In my previous stack I returned to the question of how the ‘efficacy’ of a vaccine can be manipulated by either (a) ignoring some data or (b) mis-categorizing some data. This data diddling has been pointed out and examined by many others so I wasn’t pointing out anything new, but I have been trying to approach this from an algebraic perspective to see what insights we could gain, if any. In particular, I wanted to see what impact this diddling might have; would it be a large or small impact and could that be demonstrated algebraically?
I was trying to figure out why, in Pfizer's study of children 6 months-4 years, their vaccinated group was about double the size of their placebo group (prior to unblinding when if I remember correctly 100% of the placebo group was vaccinated). And then they based efficacy on children who contracted Covid 7 days or more after Dose 3. Is this similar data diddling as you discuss above? Obviously, for those of us who live in the real world (and if we were actually dealing with a disease that seriously harmed children), what matters isn't the likelihood that our children will become ill more than 3 months after their first vaccination in the series but the likelihood our children will become ill at any point after that first shot. (Including the period post-unblinding when the placebo group went bye-bye.)
That "trial" is an absolute abomination. A total disgrace. That the government 'experts' passed the vax for kids as young as 6 months based on this "evidence" speaks volumes. I've never seen anything like it in my life - it bears the same relation to science as does Ghislaine Maxwell to child welfare.
I have no words of condemnation strong enough for the corruption that is plain to see here. There's no 'emergency' in ANY conceivable sense of that word for young kids anyway - and it just gets worse (so much worse) from there on in.
The people who passed this are contemptible assholes, in my view, who have betrayed kids, betrayed science, and betrayed their country. They should be in jail. In any just and fair and rational world they would be.
I can’t check this...out of it for too long. But if I put my tinfoil hat on...how’s the ‘new math’ working out in our public schools? Intentional? “Trust the experts”!!!
I'm no math genius, but your numbers appear to be correct. However, the one constant I'm noticing throughout all of the conclusions arrived at by the experts is that the methods used do not pass the common sense test.
Whether it's,
Using different participant numbers in test vs control groups,
Not using similar demographic populations for test vs control groups,
Factoring for a delay between injection and vaccinated, but not isolating those subjects from both vaccinated and unvaccinated.
And then when the jig is up insisting that the failed glorious goo somehow has therapeutic effects that are unproven, and unprovable considering that those infected, gooed or not, have a 99.6% survival rate, as well as, experience symptoms no more serious than the seasonal flu.
Very nice! I haven't spotted any errors in the squiggles.
Why would you have to assume that deaths are a small percentage of the populations? As far as I can see, you only need that r(u) and r(v) are invariant under changes in p or n (sorry, I should have written "n or p"...), i.e., discreteness issues.
You're probably right - but when I first wrote everything out in completeness (reducing n(v) by d(T), for example) it was a mess and I didn't think about this invariance. I'll have a check and see what happens - it would tighten up the result a wee bit.
Probably why the pfarmaceutical cos end all of their super short term tests by immediately jabbing the entire control group. You know, following TheScience
My background is theoretical physics and I don't tend to understand things more fully until I've been able to liberally sprinkle a few squiggles about. I'm not all that good with data analysis proper. I've been trying to get to a simple model where some of the dangers of seemingly 'innocent' data fudges could be made really obvious with regards to a determination of efficacy.
This simple 1 jab done on day 1 model, coupled with the goo factor approach (which arose because of the data from Alberta) has ended up with a result that surprised me - not least in just how significant the difference between the true efficacy and the manipulated efficacy can be.
If you do have some thoughts on how the Bilious Bratwurst of Bayesian Buggeration was actually manufactured I'd appreciate those.
I have a mental heuristic that has always served me well: in the absence of overwhelming evidence, assume you are not the first person to have any idea.
I suspect I'm not even the first person to have used this kind of algebraic approach either. What is important is that we all collectively keep developing our insights and approaches - and keep trying to get the word out as effectively as we can. A lot of people have woken up to the failure of the vaccines now, but I don't think a lot of people know just how flawed the 'expert' position is.
It's not my intention to diminish the importance of your statement. I was referring to the near-certainty that those establishing the analytics they'd use for reporting knew perfectly well what the math meant.
This is one of the things I simply can't get my head round. I go round in endless circles of frustration and self-doubt. OK maybe not everyone is as weird as me and actually *likes* all the algebra, but you don't need to be that nerdy to appreciate that ignoring things until 14 days have elapsed, or worse putting these stats into the wrong population is going to mess things up.
It doesn't require a PhD in stats and years of experience to see there are potentially big problems here - and yet here we are, official government departments, supposedly staffed by 'experts', seem to keep doing dumb things when analysing the data.
So I oscillate between - I must be being really stupid and missing something, and - these people can't be that stupid, they must know what they're doing isn't right.
My nice cosy pre-pandemic view of the world has been shattered and I'm still trying to put all the pieces back together into something that makes any sort of sense.
yeah - the bad cat has done some magnificent work - truly inspiring stuff. There's a lot of overlap between the algebraic approach I've adopted and EGM's work. Lots of people have also pointed this datacrime out (Fenton and colleagues, to name just one example).
At the end of the day we need everyone to keep thinking and developing new ideas and approaches to understanding the fraud that we've been subjected to. We need to say, look, here are 50 different ways to analyse this - and they all come to the same conclusion - you're calculating efficacy wrongly.
You can rewrite the fraction (the beast behind the 1-) as follows:
numerator:
(1-p)(1-g)(1-e)
denominator:
(1-p(1-g(1-e)))
So, numerator is symmetrical, denominator is kind of nested. In other words: beware the brackets!
On the one hand, this is just the otherworldly nonsense mathematicians are fond of. On the other hand, if your intention is to drive up the diddle efficacy, it might say something about which parameter to work on.
You don’t even need to be injecting anyone with anything to commit this Bayesian data fraud. Although there’ll be a more pronounced effect if you are injecting a LNP laden, TLR deregulating, spike protein manufacturing serum. And in answer to your question as to whether this is intentional obfuscation, the answer is yes. See Pfizer docs.
You're right - this kind of data manipulation causes problems even when you just split the population into 2 groups vaxxed and unvaxxed and don't do anything other than that.
I'm 100% certain the Pharma companies are going to do their level best to 'fudge' things to make their products look better. They can hardly be relied upon to be the most objective assessors in the game. They're only really interested in profit and if what they do also happens to be beneficial to people then that's a plus - as long as they can 'show' the benefit (even if none really exists) they're good to go. I don't trust their data one bit.
But when official government agencies start calculating "efficacy" based on these kinds of data diddles it needs to be shouted from the rootops!
No; even if the vaccine is a harmless placebo (nobody dying from the vaccine, and e = 0), the diddle efficacy is g / ( 1 - p ( 1 - g ) ), and you can get arbitrarily close to 100% by boosting g (replace 14 days by a longer time period), or by boosting p (vaccinate everybody).
I'm sure this 'misuse' of algebra is now considered a hate crime in Canada.
https://en.m.wikipedia.org/wiki/Omar_Alghabra
I was trying to figure out why, in Pfizer's study of children 6 months-4 years, their vaccinated group was about double the size of their placebo group (prior to unblinding when if I remember correctly 100% of the placebo group was vaccinated). And then they based efficacy on children who contracted Covid 7 days or more after Dose 3. Is this similar data diddling as you discuss above? Obviously, for those of us who live in the real world (and if we were actually dealing with a disease that seriously harmed children), what matters isn't the likelihood that our children will become ill more than 3 months after their first vaccination in the series but the likelihood our children will become ill at any point after that first shot. (Including the period post-unblinding when the placebo group went bye-bye.)
That "trial" is an absolute abomination. A total disgrace. That the government 'experts' passed the vax for kids as young as 6 months based on this "evidence" speaks volumes. I've never seen anything like it in my life - it bears the same relation to science as does Ghislaine Maxwell to child welfare.
I have no words of condemnation strong enough for the corruption that is plain to see here. There's no 'emergency' in ANY conceivable sense of that word for young kids anyway - and it just gets worse (so much worse) from there on in.
The people who passed this are contemptible assholes, in my view, who have betrayed kids, betrayed science, and betrayed their country. They should be in jail. In any just and fair and rational world they would be.
I can’t check this...out of it for too long. But if I put my tinfoil hat on...how’s the ‘new math’ working out in our public schools? Intentional? “Trust the experts”!!!
I'm no math genius, but your numbers appear to be correct. However, the one constant I'm noticing throughout all of the conclusions arrived at by the experts is that the methods used do not pass the common sense test.
Whether it's,
Using different participant numbers in test vs control groups,
Not using similar demographic populations for test vs control groups,
Factoring for a delay between injection and vaccinated, but not isolating those subjects from both vaccinated and unvaccinated.
And then when the jig is up insisting that the failed glorious goo somehow has therapeutic effects that are unproven, and unprovable considering that those infected, gooed or not, have a 99.6% survival rate, as well as, experience symptoms no more serious than the seasonal flu.
There has certainly been a lot of very suspicious things going on with regard to the data and the analysis of it.
Very nice! I haven't spotted any errors in the squiggles.
Why would you have to assume that deaths are a small percentage of the populations? As far as I can see, you only need that r(u) and r(v) are invariant under changes in p or n (sorry, I should have written "n or p"...), i.e., discreteness issues.
You're probably right - but when I first wrote everything out in completeness (reducing n(v) by d(T), for example) it was a mess and I didn't think about this invariance. I'll have a check and see what happens - it would tighten up the result a wee bit.
Probably why the pfarmaceutical cos end all of their super short term tests by immediately jabbing the entire control group. You know, following TheScience
While I don't think that these diddles were precisely how the sausage was made, this is an important demonstration of the results of diddling.
Thanks for that - much appreciated.
My background is theoretical physics and I don't tend to understand things more fully until I've been able to liberally sprinkle a few squiggles about. I'm not all that good with data analysis proper. I've been trying to get to a simple model where some of the dangers of seemingly 'innocent' data fudges could be made really obvious with regards to a determination of efficacy.
This simple 1 jab done on day 1 model, coupled with the goo factor approach (which arose because of the data from Alberta) has ended up with a result that surprised me - not least in just how significant the difference between the true efficacy and the manipulated efficacy can be.
If you do have some thoughts on how the Bilious Bratwurst of Bayesian Buggeration was actually manufactured I'd appreciate those.
Wow, that makes sense! Good find in the recesses of your brain. 👍🏽👏
I have a mental heuristic that has always served me well: in the absence of overwhelming evidence, assume you are not the first person to have any idea.
lol
that's a good heuristic.
I suspect I'm not even the first person to have used this kind of algebraic approach either. What is important is that we all collectively keep developing our insights and approaches - and keep trying to get the word out as effectively as we can. A lot of people have woken up to the failure of the vaccines now, but I don't think a lot of people know just how flawed the 'expert' position is.
It's not my intention to diminish the importance of your statement. I was referring to the near-certainty that those establishing the analytics they'd use for reporting knew perfectly well what the math meant.
This is one of the things I simply can't get my head round. I go round in endless circles of frustration and self-doubt. OK maybe not everyone is as weird as me and actually *likes* all the algebra, but you don't need to be that nerdy to appreciate that ignoring things until 14 days have elapsed, or worse putting these stats into the wrong population is going to mess things up.
It doesn't require a PhD in stats and years of experience to see there are potentially big problems here - and yet here we are, official government departments, supposedly staffed by 'experts', seem to keep doing dumb things when analysing the data.
So I oscillate between - I must be being really stupid and missing something, and - these people can't be that stupid, they must know what they're doing isn't right.
My nice cosy pre-pandemic view of the world has been shattered and I'm still trying to put all the pieces back together into something that makes any sort of sense.
It's corruption, hon. You're not dumb or naive, you're just not corrupt.
Along the same lines https://boriquagato.substack.com/p/bayesian-datacrime-defining-vaccine
yeah - the bad cat has done some magnificent work - truly inspiring stuff. There's a lot of overlap between the algebraic approach I've adopted and EGM's work. Lots of people have also pointed this datacrime out (Fenton and colleagues, to name just one example).
At the end of the day we need everyone to keep thinking and developing new ideas and approaches to understanding the fraud that we've been subjected to. We need to say, look, here are 50 different ways to analyse this - and they all come to the same conclusion - you're calculating efficacy wrongly.
absolutely agree, just adding material for the keen reader :) Thanks for your work Rudolph.
You can rewrite the fraction (the beast behind the 1-) as follows:
numerator:
(1-p)(1-g)(1-e)
denominator:
(1-p(1-g(1-e)))
So, numerator is symmetrical, denominator is kind of nested. In other words: beware the brackets!
On the one hand, this is just the otherworldly nonsense mathematicians are fond of. On the other hand, if your intention is to drive up the diddle efficacy, it might say something about which parameter to work on.
awesome - thanks. I'll have a play and see if I spot anything else of interest.
You don’t even need to be injecting anyone with anything to commit this Bayesian data fraud. Although there’ll be a more pronounced effect if you are injecting a LNP laden, TLR deregulating, spike protein manufacturing serum. And in answer to your question as to whether this is intentional obfuscation, the answer is yes. See Pfizer docs.
You're right - this kind of data manipulation causes problems even when you just split the population into 2 groups vaxxed and unvaxxed and don't do anything other than that.
I'm 100% certain the Pharma companies are going to do their level best to 'fudge' things to make their products look better. They can hardly be relied upon to be the most objective assessors in the game. They're only really interested in profit and if what they do also happens to be beneficial to people then that's a plus - as long as they can 'show' the benefit (even if none really exists) they're good to go. I don't trust their data one bit.
But when official government agencies start calculating "efficacy" based on these kinds of data diddles it needs to be shouted from the rootops!
No; even if the vaccine is a harmless placebo (nobody dying from the vaccine, and e = 0), the diddle efficacy is g / ( 1 - p ( 1 - g ) ), and you can get arbitrarily close to 100% by boosting g (replace 14 days by a longer time period), or by boosting p (vaccinate everybody).