Coronavirus (COVID-19)

[Todd_is_God]

Just now, Paco said:

I’m always open to considering alternative viewpoints, and an openness about PCR testing would be very welcome but Julia Hartley-Brewer screaming about 91% OF CASES ARE FALSE POSITIVES six months into a pandemic that’s killed nearly a million people can be safely fired into the sea.

That's not at all what she is saying, though

[parsforlife]

 

2 minutes ago, Todd_is_God said:

Unless maths is an opinion, it doesn't matter who says it.

It does if it’s Granny Danger.

[Paco]

That's not at all what she is saying, though

Well, it is.

[Marshmallo]

She's saying there that 80% of positive tests are false going by that? She's had a few before posting?

[Todd_is_God]

12 minutes ago, Paco said:

 

Well, it is.

 

No it isn't.

If it is, for example, 0.8%, and 1% of tests on any given day come back positive, then saying that 80% of the "cases" from that day are not actually infectious is neither hysterical nor false, and it still indicates 20% are not false positives.

On a day where 2,900 positive results are returned from 175,000 processed tests (roughly where we are just now), it still leaves 1,500 true positive results.

It also doesn't take away from what has gone before.

We wouldn't be having this conversation if Hancock just told us what it was. It's the need for secrecy, smoke, mirrors and bullshit that is most annoying.

Edited September 18, 2020 by Todd_is_God

[madwullie]

Also suggests the virus is more deadly than previously calculated  

Edited September 18, 2020 by madwullie

[Todd_is_God]

Just now, madwullie said:

Also suggests the virus is more deadly than previously calculated  

Yes, true. However providing you also factor in the hundreds of thousands (if not millions) of infections that were not recorded in Jan - March especially it is still not that deadly.

[JMDP]

@pandarilla and @JTS98 will be along in a minute to tell everyone to engage with clearly good faith actors such as Julia Hartley-Brewer, TalkRadio and @Todd_is_God

Edited September 18, 2020 by JMDP

[Michael W]

I can only conclude here that JHB has been sniffing glue. That tweet is a bawhair away from punting conspiracy theories. 

Edited September 18, 2020 by Michael W

[Sherrif John Bunnell]

SCAMDEMIC!

[Todd_is_God]

1 minute ago, Michael W said:

I can only conclude here that JHB has been sniffing glue. That tweet is a bawhair away from punting conspiracy theories. 

Which part of her explanation of what a FPR of 0.8% means in real terms do you disagree with?

I'm not having a go or defending anybody, but which part of the maths of what she says there are you disputing?

[djchapsticks]

58 minutes ago, Ron Aldo said:

We had the longest lockdown in Europe which logic dictates should have bought us the biggest window to develop our testing system.

As usual, the government have fucked it and it looks like Joe Public will be made to suffer the consequences in the form of yet another lockdown.

Add to the fact, we've now seen that the medium term benefit of a full lockdown followed by easing of restrictions is negligible.

All that another lockdown is going to do is kick the can down the road a few more months with more uncertainty at the end. The only certainty is that it will harm people mentally, physically and financially now with no long term benefit.

The very real impact of a fucking hapless shambles of a UK government who have fucked up every step from testing, to the 'world beating' app, to things as simple as instructions of what people can and cannot do.

Now at a stage where we've tried lockdown and there were no lasting long term benefits, the best option is to shield the vulnerable and let others work around it.

[Michael W]

9 minutes ago, Todd_is_God said:

Which part of her explanation of what a FPR of 0.8% means in real terms do you disagree with?

I'm not having a go or defending anybody, but which part of the maths of what she says there are you disputing?

I have issue with the 91% of "covid cases" (quote marks her own) are false positives. 

Now, I'm not a mathematician. But the tweet to me looks like she's put a string of things together and garbled the message to support a strange claim about how many people actually have the virus. It's an astonishing attempt to downplay it, actually

If she'd left it that 0.8% of cases undertaken may be false positives then I'd have taken no issue. That seems plausible tbh. 

Edited September 18, 2020 by Michael W

[dirty dingus]

If there was so many false positives it would have came out donkeys ago. Remember the antibody test getting poo-pooed in the media because it was something like 20% unreliable.

[Todd_is_God]

8 minutes ago, Michael W said:

I have issue with the 91% of "covid cases" (quote marks her own) are false positives. 

Now, I'm not a mathematician. But the tweet to me looks like she's put a string of things together and garbled the message to support a strange claim about how many people actually have the virus. It's an astonishing attempt to downplay it, actually

If she'd left it that 0.8% of cases undertaken may be false positives then I'd have taken no issue. That seems plausible tbh. 

But 0.8% of positives being false positives is not the same as 0.8% of tests returning false positives.

They are wildly different.

I don't know what England's daily positive % is (I don't think they publish it in the same way the SG do), but if it was 0.88% on a given date, then 91% of all positive results returned that particular day could well be false positives.

You don't need to be a mathematician to complete what is a quite simple calculation.

Edited September 18, 2020 by Todd_is_God

[Melanius Mullarkey]

10 minutes ago, Michael W said:

But the tweet to me looks like she's put a string of things together and garbled the message to support a strange claim about how many people actually have the virus.

[Marshmallo]

6 minutes ago, Todd_is_God said:

But 0.8% of positives being false positives is not the same as 0.8% of tests returning false positives.

They are wildly different.

I don't know what England's daily positive % is (I don't think they publish it in the same way the SG do), but if it was 0.88% on a given date, then 91% of all positive results returned that particular day could well be false positives.

You don't need to be a mathematician to complete what is a quite simple calculation.

You and her are talking percentage points, rather than percentages.

[Todd_is_God]

5 minutes ago, Marshmallo said:

You and her are talking percentage points, rather than percentages.

Ok, i'm (genuinely) listening.

If we process 175,000 tests on a given day, using a test which is 99.2% specific (0.8% FPR), how many false positives would we expect to be returned from those tests?

Edited September 18, 2020 by Todd_is_God

[Bairnardo]

21 minutes ago, djchapsticks said:

Add to the fact, we've now seen that the medium term benefit of a full lockdown followed by easing of restrictions is negligible.

All that another lockdown is going to do is kick the can down the road a few more months with more uncertainty at the end. The only certainty is that it will harm people mentally, physically and financially now with no long term benefit.

The very real impact of a fucking hapless shambles of a UK government who have fucked up every step from testing, to the 'world beating' app, to things as simple as instructions of what people can and cannot do.

Now at a stage where we've tried lockdown and there were no lasting long term benefits, the best option is to shield the vulnerable and let others work around it.

Genuinely think that's what they mean to do despite the term "lockdown" being banded about. Probably just don't want to piss of the older Conservative population by coming out now and saying we are going to bring back shielding. The cost to the rest of us will be a lazily (probably not at all) enforced ban on indoor socialising with other households. Any establishment which collects VAT will remain open.

[Shipa]

9 minutes ago, Todd_is_God said:

But 0.8% of positives being false positives is not the same as 0.8% of tests returning false positives.

They are wildly different.

I don't know what England's daily positive % is (I don't think they publish it in the same way the SG do), but if it was 0.88% on a given date, then 91% of all positive results returned that particular day could well be false positives.

You don't need to be a mathematician to complete what is a quite simple calculation.

While it may be the case that 91% of positive results are false positives, and that may be what she meant, it is not what she said. 0% of "covid cases" are false positives, if they were they would be false positives and not covid cases.