#1 ·
I'm honestly not quite sure where this fits in, and I see there are already a few posts hitting this same note, but what I want to bring up is pretty specific and hasn't really been mentioned yet, which is kind of a bummer.
Over on the Politics board, we’ve seen plenty of attempts at statistical analysis or people trying to interpret the raw numbers we've been getting for this subject so far.
Of course, a little mistake here and there always manages to sneak in, doesn't it?
The real issue I'm seeing is that any discussion about those interpretations just gets wiped away without much logic behind it.
The specific case that prompted me to write this is actually pretty simple, and my goal is just to point out a classic example of how this works:
We have estimates for the virus mortality rates (basically looking at deaths relative to total infections);
we have those German estimates regarding how many people are infected but undetected (which would totally change the calculated mortality rate);
and then, on the flip side, we have Korea, which is testing incredibly intensely.
Those are our starting points.
So, logically, it follows that the number of undetected cases in Korea is probably tiny. Because of that, I guess applying some sort of standard "undetected rate" would basically just be massaging the numbers, which isn't really helpful if we're trying to describe things accurately or calculate true rates.
Also, mixing and matching bits of info like a cocktail without considering the context... well, that's a path full of traps, I think.
This observation doesn't really fit in with some of the very questionable math being thrown around, so I suppose it might not be the most welcome addition to the thread.
It feels to me like that kind of approach isn't exactly well-balanced, you know?
If we can't actually talk about arguments—or if only bad statistics are allowed—then we aren't going to get very far, and I don't really see the point of the conversation or what's actually left for us to discuss here.😁
I mean, I think it would be way better if people just provided counterarguments if they actually have them.
I know emotions are running pretty high right now; it's clearly not easy for anyone to deal with this or even moderate it.😁 So, I'll just stick to the purely numerical side of things in this post of mine.
-Steve
Over on the Politics board, we’ve seen plenty of attempts at statistical analysis or people trying to interpret the raw numbers we've been getting for this subject so far.
Of course, a little mistake here and there always manages to sneak in, doesn't it?
The real issue I'm seeing is that any discussion about those interpretations just gets wiped away without much logic behind it.
The specific case that prompted me to write this is actually pretty simple, and my goal is just to point out a classic example of how this works:
We have estimates for the virus mortality rates (basically looking at deaths relative to total infections);
we have those German estimates regarding how many people are infected but undetected (which would totally change the calculated mortality rate);
and then, on the flip side, we have Korea, which is testing incredibly intensely.
Those are our starting points.
So, logically, it follows that the number of undetected cases in Korea is probably tiny. Because of that, I guess applying some sort of standard "undetected rate" would basically just be massaging the numbers, which isn't really helpful if we're trying to describe things accurately or calculate true rates.
Also, mixing and matching bits of info like a cocktail without considering the context... well, that's a path full of traps, I think.
This observation doesn't really fit in with some of the very questionable math being thrown around, so I suppose it might not be the most welcome addition to the thread.
It feels to me like that kind of approach isn't exactly well-balanced, you know?
If we can't actually talk about arguments—or if only bad statistics are allowed—then we aren't going to get very far, and I don't really see the point of the conversation or what's actually left for us to discuss here.😁
I mean, I think it would be way better if people just provided counterarguments if they actually have them.
I know emotions are running pretty high right now; it's clearly not easy for anyone to deal with this or even moderate it.😁 So, I'll just stick to the purely numerical side of things in this post of mine.
-Steve