You're getting confused between the usefulness of aggregate in establishing a stable mean output (which is true) and the usefulness of aggregate in supporting unverified postulates. The latter is a fallacy called "bandwagon appeal": there are lots of people who say this thing is true, without proving it, and therefore, it must be true.
I'm going to open a website that collects reports of dangerous interactions with Q people. I will not verify anything. But police MUST respond by uploading ALL police encounters with suspected Q people, just in case they're dangerous. Also, any random citizen is also allowed to upload any encounters they believe they had with crazy Q people.
My database has 1,000,000 reports by next month of psychotic Q people causing mischief and violence.
Based on your argument, I can then confidently state that there is an epidemic of Q-related violence in this country, because my Q-violence tip line website has a lot of reports, and according to the law of big numbers, that must mean that a significant number of those reports must be based on real life.
I'm just going to assume that only 5% of these 1,000,000 reports are worthless. That's a pretty big problem.
Except, of course, that I haven't proven that ANY Q people are capable of violence. And even if it seems logical that there must be SOME unhinged Q people out there, is there a basis for me to assume that even the vast majority of my tip line website is based off of reality? Does the fact that I have a lot of unverified reports prove that these reports reflect reality?
Or is it just the bandwagon appeal fallacy?
In this case, I have an enormous aggregate of reports suggesting Q people are violently crazy. BUT, because I am SPECIFICALLY ASKING FOR THOSE STORIES, and because my website VERIFIES NOTHING, then my website's aggregate conclusion can't be empirically supported, because I have no empirical evidence that even a single Q person was verified to have committed violence via my website. Just lots of unverified reports.
Exactly the same with VAERS.
You're getting confused between the usefulness of aggregate in establishing a stable mean output (which is true) and the usefulness of aggregate in supporting unverified postulates. The latter is a fallacy called "bandwagon appeal": there are lots of people who say this thing is true, without proving it, and therefore, it must be true.
I'm going to open a website that collects reports of dangerous interactions with Q people. I will not verify anything. But police MUST respond by uploading ALL police encounters with suspected Q people, just in case they're dangerous. Also, any random citizen is also allowed to upload any encounters they believe they had with crazy Q people.
My database has 1,000,000 reports by next month of psychotic Q people causing mischief and violence.
Based on your argument, I can then confidently state that there is an epidemic of Q-related violence in this country, because my Q-violence tip line website has a lot of reports, and according to the law of big numbers, that must mean that a significant number of those reports must be based on real life.
I'm just going to assume that only 5% of these 1,000,000 reports are worthless. That's a pretty big problem.
Except, of course, that I haven't proven that ANY Q people are capable of violence. And even if it seems logical that there must be SOME unhinged Q people out there, is there a basis for me to assume that even the vast majority of my tip line website is based off of reality? Does the fact that I have a lot of unverified reports prove that these reports reflect reality?
Or is it just the bandwagon appeal fallacy?
In this case, I have an enormous aggregate of reports suggesting Q people are violently crazy. BUT, because I am SPECIFICALLY ASKING FOR THOSE STORIES, and because my website VERIFIES NOTHING, then my website's aggregate conclusion can't be empirically supported, because I have no empirical evidence that even a single Q person was verified to have committed violence via my website.
Exactly the same with VAERS.