5 ms·

Your first stop when evaluating a massive and heterdox claim like this should be to double-check their fundamental data against other sources [1]. What you'll

Your first stop when evaluating a massive and heterdox claim like this should be to double-check their fundamental data against other sources [1]. What you'll find is that other sources do not contain the same data at all. It's not just slightly different, it's _wildly_ different. There is no large jump in mortality in April 2021. There _is_ a large jump in morality 2020. I don't have time to thoroughly examine their research, because it doesn't come close to passing a sniff test. On a separate note, it's really discouraging how this stuff gets voted up on HN. Flagged. [1] https://ourworldindata.org/excess-mortality-covid https://ourworldindata.org/excess-mortality-covid

nicoco 4y ago
This guy apparently took the time to propose counter arguments: https://www.researchgate.net/publication/362959885_A_note_on_a_couple_of_issues_in_a_recent_manuscript_on_Excess_Mortality_in_Germany_2020-2022 https://www.researchgate.net/publication/362959885_A_note_on...
wessorh 4y ago
The two researchers actually have a number of publications in data science, please sniff again. Your kind of argument is called a an appeal to an irrelevant authority. You also forgot to cite your sources or link to the data you are advocating for. Please consider taking the time to properly advocate your point
- jlmorton 4y ago
  Two people who have never published in epidemiology stumble upon a new method for calculating excess deaths, at odds with all the existing sources, whether official sources like the WHO, domestic official government sources, or other independent groups like The Economist which have tracked this closely. And lo and behold, what do they find? A wildly different result that shows no deaths from Covid and all the deaths from the vaccine. No, I'm not going to sniff again, because it stinks to the high heavens.
  - cf141q5325 4y ago
    I think we could all appreciate if anyone with an epidemiology background could explain what disqualifies data scientists from working with this dataset. edit: Not a data scientists myself obviously
    - Arnt 4y ago
      I'm not an epidemiologist, but I've written code for people in various fields. I knew code, they knew the domain. All of them knew things that I would have fucked up, if I had tried to write the code with just my programming knowledge and my big brain, without their domain knowledge and experience. A university education teaches you something. I assume that epidemiology is like the other university fields — when lay people come and show that the epidemiologists are wrong about epidemiology, they're about as likely to be right as the people who write to maths professors claiming to have found a method of trisecting an angle. And the people with the novel claim and the missing education bear the burden of proof.
      - cf141q5325 4y ago
        I dont think that comparison is universally applicable. I wouldnt rule out that this is a lot more similar to a Math Prof telling you that the p-values in for example your psychology paper are nonsense and correcting you. edit: Not saying studying Math qualifies you to design and run studies in Psychology. But once you break down tasks enough you will often find them to be the area of expertise of another discipline. Often its very easy to tell when the Professor that taught you something in university is from another department.
  - claytongulick 4y ago
    When we close ourselves off to new information, especially if it's contrary to deeply held beliefs, our opinions devolve from informed to merely dogmatic.
    - jlmorton 4y ago
      That is absurd, and little more than a platitude. Let's set the stage here: * I am not an epidemiologist * The authors of this study are not epidemiologists * The paper has not been peer-reviewed * The authors invent a new statistical method for estimating excess deaths, it is at odds with all other sources, and it just so happens to show no excess deaths from Covid until the vaccine is released In situations like this, you should always default to disbelief.
upsidesinclude 4y ago
origin_path 4y ago
You shouldn't be flagging articles simply because you disagree with them, that's not what flagging is for. Especially because you are wrong - the OurWorldInData link you provide does indeed show German excess mortality climb steeply in April 2021 and then again in March 2022. Current mortality by that data has been at the same level as it spiked to in April 2020 (first wave), but sustained since March. I don't understand how you can look at the graphs there and not see it because the data is so plain. Remember that after COVID we should be seeing negative excess mortality due to the pull-forward effect. We should not be seeing positive excess mortality after an epidemic that near exclusively takes out the elderly. The pull forward effect can be seen after the winter 2020 wave in the German data but not later. "On a separate note, it's really discouraging how this stuff gets voted up on HN. Flagged." It's really discouraging how some people visit a site meant for "intellectual curiosity" and then immediately and incorrectly flag anything that creates cognitive dissonance in their mind, especially when they go on to make misleading statements about the data.
z7 4y ago
>There is no large jump in mortality in April 2021. Eurostat reports a 16.1% increase between March and April 2021: https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Excess_mortality_-_statistics#Excess_mortality_in_the_EU_between_January_2020_and_June_2022 https://ec.europa.eu/eurostat/statistics-explained/index.php... Europe overall: "Excess mortality levels reached a third peak in April 2021 (20.9 %)" Euromomo shows a progression from baseline level in March to "substantial increase" in April: https://www.euromomo.eu/graphs-and-maps https://www.euromomo.eu/graphs-and-maps

nicoco 4y ago

This guy apparently took the time to propose counter arguments: https://www.researchgate.net/publication/362959885_A_note_on_a_couple_of_issues_in_a_recent_manuscript_on_Excess_Mortality_in_Germany_2020-2022 https://www.researchgate.net/publication/362959885_A_note_on...

wessorh 4y ago

The two researchers actually have a number of publications in data science, please sniff again. Your kind of argument is called a an appeal to an irrelevant authority. You also forgot to cite your sources or link to the data you are advocating for. Please consider taking the time to properly advocate your point

jlmorton 4y ago
Two people who have never published in epidemiology stumble upon a new method for calculating excess deaths, at odds with all the existing sources, whether official sources like the WHO, domestic official government sources, or other independent groups like The Economist which have tracked this closely. And lo and behold, what do they find? A wildly different result that shows no deaths from Covid and all the deaths from the vaccine. No, I'm not going to sniff again, because it stinks to the high heavens.
- cf141q5325 4y ago
  I think we could all appreciate if anyone with an epidemiology background could explain what disqualifies data scientists from working with this dataset. edit: Not a data scientists myself obviously
  - Arnt 4y ago
    I'm not an epidemiologist, but I've written code for people in various fields. I knew code, they knew the domain. All of them knew things that I would have fucked up, if I had tried to write the code with just my programming knowledge and my big brain, without their domain knowledge and experience. A university education teaches you something. I assume that epidemiology is like the other university fields — when lay people come and show that the epidemiologists are wrong about epidemiology, they're about as likely to be right as the people who write to maths professors claiming to have found a method of trisecting an angle. And the people with the novel claim and the missing education bear the burden of proof.
    - cf141q5325 4y ago
      I dont think that comparison is universally applicable. I wouldnt rule out that this is a lot more similar to a Math Prof telling you that the p-values in for example your psychology paper are nonsense and correcting you. edit: Not saying studying Math qualifies you to design and run studies in Psychology. But once you break down tasks enough you will often find them to be the area of expertise of another discipline. Often its very easy to tell when the Professor that taught you something in university is from another department.
- claytongulick 4y ago
  When we close ourselves off to new information, especially if it's contrary to deeply held beliefs, our opinions devolve from informed to merely dogmatic.
  - jlmorton 4y ago
    That is absurd, and little more than a platitude. Let's set the stage here: * I am not an epidemiologist * The authors of this study are not epidemiologists * The paper has not been peer-reviewed * The authors invent a new statistical method for estimating excess deaths, it is at odds with all other sources, and it just so happens to show no excess deaths from Covid until the vaccine is released In situations like this, you should always default to disbelief.

upsidesinclude 4y ago

origin_path 4y ago

You shouldn't be flagging articles simply because you disagree with them, that's not what flagging is for. Especially because you are wrong - the OurWorldInData link you provide does indeed show German excess mortality climb steeply in April 2021 and then again in March 2022. Current mortality by that data has been at the same level as it spiked to in April 2020 (first wave), but sustained since March. I don't understand how you can look at the graphs there and not see it because the data is so plain. Remember that after COVID we should be seeing negative excess mortality due to the pull-forward effect. We should not be seeing positive excess mortality after an epidemic that near exclusively takes out the elderly. The pull forward effect can be seen after the winter 2020 wave in the German data but not later. "On a separate note, it's really discouraging how this stuff gets voted up on HN. Flagged." It's really discouraging how some people visit a site meant for "intellectual curiosity" and then immediately and incorrectly flag anything that creates cognitive dissonance in their mind, especially when they go on to make misleading statements about the data.

z7 4y ago

>There is no large jump in mortality in April 2021. Eurostat reports a 16.1% increase between March and April 2021: https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Excess_mortality_-_statistics#Excess_mortality_in_the_EU_between_January_2020_and_June_2022 https://ec.europa.eu/eurostat/statistics-explained/index.php... Europe overall: "Excess mortality levels reached a third peak in April 2021 (20.9 %)" Euromomo shows a progression from baseline level in March to "substantial increase" in April: https://www.euromomo.eu/graphs-and-maps https://www.euromomo.eu/graphs-and-maps