the doctor's. The best the doctor could do would be to provide additional information. Second, what doctor will be willing to second guess an AI and get sued, when just following the AI's recommendation provides protection from any second-guessing?

Certainly. That's why smallpox is still around, polio and tuberculosis are real threats, etc. Sheesh.

A few successes amongst massive failure to cure disease. Not impressed. You realize that medicine as a field has only started to try to do science in the last century or so, right? Before that it was strictly a craft, a practice. Mostly it still is. Which is why I said (by analogy) that I'll go with the guy who builds the best chairs, not the guy who has done lots of chair research and thinks he has some cool designs for better chairs.

In Forum posts do you routinely try to assault an entire field? As for the false dichotomy between doing what is best for your patients verses doing what is the best science it's obvious based on that statement alone that you don't know all that much about health care, medicine or for that matter biology. You describe Physicians as clueless, but can you be absolutely sure that's not a projection?

Oh, I'm clueless enough. But I'm also not busy putting other people's lives in danger. And yes, I'll happily assault an entire field when it is failing. It's the only moral thing to do. Meanwhile, remember Sturgeon's Law. That 90% of all medical science is crap is pretty much typical. It would do you well to remember that.

Computer models can only be as good as the data going into them. We understand some phenomenon very well and we can make excellent computer models. But some things are very difficult to model.

AI is not "computer models", which pretty much makes the rest of your comment irrelevant.

AI is only as good as who/what programs it. And in my experience, most programmers are too "binary" in their thinking to appreciated and understand the shades of grey.

I see you are not averse to providing commentary on fields you know nothing about. Perhaps you should take your own advice and refrain in order to avoid looking ridiculous. Pretty much all of our jobs are going away over the next few decades because they will be performed much better by AI. Doctors will be among the earlier ones (pathologists and radiologists and surgeons are already on their way out), in part because humans do such an awful job and other humans so desperately want a really good job to be done. So... incentive.

Here's a paper from DeepMind about AlphaZero that y'all might find interesting. It makes it clear that in order to win the game you don't try to understand the game; instead you build a superior mind and let it figure it out. This is not your father's programming. Tesla is doing the same thing for autonomous driving, which is coming soon too. They were slowed down in that they had to build a visual perception system first, but it looks like they have that now so progress should be more rapid.