Bikers descend on Sturgis rally with few signs of pandemic

Then I’m confused as to why you keep bringing up those statistics to support your position. It is you saying subscriptions are an indicator of value. Liking content is an argument for value provided to the individual. Repeating their statistics on subscriptions really isn’t anything but herd mentality.

Well, as I haven’t been reading it I don’t have a lot of ammo. But this popped up immediately when I started investigating based on my skepticism the NYT is infallible. I don’t think it ran in the science section but that’s irrelevant to me as it’s the NYT and a science topic. A paper that is accurate on science in one section but not in another seems “interesting” for lack of a better word. Anyway, the Gov. of Arizona called them out for it and a quick look confirms there is no adjustment for testing rates. It even goes so far as to say testing rates are irrelevant. LOL.
https://www.google.com/amp/s/www.ny...ump-facebook-your-wednesday-briefing.amp.html

Yep they are profitable. That’s not being contended. I’ve applauded them repeatedly for that. Stocks are not “all” but a good indicator of sound business models. It hasn’t been shown to be a great business model for years. That was my point.

True.

Experts teaching young journalists anonymous sources are all one needs to run a story? Experts teaching journalists to quote other publications’s stories based on the same anonymous source as a second confirming source? Sorry, no, I don’t believe the institutions foisting they current crop of journalists on us are experts on good journalism. I could go on and on...

I think prizes awarded by peers in any field are not just silly but detrimental to professionalism. Certainly in journalism. This is interesting if you are interested in how winners of Pulitzers are really chosen.
https://www.japantimes.co.jp/opinio...sed-judging-in-award-of-journalism-pulitzers/

It’s not uncommon for prize winners (and hopefuls) to push the envelope and value shock or awe over accuracy. Just some random relevant reading.
https://archive.seattletimes.com/archive/?date=19980703&slug=2759312

This piece brings out the evidence awards are based more on writing style than accuracy or relevance, even when ignoring bias or backscratching. If you want to continue with the medical analogy, it like saying doctors get awards for bedside manner or tailoring of clothing instead of based on lives saved.
https://www.cjc-online.ca/index.php/journal/article/view/1743/1855

Interesting and relevant.
https://www.nytco.com/company/prize...932-pulitzer-prize-awarded-to-walter-duranty/

My point is awards that are essentially “you’re awesome, dude” reciprocated by “no dude, you’re awesome!” are mainly not helpful to a profession, especially journalism where speed and shock over accuracy is already a problem.

Yes, about that time they started realizing they had to move to digital content and it helped. But the real growth doesn’t start till the 2016 election cycle.

Anyway, from the bottom of this rabbit hole I’ll thank you for explaining why the Times has value to you. Like I said, we live different lives and enjoy different things, I’m glad the Times makes your life more enjoyable. That’s really the heart of capitalism—providing value to someone for a return. The last word on the subject is yours if you’re so inclined.

 

One is a preprint. The other containing the same information is peer reviewed.

 

What happened between May and now? Lol

 

When they ran that article what they said was absolutely true.



It's what happens when testing is not keeping up with increased cases.

 

Less infections... as noted

 

The amount of cases missed by PCR is not assumption. It’s factual based on well documented failure rates by day of infection. Although it isn’t talked about (I don’t like keeping people in the dark about how poorly PCR testing fits the needs we have to deal with this virus) it isn’t debatable. It just the way it is. I agree more cases may have been missed early because of fewer tests. My use of antibody tests adjusted for known limitations of the tests is my way of making all the PCR jazz irrelevant. There are too many unknowns with PCR. My method of adjusting antibody tests is superior I believe. I will continue to believe it is until someone offers good evidence it’s flawed.

Can you clarify what you mean by percent positive instead of number of cases? I don’t want to misunderstand your point.

I don’t think today we have any disagreement on NY data.

Maybe. I remember everyone telling me I was crazy for claiming NYC areas were approaching herd immunity. Now you seem to agree.

Reported how? By PCR or antibody testing? By PCR they are at .83% infected now. I claim they are over 30 times that figure. I claim they a little over twice as high as their current antibody testing shows.

That would be an incorrect assumption for two reasons. One, those are CDC numbers, not mine. I’ve been very clear about what my numbers are. I just reiterated above. Second, I don’t care about justifying herd immunity. I’ve been clear I don’t think it’s prudent. However, when there is a lot of evidence it exists in certain places I don’t see why we should deny it. Let’s say some modelers are correct and herd immunity IS around 30%. Wouldn’t that be something to be thankful for, not complain about? I don’t understand the hate for something positive.

If places I’m using for examples weren’t experiencing low rates of infection now vs. high rates earlier when mitigation was much greater, I would clearly be mistaken. But actual evidence on the ground supports me as well as the underlying science of testing and mathematics.

Please be clear. We are discussing two different methods of testing. What type of testing had been used on 1.7% of Sweden in March and how is it relevant to my calculations?

 

No it’s misleading precisely because they ignored facts like Brazil testing at rates 4 times less than Arizona.

 

I think it has been a great idea each year for as long as I can remember. Governor Noem is the only governor who has done the pandemic correctly.

 

LOL. June and July were peak infection months for Sweden. Where do you get your information? For two months between May and now, infections documented by PCR (which we know only accounts for a small percentage of actual infections) were double what they were in May. LOL

 

Lets just talk about 3 of your claims for now.
I recall talking about New Your State.
I think you may have brought up the bough in queens with the 50% rate. But that by no means applies to the whole city or state.

PCR testing. I try not to use cases because it depends on how much testing you do. So you will find I like % positive. It's like every day you are taking a sample of the population. So .83% positive cases doesn't mean anything. Had they test twice as many they would have found about twice as many cases. The fact that they test about 1% of there population every day makes that number better, but not as good as percent positive. Its like us testing 3.3 million a day and probably why they have kept the numbers low.
On a side note. Doesn't it seem a little strange that Corona could have an antibody rate of 50% in July, but herd immunity occurs at 30%.

 

It's percent increase. Not percent.

 

I'm just saying, they are valuable to me, and to a LOT of people. And are the most rewarded newspaper in the world.

Come on, that wasn't a scientific article and you know it.

Sometimes it's not just the "business model" but rather, providing good journalism. Some of these people there may be idealists.

There are bad journalists and good journalists, bad doctors and good doctors, bad farmers and good farmes.

All fascinating, and it the distribution of these awards got to be all equal among all journalistic organizations with people mutually praising each other, you'd have a point, but the thing is, the Times is the MOST rewarded paper in the world and that's gotta count for something, no? Sure, the Oscars are awards given by filmmakers and actors to filmmakers and actors... but if say, a film director or say, an actor, accumulates more awards than anybody else, chances are that the person is good at what he/she does, no?

Seems like a rather smooth curve to me, I don't see a big bump in 2016.

Fair enough. I like the Times. You don't. Let's leave it at that, and enjoy the other things you and I *do* have in common.

 

As far as you can remember for other editions of this otherwise nice rally, there wasn't a pandemic. What makes it boneheaded is not the rally itself - looks like great fun. What makes it boneheaded is doing it in the middle of a freaking pandemic. It could have been cancelled for this year only, like many other events were wisely cancelled.

Did you see the study showing 266,000 new cases likely stemming from this rally? See post #377, there's a link to the study. Read it.

Care to still think it was not a boneheaded idea?

 

I haven't had the time to follow the discussion between @ronv and @557 - probably very interesting but I've been busy. I'll see if I can take a look at some point.

 

I posted the worldometer link. For 30 days after May the average was 1000 per day. About equal to the period ending in May. Since then it’s been only a couple hundred per day.

Again... infections until the end of May was 6%.

So what is it now? Maybe double or slightly higher. 15% ? 17%?

Does that sound like 60% ?

No?

oh...

 

No ads, I think they only survive on members' donations (which is why I'm a donor).

 

They aren't making any money, though.

 

I saw it and I don't believe it. Sorry.

 

OK, now I've read the discussion between @557 and @Lesh and @ronv - all sides make good points, all sides being intelligent posters.

557's good points: herd immunity thresholds depend on susceptible populations and on a LARGE number of intervening factors, including the effective reproduction number which depends on population density, epidemiological controls in place, temperature and humidity, winds, etc. In terms of susceptible populations, we have blacks and Hispanics more susceptible than whites, we have people with different levels of vitamin D, different blood types, different previous exposure to other viruses of the coronavirus family, etc. Then, if you take all of this into account, herd immunity threshold is not just one minus the reverse of a global R naught number times 100, given that this would be only true if all humans in that area were equally susceptible, and if the infection was spreading homogeneously given all other conditions being equal. So, it's not like R naught = a fixed 5 therefore the HIT is a fixed 80% (1 minus 1/5 = 0.8 x 100 = 80). Some of the factors push the needed HIT up (like densely populated places where people are maskless) but some push it down, like populations that are naturally less susceptible. Furthermore, it's geographical. One given densely populated area will have more transmission among a more susceptible population so to stop it from occurring you'll need the virus to run its course through a lot of people, while in other areas where the virus has a lot of trouble spreading, it will die out sooner with a smaller percentage of infected people.

So, I'm with 557 here. HIT, globally speaking, for the whole population, is likely to be smaller than what the Mayo Clinic is globally quoting as 70%.

Now, I don't subscribe to 557's estimate of this many infections being missed. I have profound doubts about 9 cases out of 10 being missed (much less, 23 out of 24, the upper range of the CDC's guesstimate; preposterous, in my opinion). I explained elsewhere why. I proceed in the reverse order, trying to compare the infection-fatality ratio of the SARS-CoV-2 with that of the flu, and if the true ratio is only a third of the current confirmed-case fatality ratio, then the cohort of infected people responsible for these deaths can't be that much bigger. 24 times bigger? I definitely think not. I don't even think it's 10 times bigger. It's probably between 3 and 6 times bigger. But these are gut-feeling estimates too, although a relatively educated guess.

I'm with ronv in questioning that if the real HIT is closer to 30% then it's weird that more than 51% of people got infected in the Corona (aptly named) area of Queens. Especially because if we hear 577 then the number of infected people there must be even bigger... how did the virus get to infect so many if the HIT supposedly had been reached at 30%? Well, that's not absolute either because for that population maybe the R naught was too large (overcrowded poverty quarters, etc.).

Now, I'm not exactly with 577 in the argument that HIT has been reached because since PCRs have too many false negatives and antibodies fade, then we got a lot of people infected and this should have allowed for the HIT in certain places already. But wait a moment... maybe these people will have cell immunity... but if IgG neutralizing antibodies fade, then these people may STOP counting towards the HIT as they theoretically would still be able to be re-infected carriers, before their cell immunity, much more delayed, kicked in. So the argument for all this fading is actually traditionally, in epidemiology, an argument for the HIT being HARDER to reach or even impossible to reach, rather than saying "well, the antibodies have faded so more people were infected so we got a large percentage of people infected so we now have HIT."

Now, I actually think that a map like the one on post 435 by ronv, is only a temporary snapshot. Because, see, soon enough the maid from Corona, Queens, will infect the madam from the Upper West Side for whom she works. Especially in the same city, people move a lot, and these pockets of red versus pockets of beige will probably end up getting to be a lot more homogeneous, with time.

At one point the HIT may be reached, or not. I actually think it will be a lot easier to reach it with vaccines (which is traditionally the case). So the discussion is a bit academic.

How will we ever know what the HIT was and what the true infection-fatality ratio was? AFTER the end of the pandemic, and after serial, repeated serology surveys, testing the same people several times over the months. This being a huge enterprise, we simply may never know.

Me, I feel that rather than veering towards this whole herd immunity discussion, which is so full of uncertain estimates and intervening factors, it's best to think like this:

1) This is a dangerous virus that kills a few and maims many
2) HIT via natural infection is very variable according to location; it is ever-moving, and uncertain
3) The more people infected, HIT or not, the more dead and maimed people; that's a given
4) Vaccines are hopefully coming (despite the recent setbacks)
5) Let's all continue to be prudent and cautions, while we wait for the vaccines, not to kill unnecessarily people who would have survived (or avoided sequelae) if they had avoided this infection, and if in a few more months they got vaccinated
6) Meanwhile let's not worry too much about HIT. If we get good vaccines we'll get HIT accomplished

I hope that the above is helpful.

 

All right then. Just keep observing things, then. This virus has a way of proving the non-believers wrong.

 

Check out the graph in your link. June was the peak of new official cases. June figures are not “about equal” to May.

By antibody testing yes. But by May because of fading antibody levels you are missing up to 40% of March infections. They just don’t have antibody levels high enough to be detected.

The last numbers I’m aware of for antibody test results was 14% for Sweden. As I’ve explained numerous times now, and this is facts to the best of our knowledge today, antibodies are undetectable in 40% of infected persons around 2 months after infection. There is some evidence they return, but it’s thin evidence at this point. At any given point in time if you test a population for antibodies, you are going to miss 40% of actual infections from two months previous. If they don’t return, this compounds as time progresses.

Also, there is now peer reviewed data that twice as many infected individuals develop T cell immunity as develop antibodies. We do not know for sure how fast T cell immunity alone can deal with reintroduction to the virus. Probably not as well as if antibodies are present as well, but it is a degree of immunity so can’t be ignored.

Taking all the above FACTS into account, I’ve claimed 30% immunity level in Sweden.

60% is your strawman. I’ve never claimed 60% immunity or infection in Sweden. Never.

I’ve also demonstrated conclusively the old classic definition of herd immunity being in the 70% range is outdated and irrelevant to SARS-CoV-2. We must use models including heterogeneity metrics. I’m not the only one saying this anymore. In July I was certainly a lonely voice crying in the wilderness, but most experts are now having to admit 70% is not accurate. There are now models showing it may be as low as 30%. That’s why I claim some places are near herd immunity.

The fact places like Sweden and NYC have low infection rates now vs earlier when mitigations were more prevalent (especially in NYC) buttresses my conclusions. The herd immunity heterogeneity model studies I’ve linked to predict what we see in NYC today. You are welcome to critique them (even though we know you won’t read them) and point out errors missed in peer review.

Or, you can keep telling me I’m wrong without presenting evidence I am and I’ll stop responding. Discussion I like. Someone using FACTS and logic to point out my errors I like. Your opinions, thinly veiled ad hominem, and “straw people” arguments I’m bored with.

 

I don’t believe a set of antibody tests today on a population captures the true number of individuals who have been infected in the tested population. I’ve explained why repeatedly. Here are the bullet points.
—Because of antibodies fade, at any given time, a testing event will miss 40% of the individuals who were infected and recovered approximately 60 days previously. There is evidence antibodies may return at some point as a second wave, but this is preliminary. It’s been demonstrated repeatedly antibody disappearance at 60 days post C19 is about 40%.
—only half as many individuals who recover EVER have antibodies in comparison to the number who achieve T cell immunity.
—It’s possible a degree of immunity exists in the population before infection with SARS-CoV-2 from previous coronavirus infections. I don’t add this into my calculations because it’s not confirmed

Based on this, yes it’s possible the state of N.Y. could be around 50% infected. That’s only if we include all T cell immunity though. Based just on antibody testing I’m not sure where NY state claims to be on antibodies. If you have recent statewide antibody testing data I’d like to have it. Most recent data is NYC only.

I am still confused. You say you like % positive and then in the next sentence you say .83% positive doesn’t mean anything.

Percent positive cases is valuable to know but it can’t predict the actual real number of infections. It’s mostly useful to determine if testing volume is adequate or how fast infections are spreading.

Also percent positives is so dependent on variables like age of the population, whether testing is targeted to symptomatic individuals or randomly applied in mass testing scenarios, changes in testing volume, etc., etc. it’s impossible to apply it equitably across states or countries to draw conclusions.

In short, it’s worse for trying to determine total infections for herd immunity purposes than PCR data. Partially because PCR data is the numerator.

Who is “they”? Sweden has never tested more than 0.12% of their population in one day. New York State May have hit 0.5% a couple times and NYC claims to be able to test 0.5% per day but I can’t confirm they ever have.

First we don’t know what the herd immunity threshold (HIT)is. And it’s going to be different in different areas. I’m not enough of a mathematician to say definitively what I think the HIT is. I believe it’s well under 60% and over 30% and that it won’t look the same everywhere.

Second, if your above conundrum is true, it’s not really strange to me. I approach things from a systems perspective. Remember when I was lambasted for recommending wearing masks at home? Didn’t I essentially say locking people up in their homes without protection would lead to more spread not less when all else being equal you were 6 times more likely to be infected at home than in public? Remember when New York politicians acted shocked to learn infections were still happening rapidly under lockdown and stay at home orders? Heterogeneity and R naughts are two edged swords and cut both ways. Forcing people to not social distance and telling them masks magically don’t work in your house likely has the consequences I warned about. Artificially drive up R naught and herd immunity threshold with it and your scenario isn’t strange, it’s unavoidable in the short term.

 

I’ve been busy today and finally got all my notifications answered and then noticed this post. When I’m not quoted directly and not notified I miss things.

LOL, If the whole world lived like I do in the sticks, seldom interacting with more than one person at a time, and that being rare, the global HIT would be 2.3%

I’ll have to ruminate on the IFR flu comparison.

When you say “infections missed” do you mean ones exceeding the case counts on Worldometer or something else? Because by Worldometer data, Sweden is at 0.85% infected. But in late May by antibody testing they were at 6.1%. Even assuming there have been no infections since May (which I think we can agree is preposterous) we’ve already exceeded your top end of missing 6%. Do you accept antibody data or just official PCR? When I claimed parts of NYC were around 50% antibody prevalence in July, I was told that was “a baseless load of crap” and that I’m into pseudoscience. Now media is printing maps of such places in NYC and they are being posted here to show that I’m wrong now.

Yes, as I responded to @ronv, R naught and HIT can be manipulated in both directions. That’s the point of heterogeneity. I do believe stay at home orders in lunchbox sized apartments with no masks or social distancing in the home drove infections higher than more sensible mitigations would have. Again, when I advocated for wearing masks in the environment most likely to lead to infection I was mocked and again told I was wrong even though all the evidence supports my position.

So, this fading antibody thing has been bothering me ever since the first studies on the phenomenon came out. One of the first ones claimed antibodies in general faded fast, but neutralizing antibodies faded much more slowly. I think it may have been this one that started it all.
https://www.japantimes.co.jp/news/2...d-19-patients-decline-quickly-research-finds/

Then I found this tonight.
https://money.yahoo.com/genscript-announces-publication-clinical-data-224800134.html

It looks like to me the antibodies that fade to undetectable levels are just the generic class IgG/M antibodies. These new tests look for neutralizing antibodies specifically and find them even when other tests can’t find antibodies at all! So in theory, once these new tests replace what we’ve been using my accounting for the 40% of cases being undetectable will be unnecessary. We will have an accurate count of infections. This test is excellent for specificity as well so very few false positives. Also, it’s very likely neutralizing antibodies are formed in essentially all natural infections and remain at levels sufficient for neutralizing immunity. Thus T cell immunity takes the backseat again as just a failsafe second line of defense. I’m excited. This is good news for everyone.

Yeh, it was helpful. It made me go and try and figure out this fading antibody conundrum and now it turns out someone already has. We’ve all been hashing out the weaknesses of testing methods and it looks like we’ve made a great leap forward with new, improved antibody testing. Man, @CenterField, remember when I suggested the government or medical profession needed to set up a dedicated research collating and think tank organization? This information being out there and me stumbling across it is ridiculous. A guy eating his supper before the midnight caprine midwife duties commence shouldn’t be the one putting this all together. If we all die of corona we deserve it if we can’t coordinate science, politics, and medicine better than this. Sorry to rant, but this is ridiculous.

Link to manufacturer of new test.
https://www.genscript.com/covid-19-...ne-system-can-neutralize-the-coronavirus.html

 

I have not had much luck confirming your claim of 4) not having IgG antibodies after 40 days.
But I did find this one.

In the second study, published June 18 in Nature Medicine, researchers compared the immune responses of 37 asymptomatic but positive patients to an equal number with severe symptoms living in the Wanzhou District in China. They found that asymptomatic individuals reacted less strongly to infection, with 40 percent having undetectable levels of protective antibodies in the two to three months after the infection compared to 13 percent of the symptomatic patients.


Is this the one you quote at 40%?

https://www.the-scientist.com/news-opinion/studies-report-rapid-loss-of-covid-19-antibodies-67650

The .83% positive was your number. I think you calculated that by dividing Swedens total cases by the number of people in Sweden.
That is the number I don't think means anything.

Percent positive cases is valuable to know but it can’t predict the actual real number of infections. It’s mostly useful to determine if testing volume is adequate or how fast infections are spreading.

I would disagree. Both are samples. But there are a lot more PCR tests than antibody tests. If anything percent positives overstate the number of infections since most people get tested because they are sick or think they have been exposed.

But they are the most accurate.

Sorry. yes .1%. But they had that rate when they started.

Fair enough, but that doesn't give you a license to make one up.
It could just as easily be 5.6 like the initial estimates from China.

Of course. It's why social distancing works.

 

Yes that’s one of them. Ironically I just linked to that study in response to @CenterField a minute ago. Good catch on the symptomatic part. I’ll have to look up the other studies and see if I’m in error on that total 40%. I could be. That’s why I encourage fact checking. Thank you.

What’s the date on that? It’s better than anything I can find but I don’t know how to condense it down to a statewide number without a lot of algebra (which I hate). Lol

Oh, ok. Then we agree. I think that number is meaningless as well. That’s why I prefer antibody data as it’s easier in my opinion to get closer to reality.

I just don’t understand how percent positives has any more value to you than the actual positives that are the numerator of the percent positive fraction. Agree to disagree?

That’s what we’ve been led to believe. In reality PCR sucks. On the “best“ day it misses 20% of infected individuals. That’s one day out of a median 20 day shedding period. Overall it’s estimated PCR can miss over 50% of actual infections. In contrast, there are now antibody tests nearly 100% accurate (specificity as well that affects false positives). As I just posted to @CenterField, we will soon have antibody tests that make my manipulation of antibody data to account for fading antibodies irrelevant! The test also finds neutralizing antibodies in nearly all individuals regular antibody tests say are negative for antibodies.

Could be I don’t know how their testing started. I doubt any country started with more testing than now though.

So you missed the two links I provided earlier to models developed by actual professional mathematicians and epidemiologists? One peer reviewed, the other in a respected pre print journal. Both using heterogeneity to show herd immunity is likely somewhere between 20% and 50% if I remember right. Anyway, I’m not making anything up. I’m basing everything on the best evidence I can find.

Except when you are forced to lock down in tiny apartments.