Howdy, Stranger!

It looks like you're new here. Sign in or register to get started.

New Ipsos US polling finds HALF of Republican voters oppose the plan to fill the Supreme Court vacan

SystemSystem Posts: 11,002
edited September 2020 in General
New Ipsos US polling finds HALF of Republican voters oppose the plan to fill the Supreme Court vacancy before the election – politicalbetting.com

The Supreme Court building in Washington DC pic.twitter.com/coXYrJYuah

Read the full story here

«13456710

Comments

  • BenpointerBenpointer Posts: 31,281
    First?
  • 2nd. like the number of national lockdowns we will be up to by end of the month.
  • BenpointerBenpointer Posts: 31,281
    edited September 2020
    Those Republican Senators who won't support an appointment before November 3rd are not likely to support a Trump nominee afterwards if Trump loses.

    On the other hand if Trump wins, why rush?

    So yes, I think the appointment takes place after inauguration, whatever the election outcome.
  • Quick. Someone has to let them know that some retired bloke who disagrees with the lockdown thinks it's almost certain that these people don't have the virus.
  • Perhaps at least one of the journalists can ask them what the FPR (False Positive) rate is for the PCR test they are using in the 100,000s?

    The public needs to know the answer.
  • RobDRobD Posts: 58,941

    Perhaps at least one of the journalists can ask them what the FPR (False Positive) rate is for the PCR test they are using in the 100,000s?

    The public needs to know the answer.
    Regardless of what the actual value is, it shouldn't change with time, should it?
  • Philip_ThompsonPhilip_Thompson Posts: 65,826
    edited September 2020

    Perhaps at least one of the journalists can ask them what the FPR (False Positive) rate is for the PCR test they are using in the 100,000s?

    The public needs to know the answer.
    If there is a high FPR then that means the rising positivity rate is an even bigger problem not a smaller one.

    This piece by Dr Michael Yeadon is absolutely devastating, if correct. It utterly blows the current covid panic and crisis completely and utterly out of the water. We are being deluged with case figures and scary graphs showing huge leaps and seeing demands for lockdown 2.0, when

    "The likelihood of an apparently positive case being a false positive is between 89-94%, or near-certainty."

    https://lockdownsceptics.org/lies-damned-lies-and-health-statistics-the-deadly-danger-of-false-positives/


    I'm really hoping some other scientists can show the flaw in his workings, because if not then we are living through easily the biggest public policy disaster since the War. We are about to cause more untold misery, mental health breakdowns, smashing our economy to pieces and killing many through neglect and lack of normal NHS treatment based on numbers that are a false. Not by a bit, but massively, massively wrong.


    What tests are they using in other countries?

    I want to see a public debate on this next week. I want MPs demanding answers from Hancock on this.

    Looks like some of these false positives are finding themselves in ICU.
    How many are have been admitted to ICU in say, the last week? He's not saying there are no + cases. Just that the real number is ≈10x less than the number Huw Edwards tells us at 10pm every night.
    Is there any mechanism by which the claims of the lockdown sceptics are falsifiable? For instance, if we do nothing and there's a large rise in hospitalizations and deaths, will that be proof that the cases were real after all, and what responsibility would the sceptics take for such an outcome?
    The "false positive" theory is falsifiable by looking at the positivity rate.

    If there was a high(ish) false positive rate then that should be a baseline from which the false positives should form thus the net cases would be the difference above that. The positivity rate then would vary very limitedly.

    EG the link hypothesises a 0.8% false positive rate and a 0.85% positivity rate thus a 0.05% real positive rate. So far so good. But then if we go from that to a 1.1% positive rate then we have seen a 0.25% increase in the positivity rate. If there is not a major false positive issue then that is a 30% increase in cases. If there is that hypothesised false positive scenario then that is a 500% increase in cases. A 500% increase in cases is far, far worse than a 30% increase in cases.
    EDIT: See the changing positivity rate here.
    image
  • RobD said:

    Perhaps at least one of the journalists can ask them what the FPR (False Positive) rate is for the PCR test they are using in the 100,000s?

    The public needs to know the answer.
    Regardless of what the actual value is, it shouldn't change with time, should it?
    From the Yeadon article:

    "The actual or operational false positive rate differs, sometimes substantially, under different settings, technical operators, detection methods and equipment."
  • BluestBlueBluestBlue Posts: 4,556

    Quick. Someone has to let them know that some retired bloke who disagrees with the lockdown thinks it's almost certain that these people don't have the virus.
    But Retired Bloke's written an article! If that doesn't convince you that none of the cases are real and that none of the dead people are dead, then what will?
  • Yeadon:

    "What do we know about the false positive rate? Well, we do know that the Government’s own scientists were very concerned about it, and a report on this problem was sent to SAGE dated June 3rd 2020. I quote: “Unless we understand the operational false positive rate of the UK’s RT-PCR testing system, we risk over-estimating the COVID-19 incidence, the demand on track and trace and the extent of asymptomatic infection”.

    In that same report, the authors helpfully listed the lowest to highest false positive rate of dozens of tests using the same technology. The lowest value for false positive rate was 0.8%. "

  • HYUFDHYUFD Posts: 116,712
  • Yeadon:

    "What do we know about the false positive rate? Well, we do know that the Government’s own scientists were very concerned about it, and a report on this problem was sent to SAGE dated June 3rd 2020. I quote: “Unless we understand the operational false positive rate of the UK’s RT-PCR testing system, we risk over-estimating the COVID-19 incidence, the demand on track and trace and the extent of asymptomatic infection”.

    In that same report, the authors helpfully listed the lowest to highest false positive rate of dozens of tests using the same technology. The lowest value for false positive rate was 0.8%. "

    Doesn't explain why the positivity rate is going up consistently.

    And it means the positivity rates increase is even worse not better. It is the R, the increase in case percentages, that is the problem not the baseload of cases.
  • RobDRobD Posts: 58,941

    RobD said:

    Perhaps at least one of the journalists can ask them what the FPR (False Positive) rate is for the PCR test they are using in the 100,000s?

    The public needs to know the answer.
    Regardless of what the actual value is, it shouldn't change with time, should it?
    From the Yeadon article:

    "The actual or operational false positive rate differs, sometimes substantially, under different settings, technical operators, detection methods and equipment."
    Sounds like it'd move somewhat randomly, not consistently in one direction.
  • FF43FF43 Posts: 15,545
    FPT

    This piece by Dr Michael Yeadon is absolutely devastating, if correct. It utterly blows the current covid panic and crisis completely and utterly out of the water. We are being deluged with case figures and scary graphs showing huge leaps and seeing demands for lockdown 2.0, when

    "The likelihood of an apparently positive case being a false positive is between 89-94%, or near-certainty."

    https://lockdownsceptics.org/lies-damned-lies-and-health-statistics-the-deadly-danger-of-false-positives/


    I'm really hoping some other scientists can show the flaw in his workings, because if not then we are living through easily the biggest public policy disaster since the War. We are about to cause more untold misery, mental health breakdowns, smashing our economy to pieces and killing many through neglect and lack of normal NHS treatment based on numbers that are a false. Not by a bit, but massively, massively wrong.


    What tests are they using in other countries?

    I want to see a public debate on this next week. I want MPs demanding answers from Hancock on this.

    Jason Leitch, Clinical Director at the Scottish Government was talking about what I assume is the problem referred to here. The test tells you the virus is present, but it doesn't tell you what you really want to know: is the person infectious?

    The lockdownsceptics man is barking up the wrong tree, I suspect. There is nothing massively wrong with the statistics. The test tests what it does. If there are ten times more positives it's because the epidemic has grown tenfold. What does he want? Stop all testing so we have no idea what's going on?

    The problem is people being asked to self isolate because they are no longer infectious or because they have come into contact with someone who is no longer infectious. Also Test and Protect can't focus on the traces that really matter because they don't know who is infectious and who isn't.

  • BenpointerBenpointer Posts: 31,281

    Perhaps at least one of the journalists can ask them what the FPR (False Positive) rate is for the PCR test they are using in the 100,000s?

    The public needs to know the answer.
    Good job Operation Moonshot is, er... total bollocks made up on the spur of the moment by our Chief State Liar
  • Scott_xP said:
    Only six months.

    I don't believe them. If there is no working and mass available vaccine by next spring then we go through the cycle again.
  • Scott_xP said:
    Only six months.

    I don't believe them. If there is no working and mass available vaccine by next spring then we go through the cycle again.
    I believe them, we'll have a mass available vaccine by then.
  • OmniumOmnium Posts: 9,688
    HYUFD said:
    More about total debt really given these are exceptional measures.
  • rottenboroughrottenborough Posts: 58,043
    edited September 2020

    Yeadon:

    "What do we know about the false positive rate? Well, we do know that the Government’s own scientists were very concerned about it, and a report on this problem was sent to SAGE dated June 3rd 2020. I quote: “Unless we understand the operational false positive rate of the UK’s RT-PCR testing system, we risk over-estimating the COVID-19 incidence, the demand on track and trace and the extent of asymptomatic infection”.

    In that same report, the authors helpfully listed the lowest to highest false positive rate of dozens of tests using the same technology. The lowest value for false positive rate was 0.8%. "

    Doesn't explain why the positivity rate is going up consistently.

    And it means the positivity rates increase is even worse not better. It is the R, the increase in case percentages, that is the problem not the baseload of cases.
    He's not saying there aren't more genuine cases (if I have understood things), he's saying the high numbers we see all over our TV screens and which are driving public policy are wrong. 10x wrong.



  • NigelbNigelb Posts: 61,575
    That is an interesting poll.
    It tends to confirm the belief that perceived blatant unfairness often transcends voters’ political affiliation.
  • Omnium said:

    HYUFD said:
    More about total debt really given these are exceptional measures.
    True. But the structural deficit is probably in big trouble as well as when we lockdown again which this government now seems hell bent on, the economy will shrink (and is less likely to bounce back every time we go through this). Tax base will start to shrink.

  • Philip_ThompsonPhilip_Thompson Posts: 65,826
    edited September 2020

    Yeadon:

    "What do we know about the false positive rate? Well, we do know that the Government’s own scientists were very concerned about it, and a report on this problem was sent to SAGE dated June 3rd 2020. I quote: “Unless we understand the operational false positive rate of the UK’s RT-PCR testing system, we risk over-estimating the COVID-19 incidence, the demand on track and trace and the extent of asymptomatic infection”.

    In that same report, the authors helpfully listed the lowest to highest false positive rate of dozens of tests using the same technology. The lowest value for false positive rate was 0.8%. "

    Doesn't explain why the positivity rate is going up consistently.

    And it means the positivity rates increase is even worse not better. It is the R, the increase in case percentages, that is the problem not the baseload of cases.
    He's not saying there aren't more genuine cases (if I have understood things), he's saying the high numbers we see all over our TV screens and which are driving public policy are wrong. 10x wrong.



    I hope he's not right because if it is then that makes the increase dramatically worse.

    Lets say that hypothetically there were 3000 cases but 90% were false, then there were just 300 real cases. 2700 false cases should remain the baseline.

    If suddenly ceteris paribus we have 5000 cases with 2700 being false then now we have 2300 real cases.

    If there aren't so many false positives then the caseload has gone up by just 66%.
    If there are so many false positives then the caseload has gone up by 666%.

    The worrying news isn't the number of cases it is the increase in cases. That is the problem and if your theory is right then the increase is more concerning not less.
  • Scott_xP said:
    Only six months.

    I don't believe them. If there is no working and mass available vaccine by next spring then we go through the cycle again.
    I believe them, we'll have a mass available vaccine by then.
    Let's hope so, because the only other way out is Sweden and for whatever reasons that will either not work here because we can't be trusted to obey rules and apply common sense (possibly correct) or will never be tried because the Ferguson model is seen as the dominant driver.
  • NigelbNigelb Posts: 61,575

    Yeadon:

    "What do we know about the false positive rate? Well, we do know that the Government’s own scientists were very concerned about it, and a report on this problem was sent to SAGE dated June 3rd 2020. I quote: “Unless we understand the operational false positive rate of the UK’s RT-PCR testing system, we risk over-estimating the COVID-19 incidence, the demand on track and trace and the extent of asymptomatic infection”.

    In that same report, the authors helpfully listed the lowest to highest false positive rate of dozens of tests using the same technology. The lowest value for false positive rate was 0.8%. "

    Doesn't explain why the positivity rate is going up consistently.

    And it means the positivity rates increase is even worse not better. It is the R, the increase in case percentages, that is the problem not the baseload of cases.
    He's not saying there aren't more genuine cases (if I have understood things), he's saying the high numbers we see all over our TV screens and which are driving public policy are wrong. 10x wrong.

    But his argument only makes even theoretical sense if you posit a steadily increasing percentage rate of false positives. Which seems unlikely, to put it mildly.

  • BenpointerBenpointer Posts: 31,281

    Perhaps at least one of the journalists can ask them what the FPR (False Positive) rate is for the PCR test they are using in the 100,000s?

    The public needs to know the answer.
    If there is a high FPR then that means the rising positivity rate is an even bigger problem not a smaller one.

    This piece by Dr Michael Yeadon is absolutely devastating, if correct. It utterly blows the current covid panic and crisis completely and utterly out of the water. We are being deluged with case figures and scary graphs showing huge leaps and seeing demands for lockdown 2.0, when

    "The likelihood of an apparently positive case being a false positive is between 89-94%, or near-certainty."

    https://lockdownsceptics.org/lies-damned-lies-and-health-statistics-the-deadly-danger-of-false-positives/


    I'm really hoping some other scientists can show the flaw in his workings, because if not then we are living through easily the biggest public policy disaster since the War. We are about to cause more untold misery, mental health breakdowns, smashing our economy to pieces and killing many through neglect and lack of normal NHS treatment based on numbers that are a false. Not by a bit, but massively, massively wrong.


    What tests are they using in other countries?

    I want to see a public debate on this next week. I want MPs demanding answers from Hancock on this.

    Looks like some of these false positives are finding themselves in ICU.
    How many are have been admitted to ICU in say, the last week? He's not saying there are no + cases. Just that the real number is ≈10x less than the number Huw Edwards tells us at 10pm every night.
    Is there any mechanism by which the claims of the lockdown sceptics are falsifiable? For instance, if we do nothing and there's a large rise in hospitalizations and deaths, will that be proof that the cases were real after all, and what responsibility would the sceptics take for such an outcome?
    The "false positive" theory is falsifiable by looking at the positivity rate.

    If there was a high(ish) false positive rate then that should be a baseline from which the false positives should form thus the net cases would be the difference above that. The positivity rate then would vary very limitedly.

    EG the link hypothesises a 0.8% false positive rate and a 0.85% positivity rate thus a 0.05% real positive rate. So far so good. But then if we go from that to a 1.1% positive rate then we have seen a 0.25% increase in the positivity rate. If there is not a major false positive issue then that is a 30% increase in cases. If there is that hypothesised false positive scenario then that is a 500% increase in cases. A 500% increase in cases is far, far worse than a 30% increase in cases.
    EDIT: See the changing positivity rate here.
    image
    Hang on. That's graph is headed "Estimated percentage of the population in England testing positive", it's not titled "Estimated percentage of tests in England testing positive".

    Surely more tests = more positive tests, even if the infection rate stayed the same?
  • IshmaelZIshmaelZ Posts: 21,830

    Quick. Someone has to let them know that some retired bloke who disagrees with the lockdown thinks it's almost certain that these people don't have the virus.
    But Retired Bloke's written an article! If that doesn't convince you that none of the cases are real and that none of the dead people are dead, then what will?
    The arithmetic isn't difficult. Say there are 60 million people in the country and one of them has the disease. You test all of them. If your test has an FPR of 0.8% (that is, is 99.2% accurate) you find 480,001 cases when actually there is 1. This is how false positives work, and is why when an initial mammogram detects breast cancer you conduct further tests rather than lopping the offending breast straight off. You do not make substantive decisions on the basis of the first test, which is what we are now doing.

    This is a problem. Rather than taking the perplexingly ad hominem point that he is retired, explain why it is not a problem please?
  • This is an absolute disaster if the papers are right.
  • Perhaps at least one of the journalists can ask them what the FPR (False Positive) rate is for the PCR test they are using in the 100,000s?

    The public needs to know the answer.
    If there is a high FPR then that means the rising positivity rate is an even bigger problem not a smaller one.

    This piece by Dr Michael Yeadon is absolutely devastating, if correct. It utterly blows the current covid panic and crisis completely and utterly out of the water. We are being deluged with case figures and scary graphs showing huge leaps and seeing demands for lockdown 2.0, when

    "The likelihood of an apparently positive case being a false positive is between 89-94%, or near-certainty."

    https://lockdownsceptics.org/lies-damned-lies-and-health-statistics-the-deadly-danger-of-false-positives/


    I'm really hoping some other scientists can show the flaw in his workings, because if not then we are living through easily the biggest public policy disaster since the War. We are about to cause more untold misery, mental health breakdowns, smashing our economy to pieces and killing many through neglect and lack of normal NHS treatment based on numbers that are a false. Not by a bit, but massively, massively wrong.


    What tests are they using in other countries?

    I want to see a public debate on this next week. I want MPs demanding answers from Hancock on this.

    Looks like some of these false positives are finding themselves in ICU.
    How many are have been admitted to ICU in say, the last week? He's not saying there are no + cases. Just that the real number is ≈10x less than the number Huw Edwards tells us at 10pm every night.
    Is there any mechanism by which the claims of the lockdown sceptics are falsifiable? For instance, if we do nothing and there's a large rise in hospitalizations and deaths, will that be proof that the cases were real after all, and what responsibility would the sceptics take for such an outcome?
    The "false positive" theory is falsifiable by looking at the positivity rate.

    If there was a high(ish) false positive rate then that should be a baseline from which the false positives should form thus the net cases would be the difference above that. The positivity rate then would vary very limitedly.

    EG the link hypothesises a 0.8% false positive rate and a 0.85% positivity rate thus a 0.05% real positive rate. So far so good. But then if we go from that to a 1.1% positive rate then we have seen a 0.25% increase in the positivity rate. If there is not a major false positive issue then that is a 30% increase in cases. If there is that hypothesised false positive scenario then that is a 500% increase in cases. A 500% increase in cases is far, far worse than a 30% increase in cases.
    EDIT: See the changing positivity rate here.
    image
    Hang on. That's graph is headed "Estimated percentage of the population in England testing positive", it's not titled "Estimated percentage of tests in England testing positive".

    Surely more tests = more positive tests, even if the infection rate stayed the same?
    Because its a scientific survey of the population. They're already taking the quantity of tests relative to the population into account. Thus making week-on-week comparisons legitimate.
  • Nigelb said:

    Yeadon:

    "What do we know about the false positive rate? Well, we do know that the Government’s own scientists were very concerned about it, and a report on this problem was sent to SAGE dated June 3rd 2020. I quote: “Unless we understand the operational false positive rate of the UK’s RT-PCR testing system, we risk over-estimating the COVID-19 incidence, the demand on track and trace and the extent of asymptomatic infection”.

    In that same report, the authors helpfully listed the lowest to highest false positive rate of dozens of tests using the same technology. The lowest value for false positive rate was 0.8%. "

    Doesn't explain why the positivity rate is going up consistently.

    And it means the positivity rates increase is even worse not better. It is the R, the increase in case percentages, that is the problem not the baseload of cases.
    He's not saying there aren't more genuine cases (if I have understood things), he's saying the high numbers we see all over our TV screens and which are driving public policy are wrong. 10x wrong.

    But his argument only makes even theoretical sense if you posit a steadily increasing percentage rate of false positives. Which seems unlikely, to put it mildly.

    It's the number of tests thats going up, not the % of false positives. The % is taken of the total number of tests. So the more tests the more false ones you get.

    But maybe read the article, it is a long time since I studied maths so maybe I have got the wrong end of all this.

    All I'm saying is we need some debate on all this and we aren't getting it. Hancock appears unwilling to discuss the FPR.
  • OmniumOmnium Posts: 9,688

    Omnium said:

    HYUFD said:
    More about total debt really given these are exceptional measures.
    True. But the structural deficit is probably in big trouble as well as when we lockdown again which this government now seems hell bent on, the economy will shrink (and is less likely to bounce back every time we go through this). Tax base will start to shrink.

    Yep.

    In real terms the economy will shrink, but the only escape route for the government is massive printing of money, so perhaps not in nominal terms. It is the right escape route as it's basically a blanket tax on everything. The normal resultant inflationary risk is there, but probably won't come into play.
  • JonathanJonathan Posts: 20,901
    The genius of right wing politics is to claim that government doesn’t work in opposition and through their incompetence generate evidence for their views in office.
  • IshmaelZ said:

    Quick. Someone has to let them know that some retired bloke who disagrees with the lockdown thinks it's almost certain that these people don't have the virus.
    But Retired Bloke's written an article! If that doesn't convince you that none of the cases are real and that none of the dead people are dead, then what will?
    The arithmetic isn't difficult. Say there are 60 million people in the country and one of them has the disease. You test all of them. If your test has an FPR of 0.8% (that is, is 99.2% accurate) you find 480,001 cases when actually there is 1. This is how false positives work, and is why when an initial mammogram detects breast cancer you conduct further tests rather than lopping the offending breast straight off. You do not make substantive decisions on the basis of the first test, which is what we are now doing.

    This is a problem. Rather than taking the perplexingly ad hominem point that he is retired, explain why it is not a problem please?
    Because the worrying news isn't the cases it is the increasing rate of cases.

    If the number of cases were there but flat then that would not be concerning. It is the increase that is concerning. Unless the FPR rate is consistently rising that makes no sense whatsoever.
  • FF43FF43 Posts: 15,545
    FF43 said:

    FPT

    This piece by Dr Michael Yeadon is absolutely devastating, if correct. It utterly blows the current covid panic and crisis completely and utterly out of the water. We are being deluged with case figures and scary graphs showing huge leaps and seeing demands for lockdown 2.0, when

    "The likelihood of an apparently positive case being a false positive is between 89-94%, or near-certainty."

    https://lockdownsceptics.org/lies-damned-lies-and-health-statistics-the-deadly-danger-of-false-positives/


    I'm really hoping some other scientists can show the flaw in his workings, because if not then we are living through easily the biggest public policy disaster since the War. We are about to cause more untold misery, mental health breakdowns, smashing our economy to pieces and killing many through neglect and lack of normal NHS treatment based on numbers that are a false. Not by a bit, but massively, massively wrong.


    What tests are they using in other countries?

    I want to see a public debate on this next week. I want MPs demanding answers from Hancock on this.

    Jason Leitch, Clinical Director at the Scottish Government was talking about what I assume is the problem referred to here. The test tells you the virus is present, but it doesn't tell you what you really want to know: is the person infectious?

    The lockdownsceptics man is barking up the wrong tree, I suspect. There is nothing massively wrong with the statistics. The test tests what it does. If there are ten times more positives it's because the epidemic has grown tenfold. What does he want? Stop all testing so we have no idea what's going on?

    The problem is people being asked to self isolate because they are no longer infectious or because they have come into contact with someone who is no longer infectious. Also Test and Protect can't focus on the traces that really matter because they don't know who is infectious and who isn't.

    If there are ten times more positives it's because the epidemic has grown tenfold.

    Thinking about this further, a tenfold increase in test positives will actually underestimate the increase in cases in this situation. The base date tests will include those holding the virus from earlier while tests today will be picking up additional cases that are all new. The reverse happens on a fall in cases, where the fall will be overestimated.
  • eristdooferistdoof Posts: 4,861

    This piece by Dr Michael Yeadon is absolutely devastating, if correct. It utterly blows the current covid panic and crisis completely and utterly out of the water. We are being deluged with case figures and scary graphs showing huge leaps and seeing demands for lockdown 2.0, when

    "The likelihood of an apparently positive case being a false positive is between 89-94%, or near-certainty."

    https://lockdownsceptics.org/lies-damned-lies-and-health-statistics-the-deadly-danger-of-false-positives/


    I'm really hoping some other scientists can show the flaw in his workings, because if not then we are living through easily the biggest public policy disaster since the War. We are about to cause more untold misery, mental health breakdowns, smashing our economy to pieces and killing many through neglect and lack of normal NHS treatment based on numbers that are a false. Not by a bit, but massively, massively wrong.


    What tests are they using in other countries?

    I want to see a public debate on this next week. I want MPs demanding answers from Hancock on this.

    You are assuming the epidemiologists advising the government have never heard of this False positive rate (FPR) problem which occurs when lots of tests are done with a low level of true positives. This is very basic probability. I teach it in service courses (ie to non-maths students). It is entry level epidemiology. If the virologists and epidemiologists are warning that there is a real problem, this will not be because of a high FPR, they will have taken account of the FPR aready.

    The number of false positives will be proportional to the number of tests. If the number of positive cases doubles, but the number of tests in that same period has gone up by 10%, then the increase will not have be dominated by false positives. Also if the number of positives almost entirely consists of false, positives the R value will not increase and this will be picked up by the very short infection chains, as a false positive by definition cannot infect anybody else.

    Good night!
  • dixiedeandixiedean Posts: 27,837

    This is an absolute disaster if the papers are right.

    Owt specific or just in general?
  • HYUFDHYUFD Posts: 116,712
    edited September 2020

    Scott_xP said:
    Only six months.

    I don't believe them. If there is no working and mass available vaccine by next spring then we go through the cycle again.
    I believe them, we'll have a mass available vaccine by then.
    Let's hope so, because the only other way out is Sweden and for whatever reasons that will either not work here because we can't be trusted to obey rules and apply common sense (possibly correct) or will never be tried because the Ferguson model is seen as the dominant driver.
    South Korea would be the better model than Sweden as it involves masks as well as social distancing while also trying to avoid the economic damage of a full lockdown, however overall yes, it could be years until we get a vaccine, in theory we could never get a vaccine, in which case we cannot spend the rest of our lives like this
  • IshmaelZIshmaelZ Posts: 21,830
    Nigelb said:

    Yeadon:

    "What do we know about the false positive rate? Well, we do know that the Government’s own scientists were very concerned about it, and a report on this problem was sent to SAGE dated June 3rd 2020. I quote: “Unless we understand the operational false positive rate of the UK’s RT-PCR testing system, we risk over-estimating the COVID-19 incidence, the demand on track and trace and the extent of asymptomatic infection”.

    In that same report, the authors helpfully listed the lowest to highest false positive rate of dozens of tests using the same technology. The lowest value for false positive rate was 0.8%. "

    Doesn't explain why the positivity rate is going up consistently.

    And it means the positivity rates increase is even worse not better. It is the R, the increase in case percentages, that is the problem not the baseload of cases.
    He's not saying there aren't more genuine cases (if I have understood things), he's saying the high numbers we see all over our TV screens and which are driving public policy are wrong. 10x wrong.

    But his argument only makes even theoretical sense if you posit a steadily increasing percentage rate of false positives. Which seems unlikely, to put it mildly.

    He is surely saying the % rate is absolute at 0.8%. Test a thousand, get 8. Test a million, get 8,000. So if the *true* rate is also 0.8% your stats are "only" 50% out (and get better as the true rate rises), but if only 3 people have it they are ludicrously wrong.
  • Yeadon:

    "What do we know about the false positive rate? Well, we do know that the Government’s own scientists were very concerned about it, and a report on this problem was sent to SAGE dated June 3rd 2020. I quote: “Unless we understand the operational false positive rate of the UK’s RT-PCR testing system, we risk over-estimating the COVID-19 incidence, the demand on track and trace and the extent of asymptomatic infection”.

    In that same report, the authors helpfully listed the lowest to highest false positive rate of dozens of tests using the same technology. The lowest value for false positive rate was 0.8%. "

    Doesn't explain why the positivity rate is going up consistently.

    And it means the positivity rates increase is even worse not better. It is the R, the increase in case percentages, that is the problem not the baseload of cases.
    He's not saying there aren't more genuine cases (if I have understood things), he's saying the high numbers we see all over our TV screens and which are driving public policy are wrong. 10x wrong.



    I hope he's not right because if it is then that makes the increase dramatically worse.

    Lets say that hypothetically there were 3000 cases but 90% were false, then there were just 300 real cases. 2700 false cases should remain the baseline.

    If suddenly ceteris paribus we have 5000 cases with 2700 being false then now we have 2300 real cases.

    If there aren't so many false positives then the caseload has gone up by just 66%.
    If there are so many false positives then the caseload has gone up by 666%.

    The worrying news isn't the number of cases it is the increase in cases. That is the problem and if your theory is right then the increase is more concerning not less.
    I don't understand your baseline, but maybe it is getting too late for my brain.

    How many were tested in your second case?
  • Nigelb said:

    Yeadon:

    "What do we know about the false positive rate? Well, we do know that the Government’s own scientists were very concerned about it, and a report on this problem was sent to SAGE dated June 3rd 2020. I quote: “Unless we understand the operational false positive rate of the UK’s RT-PCR testing system, we risk over-estimating the COVID-19 incidence, the demand on track and trace and the extent of asymptomatic infection”.

    In that same report, the authors helpfully listed the lowest to highest false positive rate of dozens of tests using the same technology. The lowest value for false positive rate was 0.8%. "

    Doesn't explain why the positivity rate is going up consistently.

    And it means the positivity rates increase is even worse not better. It is the R, the increase in case percentages, that is the problem not the baseload of cases.
    He's not saying there aren't more genuine cases (if I have understood things), he's saying the high numbers we see all over our TV screens and which are driving public policy are wrong. 10x wrong.

    But his argument only makes even theoretical sense if you posit a steadily increasing percentage rate of false positives. Which seems unlikely, to put it mildly.

    It's the number of tests thats going up, not the % of false positives. The % is taken of the total number of tests. So the more tests the more false ones you get.

    But maybe read the article, it is a long time since I studied maths so maybe I have got the wrong end of all this.

    All I'm saying is we need some debate on all this and we aren't getting it. Hancock appears unwilling to discuss the FPR.
    That's not true. The % of cases coming back positive is going up.

    If the % coming back positive was flat then yes it could be explained by the # of tests going up. But that's not the case.
  • BenpointerBenpointer Posts: 31,281

    Perhaps at least one of the journalists can ask them what the FPR (False Positive) rate is for the PCR test they are using in the 100,000s?

    The public needs to know the answer.
    If there is a high FPR then that means the rising positivity rate is an even bigger problem not a smaller one.

    This piece by Dr Michael Yeadon is absolutely devastating, if correct. It utterly blows the current covid panic and crisis completely and utterly out of the water. We are being deluged with case figures and scary graphs showing huge leaps and seeing demands for lockdown 2.0, when

    "The likelihood of an apparently positive case being a false positive is between 89-94%, or near-certainty."

    https://lockdownsceptics.org/lies-damned-lies-and-health-statistics-the-deadly-danger-of-false-positives/


    I'm really hoping some other scientists can show the flaw in his workings, because if not then we are living through easily the biggest public policy disaster since the War. We are about to cause more untold misery, mental health breakdowns, smashing our economy to pieces and killing many through neglect and lack of normal NHS treatment based on numbers that are a false. Not by a bit, but massively, massively wrong.


    What tests are they using in other countries?

    I want to see a public debate on this next week. I want MPs demanding answers from Hancock on this.

    Looks like some of these false positives are finding themselves in ICU.
    How many are have been admitted to ICU in say, the last week? He's not saying there are no + cases. Just that the real number is ≈10x less than the number Huw Edwards tells us at 10pm every night.
    Is there any mechanism by which the claims of the lockdown sceptics are falsifiable? For instance, if we do nothing and there's a large rise in hospitalizations and deaths, will that be proof that the cases were real after all, and what responsibility would the sceptics take for such an outcome?
    The "false positive" theory is falsifiable by looking at the positivity rate.

    If there was a high(ish) false positive rate then that should be a baseline from which the false positives should form thus the net cases would be the difference above that. The positivity rate then would vary very limitedly.

    EG the link hypothesises a 0.8% false positive rate and a 0.85% positivity rate thus a 0.05% real positive rate. So far so good. But then if we go from that to a 1.1% positive rate then we have seen a 0.25% increase in the positivity rate. If there is not a major false positive issue then that is a 30% increase in cases. If there is that hypothesised false positive scenario then that is a 500% increase in cases. A 500% increase in cases is far, far worse than a 30% increase in cases.
    EDIT: See the changing positivity rate here.
    image
    Hang on. That's graph is headed "Estimated percentage of the population in England testing positive", it's not titled "Estimated percentage of tests in England testing positive".

    Surely more tests = more positive tests, even if the infection rate stayed the same?
    Because its a scientific survey of the population. They're already taking the quantity of tests relative to the population into account. Thus making week-on-week comparisons legitimate.
    How is that different to "percentage of tests giving positive results"?

    And how does it deal with the false positives - the challenge presented by which is well aired in several other posts on this thread?
  • HYUFDHYUFD Posts: 116,712
    Scott_xP said:
    Hancock and Gove most supportive of another lockdown apparently, Patal and Sunak opposed
  • HYUFD said:

    Scott_xP said:
    Only six months.

    I don't believe them. If there is no working and mass available vaccine by next spring then we go through the cycle again.
    I believe them, we'll have a mass available vaccine by then.
    Let's hope so, because the only other way out is Sweden and for whatever reasons that will either not work here because we can't be trusted to obey rules and apply common sense (possibly correct) or will never be tried because the Ferguson model is seen as the dominant driver.
    South Korea would be the better model than Sweden as it involves masks as well as social distancing while also trying to avoid the economic damage of a full lockdown, however overall yes, it could be years until we get a vaccine, in theory we could never get a vaccine, in which case we cannot spend the rest of our lives like this
    I kinda ruled out SK because I doubt we would put up with the testing regime they seem to have used, but I haven't really been following what they have been doing recently.
  • Perhaps at least one of the journalists can ask them what the FPR (False Positive) rate is for the PCR test they are using in the 100,000s?

    The public needs to know the answer.
    If there is a high FPR then that means the rising positivity rate is an even bigger problem not a smaller one.

    This piece by Dr Michael Yeadon is absolutely devastating, if correct. It utterly blows the current covid panic and crisis completely and utterly out of the water. We are being deluged with case figures and scary graphs showing huge leaps and seeing demands for lockdown 2.0, when

    "The likelihood of an apparently positive case being a false positive is between 89-94%, or near-certainty."

    https://lockdownsceptics.org/lies-damned-lies-and-health-statistics-the-deadly-danger-of-false-positives/


    I'm really hoping some other scientists can show the flaw in his workings, because if not then we are living through easily the biggest public policy disaster since the War. We are about to cause more untold misery, mental health breakdowns, smashing our economy to pieces and killing many through neglect and lack of normal NHS treatment based on numbers that are a false. Not by a bit, but massively, massively wrong.


    What tests are they using in other countries?

    I want to see a public debate on this next week. I want MPs demanding answers from Hancock on this.

    Looks like some of these false positives are finding themselves in ICU.
    How many are have been admitted to ICU in say, the last week? He's not saying there are no + cases. Just that the real number is ≈10x less than the number Huw Edwards tells us at 10pm every night.
    Is there any mechanism by which the claims of the lockdown sceptics are falsifiable? For instance, if we do nothing and there's a large rise in hospitalizations and deaths, will that be proof that the cases were real after all, and what responsibility would the sceptics take for such an outcome?
    The "false positive" theory is falsifiable by looking at the positivity rate.

    If there was a high(ish) false positive rate then that should be a baseline from which the false positives should form thus the net cases would be the difference above that. The positivity rate then would vary very limitedly.

    EG the link hypothesises a 0.8% false positive rate and a 0.85% positivity rate thus a 0.05% real positive rate. So far so good. But then if we go from that to a 1.1% positive rate then we have seen a 0.25% increase in the positivity rate. If there is not a major false positive issue then that is a 30% increase in cases. If there is that hypothesised false positive scenario then that is a 500% increase in cases. A 500% increase in cases is far, far worse than a 30% increase in cases.
    EDIT: See the changing positivity rate here.
    image
    Hang on. That's graph is headed "Estimated percentage of the population in England testing positive", it's not titled "Estimated percentage of tests in England testing positive".

    Surely more tests = more positive tests, even if the infection rate stayed the same?
    Because its a scientific survey of the population. They're already taking the quantity of tests relative to the population into account. Thus making week-on-week comparisons legitimate.
    How is that different to "percentage of tests giving positive results"?

    And how does it deal with the false positives - the challenge presented by which is well aired in several other posts on this thread?
    Because the ONS survey is a scientific survey that has nothing to do with Pillar 1 and Pillar 2 testing. Consider it more akin to the difference between a scientific Opinion Poll and a Voodoo Poll.

    The percentage of tests giving positive results (in Pillar One and Pillar Two) has gone up as well as the ONS survey data going up.
  • CharlesCharles Posts: 35,758
    HYUFD said:
    If you are printing money then the size of the deficit is meaningless
  • RobDRobD Posts: 58,941

    Perhaps at least one of the journalists can ask them what the FPR (False Positive) rate is for the PCR test they are using in the 100,000s?

    The public needs to know the answer.
    If there is a high FPR then that means the rising positivity rate is an even bigger problem not a smaller one.

    This piece by Dr Michael Yeadon is absolutely devastating, if correct. It utterly blows the current covid panic and crisis completely and utterly out of the water. We are being deluged with case figures and scary graphs showing huge leaps and seeing demands for lockdown 2.0, when

    "The likelihood of an apparently positive case being a false positive is between 89-94%, or near-certainty."

    https://lockdownsceptics.org/lies-damned-lies-and-health-statistics-the-deadly-danger-of-false-positives/


    I'm really hoping some other scientists can show the flaw in his workings, because if not then we are living through easily the biggest public policy disaster since the War. We are about to cause more untold misery, mental health breakdowns, smashing our economy to pieces and killing many through neglect and lack of normal NHS treatment based on numbers that are a false. Not by a bit, but massively, massively wrong.


    What tests are they using in other countries?

    I want to see a public debate on this next week. I want MPs demanding answers from Hancock on this.

    Looks like some of these false positives are finding themselves in ICU.
    How many are have been admitted to ICU in say, the last week? He's not saying there are no + cases. Just that the real number is ≈10x less than the number Huw Edwards tells us at 10pm every night.
    Is there any mechanism by which the claims of the lockdown sceptics are falsifiable? For instance, if we do nothing and there's a large rise in hospitalizations and deaths, will that be proof that the cases were real after all, and what responsibility would the sceptics take for such an outcome?
    The "false positive" theory is falsifiable by looking at the positivity rate.

    If there was a high(ish) false positive rate then that should be a baseline from which the false positives should form thus the net cases would be the difference above that. The positivity rate then would vary very limitedly.

    EG the link hypothesises a 0.8% false positive rate and a 0.85% positivity rate thus a 0.05% real positive rate. So far so good. But then if we go from that to a 1.1% positive rate then we have seen a 0.25% increase in the positivity rate. If there is not a major false positive issue then that is a 30% increase in cases. If there is that hypothesised false positive scenario then that is a 500% increase in cases. A 500% increase in cases is far, far worse than a 30% increase in cases.
    EDIT: See the changing positivity rate here.
    image
    Hang on. That's graph is headed "Estimated percentage of the population in England testing positive", it's not titled "Estimated percentage of tests in England testing positive".

    Surely more tests = more positive tests, even if the infection rate stayed the same?
    Because its a scientific survey of the population. They're already taking the quantity of tests relative to the population into account. Thus making week-on-week comparisons legitimate.
    How is that different to "percentage of tests giving positive results"?

    And how does it deal with the false positives - the challenge presented by which is well aired in several other posts on this thread?
    Because tests aren't distributed randomly. They are focused on areas with outbreaks, for example.
  • geoffwgeoffw Posts: 8,092

    Perhaps at least one of the journalists can ask them what the FPR (False Positive) rate is for the PCR test they are using in the 100,000s?

    The public needs to know the answer.
    fpt

    What if the virus isn't weaker, nor are our bodies better at coping with it as we've had some exposure. What if the low numbers need hospital compared to the huge spike in the cases is because of the false positives?

    Despite the somewhat breathless and dramatic style, Dr MikeYeadon is right about the potentially seriously misleading impression that false positives can give about the scale of infection. It seems to me that there are clearly issues here which should be debated openly in public. Kudos to Julia Hartley-Brewer for her efforts in that direction.


  • eristdoof said:

    This piece by Dr Michael Yeadon is absolutely devastating, if correct. It utterly blows the current covid panic and crisis completely and utterly out of the water. We are being deluged with case figures and scary graphs showing huge leaps and seeing demands for lockdown 2.0, when

    "The likelihood of an apparently positive case being a false positive is between 89-94%, or near-certainty."

    https://lockdownsceptics.org/lies-damned-lies-and-health-statistics-the-deadly-danger-of-false-positives/


    I'm really hoping some other scientists can show the flaw in his workings, because if not then we are living through easily the biggest public policy disaster since the War. We are about to cause more untold misery, mental health breakdowns, smashing our economy to pieces and killing many through neglect and lack of normal NHS treatment based on numbers that are a false. Not by a bit, but massively, massively wrong.


    What tests are they using in other countries?

    I want to see a public debate on this next week. I want MPs demanding answers from Hancock on this.

    You are assuming the epidemiologists advising the government have never heard of this False positive rate (FPR) problem which occurs when lots of tests are done with a low level of true positives. This is very basic probability. I teach it in service courses (ie to non-maths students). It is entry level epidemiology. If the virologists and epidemiologists are warning that there is a real problem, this will not be because of a high FPR, they will have taken account of the FPR aready.

    The number of false positives will be proportional to the number of tests. If the number of positive cases doubles, but the number of tests in that same period has gone up by 10%, then the increase will not have be dominated by false positives. Also if the number of positives almost entirely consists of false, positives the R value will not increase and this will be picked up by the very short infection chains, as a false positive by definition cannot infect anybody else.

    Good night!
    As he writes:

    "What do we know about the false positive rate? Well, we do know that the Government’s own scientists were very concerned about it, and a report on this problem was sent to SAGE dated June 3rd 2020. I quote: “Unless we understand the operational false positive rate of the UK’s RT-PCR testing system, we risk over-estimating the COVID-19 incidence, the demand on track and trace and the extent of asymptomatic infection”. In that same report, the authors helpfully listed the lowest to highest false positive rate of dozens of tests using the same technology. The lowest value for false positive rate was 0.8%."

  • edbedb Posts: 65
    False +ves should have a different distribution too - scattered across the country proportional to tests not hotspots of cases.
    Should be pretty easy to tell the difference.
  • AlistairAlistair Posts: 23,670
    FTPT
    geoffw said:

    What if the virus isn't weaker, nor are our bodies better at coping with it as we've had some exposure. What if the low numbers need hospital compared to the huge spike in the cases is because of the false positives?

    Despite the somewhat breathless and dramatic style, Dr MikeYeadon is right about the potentially seriously misleading impression that false positives can give about the scale of infection. It seems to me that there are clearly issues here which should be debated openly in public. Kudos to Julia Hartley-Brewer for her efforts in that direction.

    Having got through the opening 700 paragraphs of his article and actually got to some numbers and assumptions I can immediatly spot the bollocks.

    He has assumed that the people getting pillar 2 tests are a completely random sampling of the population.

    He writes a lot of purple prose to try and justify this but that is unsupportable.

    The second point is if the case rate is driven by the alleged massive false postivie rate we should have seen an enourmas rise in cases for 2-17 year olds in Scotland when there was the massive post school start spike in tests for that age group.

    There was no massive spike in the case rate for that age group.

    Reality is once again staring the Covid sceptic straight in the eye and the Covid Sceptic does not blink.

  • HYUFD said:

    Scott_xP said:
    Hancock and Gove most supportive of another lockdown apparently, Patal and Sunak opposed
    Gove is probably most worried that a laissez faire policy will push support for Scottish independence even higher.
  • Yeadon:

    "What do we know about the false positive rate? Well, we do know that the Government’s own scientists were very concerned about it, and a report on this problem was sent to SAGE dated June 3rd 2020. I quote: “Unless we understand the operational false positive rate of the UK’s RT-PCR testing system, we risk over-estimating the COVID-19 incidence, the demand on track and trace and the extent of asymptomatic infection”.

    In that same report, the authors helpfully listed the lowest to highest false positive rate of dozens of tests using the same technology. The lowest value for false positive rate was 0.8%. "

    Doesn't explain why the positivity rate is going up consistently.

    And it means the positivity rates increase is even worse not better. It is the R, the increase in case percentages, that is the problem not the baseload of cases.
    He's not saying there aren't more genuine cases (if I have understood things), he's saying the high numbers we see all over our TV screens and which are driving public policy are wrong. 10x wrong.



    I hope he's not right because if it is then that makes the increase dramatically worse.

    Lets say that hypothetically there were 3000 cases but 90% were false, then there were just 300 real cases. 2700 false cases should remain the baseline.

    If suddenly ceteris paribus we have 5000 cases with 2700 being false then now we have 2300 real cases.

    If there aren't so many false positives then the caseload has gone up by just 66%.
    If there are so many false positives then the caseload has gone up by 666%.

    The worrying news isn't the number of cases it is the increase in cases. That is the problem and if your theory is right then the increase is more concerning not less.
    I don't understand your baseline, but maybe it is getting too late for my brain.

    How many were tested in your second case?
    I said ceteris paribus so the same number of people tested in both cases.

    In reality in recent weeks our number of tests conducted have gone up by about 33% while the number of positive cases has gone up by about 100%. That can't be explained by increased testing. Nor can the ONS survey data.

    The problem is tilting at the wrong argument. Nobody is that worried about the base amount of cases, had the number of cases remained flat then that would be fine, it is when the number of cases is going up potentially exponentially that it is worrying. The false positive argument doesn't explain why the positivity rate is increasing - and means the increase in positivity rate is worse not better than we think.
  • GallowgateGallowgate Posts: 19,072
    I’m old enough to remember everyone being told to go back to their offices.
  • So Eat Out To Help Out has been a bit of a waste then
  • BenpointerBenpointer Posts: 31,281

    Yeadon:

    "What do we know about the false positive rate? Well, we do know that the Government’s own scientists were very concerned about it, and a report on this problem was sent to SAGE dated June 3rd 2020. I quote: “Unless we understand the operational false positive rate of the UK’s RT-PCR testing system, we risk over-estimating the COVID-19 incidence, the demand on track and trace and the extent of asymptomatic infection”.

    In that same report, the authors helpfully listed the lowest to highest false positive rate of dozens of tests using the same technology. The lowest value for false positive rate was 0.8%. "

    Doesn't explain why the positivity rate is going up consistently.

    And it means the positivity rates increase is even worse not better. It is the R, the increase in case percentages, that is the problem not the baseload of cases.
    He's not saying there aren't more genuine cases (if I have understood things), he's saying the high numbers we see all over our TV screens and which are driving public policy are wrong. 10x wrong.



    I hope he's not right because if it is then that makes the increase dramatically worse.

    Lets say that hypothetically there were 3000 cases but 90% were false, then there were just 300 real cases. 2700 false cases should remain the baseline.

    If suddenly ceteris paribus we have 5000 cases with 2700 being false then now we have 2300 real cases.

    If there aren't so many false positives then the caseload has gone up by just 66%.
    If there are so many false positives then the caseload has gone up by 666%.

    The worrying news isn't the number of cases it is the increase in cases. That is the problem and if your theory is right then the increase is more concerning not less.
    I don't understand your baseline, but maybe it is getting too late for my brain.

    How many were tested in your second case?
    I said ceteris paribus so the same number of people tested in both cases.

    In reality in recent weeks our number of tests conducted have gone up by about 33% while the number of positive cases has gone up by about 100%. That can't be explained by increased testing. Nor can the ONS survey data.

    The problem is tilting at the wrong argument. Nobody is that worried about the base amount of cases, had the number of cases remained flat then that would be fine, it is when the number of cases is going up potentially exponentially that it is worrying. The false positive argument doesn't explain why the positivity rate is increasing - and means the increase in positivity rate is worse not better than we think.
    Maybe "all things being equal" would have been clearer for us plebs who slogged our way through stat schools? :wink:
  • geoffw said:

    Perhaps at least one of the journalists can ask them what the FPR (False Positive) rate is for the PCR test they are using in the 100,000s?

    The public needs to know the answer.
    fpt

    What if the virus isn't weaker, nor are our bodies better at coping with it as we've had some exposure. What if the low numbers need hospital compared to the huge spike in the cases is because of the false positives?

    Despite the somewhat breathless and dramatic style, Dr MikeYeadon is right about the potentially seriously misleading impression that false positives can give about the scale of infection. It seems to me that there are clearly issues here which should be debated openly in public. Kudos to Julia Hartley-Brewer for her efforts in that direction.


    Which would be great if the positivity rate were stable.

    It is not.

    It is the increasing positivity rate that is the worry.
  • HYUFDHYUFD Posts: 116,712
    edited September 2020

    HYUFD said:

    Scott_xP said:
    Hancock and Gove most supportive of another lockdown apparently, Patal and Sunak opposed
    Gove is probably most worried that a laissez faire policy will push support for Scottish independence even higher.
    I don't think anyone is arguing for laissez faire, after all Boris introduced the rule of 6 before Sturgeon, there is also the argument that if Sturgeon introduces another Scottish lockdown it might crash the Scottish economy while England if it avoids full lockdown while maintaining precautions returns to economic growth
  • https://twitter.com/rowlsmanthorpe/status/1307790171589533697

    The only thing that brought our cases down was a full lockdown.

    We're headed there again
  • BluestBlueBluestBlue Posts: 4,556
    edited September 2020
    IshmaelZ said:

    Quick. Someone has to let them know that some retired bloke who disagrees with the lockdown thinks it's almost certain that these people don't have the virus.
    But Retired Bloke's written an article! If that doesn't convince you that none of the cases are real and that none of the dead people are dead, then what will?
    The arithmetic isn't difficult. Say there are 60 million people in the country and one of them has the disease. You test all of them. If your test has an FPR of 0.8% (that is, is 99.2% accurate) you find 480,001 cases when actually there is 1. This is how false positives work, and is why when an initial mammogram detects breast cancer you conduct further tests rather than lopping the offending breast straight off. You do not make substantive decisions on the basis of the first test, which is what we are now doing.

    This is a problem. Rather than taking the perplexingly ad hominem point that he is retired, explain why it is not a problem please?
    Yes, I understand the problem - if you conduct a large number of tests on a population with few real cases of the infection, all you'll generate is a large number of false positives in proportion to the error rate of the test.

    As a consequence, the aim of the sceptics is to persuade us to take no lockdown measures - since most of the cases don't exist in their view, it's a useless and wasteful effort.

    My overall issue is that in the current circumstances, the burden of proof falls on them to explain why we should take the risk of repeating the experience of the spring, when both cases and deaths did rise dramatically (and the cases then were if anything greatly underestimated). What prevents that from happening again if they're wrong? Are the cases really rising in exact proportion to the increase in testing, which is what we would expect if their hypothesis is correct? Why are hospitalizations and deaths rising in other countries with rising cases rates? Are their tests more accurate? What has changed to prevent the spread of the virus now that normal social mixing has largely resumed?

    All this needs to be accounted for before we get blithely told to risk life and limb by people who are in many cases obviously working backwards from their own ideological preferences and wishful thinking.
  • dixiedeandixiedean Posts: 27,837
    The virus is back. What we about it now is a choice.
    But wishing it hadn't returned and then clutching at straws to prove one's desired situation just clouds the real issue.
    Which is what do we do about it.
  • HYUFDHYUFD Posts: 116,712
    Charles said:

    HYUFD said:
    If you are printing money then the size of the deficit is meaningless
    However the resultant rise in inflation assuming productivity does not rise at the same rate is not
  • IshmaelZIshmaelZ Posts: 21,830

    Yeadon:

    "What do we know about the false positive rate? Well, we do know that the Government’s own scientists were very concerned about it, and a report on this problem was sent to SAGE dated June 3rd 2020. I quote: “Unless we understand the operational false positive rate of the UK’s RT-PCR testing system, we risk over-estimating the COVID-19 incidence, the demand on track and trace and the extent of asymptomatic infection”.

    In that same report, the authors helpfully listed the lowest to highest false positive rate of dozens of tests using the same technology. The lowest value for false positive rate was 0.8%. "

    Doesn't explain why the positivity rate is going up consistently.

    And it means the positivity rates increase is even worse not better. It is the R, the increase in case percentages, that is the problem not the baseload of cases.
    He's not saying there aren't more genuine cases (if I have understood things), he's saying the high numbers we see all over our TV screens and which are driving public policy are wrong. 10x wrong.



    I hope he's not right because if it is then that makes the increase dramatically worse.

    Lets say that hypothetically there were 3000 cases but 90% were false, then there were just 300 real cases. 2700 false cases should remain the baseline.

    If suddenly ceteris paribus we have 5000 cases with 2700 being false then now we have 2300 real cases.

    If there aren't so many false positives then the caseload has gone up by just 66%.
    If there are so many false positives then the caseload has gone up by 666%.

    The worrying news isn't the number of cases it is the increase in cases. That is the problem and if your theory is right then the increase is more concerning not less.
    I don't understand your baseline, but maybe it is getting too late for my brain.

    How many were tested in your second case?
    He is sort of right but he has misstated it (of his 5,000 90% or 4,500 should be false positives).

    Say 100 tests a day, fpr 10%.

    Day 1 100 tests - 20 positives = 10 true cases
    Day 2 100 tests - 30 positives = 20 true cases.

    So the rise is *worse* than the rise in positives makes it look - your rate has risen by 100% from 10 to 20, not 50% 20 to 30.
  • HYUFDHYUFD Posts: 116,712
    dixiedean said:

    The virus is back. What we about it now is a choice.
    But wishing it hadn't returned and then clutching at straws to prove one's desired situation just clouds the real issue.
    Which is what do we do about it.

    The virus has never gone away, we have never had 0 cases and it won't until we get a vaccine at least
  • GallowgateGallowgate Posts: 19,072
    Lockdown again is a waste of time. Compliance will be nowhere near what it was in the last one. That much is obvious from one weekend of the North East “lockdown”. Everyone is just getting on with it, ignoring the rules but being careful.
  • BenpointerBenpointer Posts: 31,281
    Spare a though for OGH. Interesting thread header about the prospects for the SC nomination...

    Comments dominated by discussion of testing stats, false positives, Bayes' Theorem... and some Retired Bloke.
  • geoffwgeoffw Posts: 8,092

    Sigh. Someone discovered Bayes' Theorem and thinks that none of the scientists advising the government understand it.

    If nothing else, trust the ONS survey.

    --AS

    Well yes we should assume that the scientists understand it (though it is not common to find them well trained in probability theory), but the issue is the prominence given in public discussions to the estimated number of positives, which amounts to a kind of scaremongering.

  • PulpstarPulpstar Posts: 75,842
    False positives will be a constant rate, so if the rate is increasing it isnt due to false positives
  • geoffwgeoffw Posts: 8,092

    Yeadon:

    "What do we know about the false positive rate? Well, we do know that the Government’s own scientists were very concerned about it, and a report on this problem was sent to SAGE dated June 3rd 2020. I quote: “Unless we understand the operational false positive rate of the UK’s RT-PCR testing system, we risk over-estimating the COVID-19 incidence, the demand on track and trace and the extent of asymptomatic infection”.

    In that same report, the authors helpfully listed the lowest to highest false positive rate of dozens of tests using the same technology. The lowest value for false positive rate was 0.8%. "

    Doesn't explain why the positivity rate is going up consistently.

    And it means the positivity rates increase is even worse not better. It is the R, the increase in case percentages, that is the problem not the baseload of cases.
    He's not saying there aren't more genuine cases (if I have understood things), he's saying the high numbers we see all over our TV screens and which are driving public policy are wrong. 10x wrong.



    I hope he's not right because if it is then that makes the increase dramatically worse.

    Lets say that hypothetically there were 3000 cases but 90% were false, then there were just 300 real cases. 2700 false cases should remain the baseline.

    If suddenly ceteris paribus we have 5000 cases with 2700 being false then now we have 2300 real cases.

    If there aren't so many false positives then the caseload has gone up by just 66%.
    If there are so many false positives then the caseload has gone up by 666%.

    The worrying news isn't the number of cases it is the increase in cases. That is the problem and if your theory is right then the increase is more concerning not less.
    I don't understand your baseline, but maybe it is getting too late for my brain.

    How many were tested in your second case?
    I said ceteris paribus so the same number of people tested in both cases.

    In reality in recent weeks our number of tests conducted have gone up by about 33% while the number of positive cases has gone up by about 100%. That can't be explained by increased testing. Nor can the ONS survey data.

    The problem is tilting at the wrong argument. Nobody is that worried about the base amount of cases, had the number of cases remained flat then that would be fine, it is when the number of cases is going up potentially exponentially that it is worrying. The false positive argument doesn't explain why the positivity rate is increasing - and means the increase in positivity rate is worse not better than we think.
    Maybe "all things being equal" would have been clearer for us plebs who slogged our way through stat schools? :wink:
    He meant pari passu.

  • Spare a though for OGH. Interesting thread header about the prospects for the SC nomination...

    Comments dominated by discussion of testing stats, false positives, Bayes' Theorem... and some Retired Bloke.

    I suspect OGH by now is probably used to below the line conversation completely ignoring the thread header.

    There is something rather sweetly innocent about the site having an off topic button to press on the below the line posts.
  • geoffw said:

    Sigh. Someone discovered Bayes' Theorem and thinks that none of the scientists advising the government understand it.

    If nothing else, trust the ONS survey.

    --AS

    Well yes we should assume that the scientists understand it (though it is not common to find them well trained in probability theory), but the issue is the prominence given in public discussions to the estimated number of positives, which amounts to a kind of scaremongering.

    I know what you mean, but they wouldn't publish numbers that they knew to be misleading. (Or at least one of the SAGE scientists would be sounding off about it in the media already.) As someone pointed out above, the Bayes' calculation of 90%ish of all positives being false is simply wrong when the people being tested aren't a random sample of the population. It's a misunderstanding which is getting too much attention. IMO this false positive thing is a distraction and does not materially affect the numbers.

    --AS
  • I wonder if in the North theres a "fuck the Tories" attitude to not following the rules
  • I wonder if in the North theres a "fuck the Tories" attitude to not following the rules

    Unlikely. Especially since the North voted Tory only last year.
  • I wonder if in the North theres a "fuck the Tories" attitude to not following the rules

    Unlikely. Especially since the North voted Tory only last year.
    Things change. Cummings can't have helped.
  • I wonder if in the North theres a "fuck the Tories" attitude to not following the rules

    Unlikely. Especially since the North voted Tory only last year.
    Things change. Cummings can't have helped.
    Most people I know are more worried about giving the virus to their relatives and loved ones than Cummings.
  • GallowgateGallowgate Posts: 19,072

    I wonder if in the North theres a "fuck the Tories" attitude to not following the rules

    Unlikely. Especially since the North voted Tory only last year.
    There’s plenty of Labour constituencies around me.
  • If they are talking about 6 month lockdowns, they obviously don't think a vaccine is coming anytime soon.
  • Can those people who want a lockdown say what level of covid infections and deaths they are prepared to accept ?
  • RobDRobD Posts: 58,941

    I wonder if in the North theres a "fuck the Tories" attitude to not following the rules

    That'd be a real stupid reason to do so.
  • JonathanJonathan Posts: 20,901

    Lockdown again is a waste of time. Compliance will be nowhere near what it was in the last one. That much is obvious from one weekend of the North East “lockdown”. Everyone is just getting on with it, ignoring the rules but being careful.

    If the hospitals start to fill and we sick people in corridors, we will lockdown and people will obey it.
  • The media must be held to blame for the idea it would all be over by Christmas
  • BluestBlueBluestBlue Posts: 4,556

    I wonder if in the North theres a "fuck the Tories" attitude to not following the rules

    Unlikely. Especially since the North voted Tory only last year.
    Things change. Cummings can't have helped.
    That must be why the lockdown failed to work after the story broke.

    Oh wait, the exact opposite happened, despite all the predictions on here.
  • Jonathan said:

    Lockdown again is a waste of time. Compliance will be nowhere near what it was in the last one. That much is obvious from one weekend of the North East “lockdown”. Everyone is just getting on with it, ignoring the rules but being careful.

    If the hospitals start to fill and we sick people in corridors, we will lockdown and people will obey it.
    And its too late by then.
  • Those Republican Senators who won't support an appointment before November 3rd are not likely to support a Trump nominee afterwards if Trump loses.

    On the other hand if Trump wins, why rush?

    So yes, I think the appointment takes place after inauguration, whatever the election outcome.

    The only problem there is that the Senate numbers will change.

    So it might be impossible to get a conservative judge even if Trump did win.

    Of course almost any judge would be more conservative than RBG was but I suspect some Republicans are hoping for someone fully to their liking.
  • GallowgateGallowgate Posts: 19,072

    I wonder if in the North theres a "fuck the Tories" attitude to not following the rules

    Unlikely. Especially since the North voted Tory only last year.
    Things change. Cummings can't have helped.
    That must be why the lockdown failed to work after the story broke.

    Oh wait, the exact opposite happened, despite all the predictions on here.
    Deny it all you want but the Cummings story did cut through, and it very much seemed to be the trigger point for people starting to more liberally interpret the rules.
  • kinabalukinabalu Posts: 38,851

    Scott_xP said:
    Only six months.

    I don't believe them. If there is no working and mass available vaccine by next spring then we go through the cycle again.
    I believe them, we'll have a mass available vaccine by then.
    June 2021 for that. Not so long really.
  • I wonder if in the North theres a "fuck the Tories" attitude to not following the rules

    Unlikely. Especially since the North voted Tory only last year.
    There’s plenty of Labour constituencies around me.
    Not as many as there were.
  • I wonder if in the North theres a "fuck the Tories" attitude to not following the rules

    Unlikely. Especially since the North voted Tory only last year.
    Things change. Cummings can't have helped.
    Most people I know are more worried about giving the virus to their relatives and loved ones than Cummings.
    The government have really missed the messaging getting bogged down with grass up your neighbour and there could be a big fine if you are naughty.

    The messaging should be focused on do this otherwise your loved ones will suffer horribly and might well die...insert videos of people struggling to breath, crying relatives and recovered people who still look like a ghost.
  • Pulpstar said:

    False positives will be a constant rate, so if the rate is increasing it isnt due to false positives

    I wonder if false positives might be affected by previous exposure ie if someone was infected previously but still had residual amounts of virus in their system when they were tested weeks or months later.

    I know someone who this might apply to.
  • OmniumOmnium Posts: 9,688

    Can those people who want a lockdown say what level of covid infections and deaths they are prepared to accept ?

    Sort of interesting as to how bad it would have to get to convince those that don't want a lockdown too.

    Hospitals failing, unburied dead. I'm not at all sure that stomaching that sort of thing isn't the better choice, but I can see why no politician would allow it, and if I was a politician I'd run away from such a choice too.

  • I presume the great bog roll hoarding will start again this week.
  • dixiedeandixiedean Posts: 27,837

    Lockdown again is a waste of time. Compliance will be nowhere near what it was in the last one. That much is obvious from one weekend of the North East “lockdown”. Everyone is just getting on with it, ignoring the rules but being careful.

    My impression is that those who follow the rules are following them. Those that aren't aren't.
    The followers are hurting the economy, the non-followers are spreading the disease just as before.
    So the only effect really has been to keep the compliant away from spending any money.
    If they want to stop the disease then it is those who won't comply who need dealing with.
    As I said that is a choice.
  • If the cases don't increase exponentially, then that's fine. But they are again
  • dixiedeandixiedean Posts: 27,837

    I wonder if in the North theres a "fuck the Tories" attitude to not following the rules

    Unlikely. Especially since the North voted Tory only last year.
    It did? En masse?
  • BluestBlueBluestBlue Posts: 4,556
    edited September 2020

    I wonder if in the North theres a "fuck the Tories" attitude to not following the rules

    Unlikely. Especially since the North voted Tory only last year.
    Things change. Cummings can't have helped.
    That must be why the lockdown failed to work after the story broke.

    Oh wait, the exact opposite happened, despite all the predictions on here.
    Deny it all you want but the Cummings story did cut through, and it very much seemed to be the trigger point for people starting to more liberally interpret the rules.
    Did infections fall drastically during the lockdown after the Cummings story broke? I think you know they did - I've still got Malmesbury's charts to prove it - so why try to maintain this bollocks narrative? If people are being more liberal now, it's because we're more than 6 months into this thing, and naturally sick of it.

    Ironically, given the rise of the sceptic tendency, if Cummings did the same thing again today he'd be treated like a hero by them!
  • GallowgateGallowgate Posts: 19,072

    I wonder if in the North theres a "fuck the Tories" attitude to not following the rules

    Unlikely. Especially since the North voted Tory only last year.
    There’s plenty of Labour constituencies around me.
    Not as many as there were.
    So what? “The North voted Tory” is clearly bollocks.
This discussion has been closed.