Nate Silver - the freak out about the Israeli vaccine effectiveness is due to poor analysis.

[Peppertoo]

[link:

Link to tweet

|]

Go to the internal links.

The analysis was poorly done. Most likely, the vaccine has an 80 ish percentage effectiveness at preventing infection.

[Johnny2X2X]

1. Yup

And this study even said as much in their release, they said they weren't ready for peer review and their data was incomplete.

Link to tweet

?s=20

[48656c6c6f20]

2. So the sky isn't falling?

[PortTack]

3. I no longer trust Nate silver or 538. they have been wrong the last 3 election cycles due to poor

Methodology, and refuse to admit it, or change their methods. I will wait to see what the medical experts that analyze these kinds of studies say

[Hortensis]

5. Umhm, and due to what else? Is the enormously influential political polling industry

really the only one that doesn't have many institutions working to elect Republicans? Like CNN, MSNBC, AP, NYT, and most other MSM among media, but not the pollsters? Aside from those mentioned who set up their own polling functions to provide support for their analysis, of course.

[brooklynite]

7. What do you mean "they have been wrong"?

They don't do polling; they evaluate available data.

[Fiendish Thingy]

11. No they haven't- 538 is a poll aggregator, and does not have its own methodology/polling

538 analyzes other polls, and grades other polling houses’ methodology.

They were the most accurate predictor in 2016, giving Trump a 1 in 3 chance of winning, and predicting the razor thin margins in the rust belt.

[Mariana]

18. Some people read "1 in 3 chance" and interpret it as "zero chance".

So they insist that 538 was wrong.

[Bucky]

22. Having worked in the polling industry for many years

I can tell you one in three means "unlikely but less than a fluke", which is the perfect description for the Trump presidency

[uponit7771]

12. +1, They're weighting of right wing approval rating polls is horrible

[Peppertoo]

19. 538 has performed very well

[Bucky]

20. Sometimes 538 delivers news that I don't like to hear

Not sure that's the same as being "untrustworthy"

[TheBlackAdder]

4. I saw there was a British Health report that just being vaccinated, Delta better controlled.

.

You have to check out the additional findings that follow the pages I being at, such as the UK report Page 38-39 and more.

https://www.democraticunderground.com/?com=view_post&forum=1002&pid=15631125

.

[Scrivener7]

6. That's good. But 80% effectiveness against a variant that is 225% more contagious

than the original variant

https://www.npr.org/sections/health-shots/2021/07/17/1017075240/delta-variant-is-spreading-fast-and-new-cases-are-rising-is-time-to-mask-up-agai

means we still have a decent chance of getting the virus. Though I know I am much less likely to die or be hospitalized, we don't know how subject to debilitating long covid vaccinated people are.

Still need to be very careful.

[totodeinhere]

14. Long Covid is a problem but it's rare. Of course we have to keep studying it and trying to

determine the best therapies for it.

But I have always been careful even before this pandemic started. So I agree with being careful. But having said that, fully vaccinated people, even people with underlying conditions, are much better off if fully vaccinated.

If we concentrate too much on the rare exceptions where the fully vaccinated get serious breakthrough infections we are inadvertently giving people just another excuse for not getting vaccinated. We are already hearing from the right "why get vaccinated when it doesn't work?" Of course in most cases it does indeed work.

[hlthe2b]

8. The thread "unwrap" for Dvir Aran (Israeli data analyst)

His assessment is highly probable/credible.

https://threadreaderapp.com/thread/1420059122725183491.html

So why does it look like there is no difference until mid-July? @AArgoetti looked at the trends of cases in cities stratified by their vaccination rates.
We can see that cities on the lower quartile had low number of cases until recently, but since mid-July, most cases are coming from those cities.

This is what we got wrong. This “wave” started from cities with high vaccination rate and couldn’t “find” unvaccined adults at risk. The denominator we need to use until mid-July is >95% vaccination rate and not the country’s average.

So to summarize, the vaccines are highly effective in preventing severe cases and probably also symptomatic diseases, even for those vaccinated early.
Just to correct one misconception from my thread - many of you understood that we think there is no waning immunity. This analysis does not reject the waning hypothesis, it just shows that the vaccines are still highly effective even with possible waning.

[Peppertoo]

23. Agreed / thanks

[Yo_Mama_Been_Loggin]

9. Yet the right is jumping on the Israeli estimate story like flies on a turd.

[muriel_volestrangler]

10. But that is about "severe cases". The earlier report was about symptomatic illness.

Here:

Pfizer and BioNTech's Covid-19 vaccine is just 39% effective in Israel where the delta variant is the dominant strain, but still provides strong protection against severe illness and hospitalization, according to a new report from the country's Health Ministry.

The efficacy figure, which is based on an unspecified number of people between June 20 and July 17, is down from an earlier estimate of 64% two weeks ago and conflicts with data out of the U.K. that found the shot was 88% effective against symptomatic disease caused by the variant.

However, the two-dose vaccine still works very well in preventing people from getting seriously sick, demonstrating 88% effectiveness against hospitalization and 91% effectiveness against severe illness, according to the Israeli data published Thursday.

https://www.democraticunderground.com/100215656268

I think Silver needs to find people who were actually "freaking out" about severe illness, and talk to them directly, rather than broadbrushing the world with "a lot of you" as if he knows something special. "Probably also" is a claim without the data to back it up.

[uponit7771]

13. +1, ... the "...probably.." qualifier is the thing that sticks out in the root tweet. It's a guess

.... At what the efficacy of symptomatic is versus some heartline numbers that were presented by Israel first.

Where's the CDC has numbers on this with the Delta variant but they're only counting critical and lethal outcomes of infection versus just infection.

CDC is basically saying even with Delta around 95% of people end up in the hospital or dead

19% of people have long haul symptoms from the original variant I still don't want to catch Delta

[Peppertoo]

21. There have been people freaking out about the Israeli data but here and elsewhere

[muriel_volestrangler]

24. That doesn't make Silver correct, though. He's a professional statistician

and he didn't aim his sloppy claim at people on DU. His tweets are worse than he has shown the original Israeli report to have been.

[Peppertoo]

26. I think his tweets are valid

People all too eager to seek out outliers especially if they conform to their preconceived notions.

[Doc Sportello]

15. That was pretty obvious when the report first came out

At least from a scientific point of view. But, caveats often don't make it out to the public. And people on here were freaking out about it and when I pointed out the questionable elements of this study the poster just doubled down.

[ProfessorGAC]

17. I Know What You Mean

I think some folks really need to be out front of the worst case scenario, that they lose some of their analytical reasoning.
Without doing a deep dive into every data analysis used, I always found it concerning that the Israeli findings were so radically different than other populations under study. Just seemed peculiar that other results were so much more positive.

[BlueLucy]

16. I tried to get a 3rd Pfizer

but my pharmacy would not let me. I'm going to try my DR.

[Scrivener7]

25. That's not how it works. This is not a good idea.