Faith is not "without evidence" argument

AProudLefty said:
That's what I want to know.

How do "predictable random numbers"and "unpredictable random numbers" apply to polling?

And how in the world are the random numbers predictable?
Click to expand...

Yep. There's basically two options that are being compared to polling. One is throwing darts at a dart board - you can hit the same number twice. The other is picking numbers from one of those bingo, spinning cages - once you pick a number, it can't be picked again. Neither, despite the flaws of polling, makes any sense
 
ZenMode said:
How do either apply to polling?
Click to expand...

Polls commit a math error.
Either:
1) The 'data' is not published (as is required for a statistical summary).
2) The data is not free of bias (as is required for a statistical summary).
3) The collection method is unknown, since the questions are unknown or the order of appearance of them is unknown or the number of people polled is unknown or the time the poll is unknown.

Polls do not declare and justify a variance either, which is required to calculate the margin of error value (a required number in a summary result).
 
AProudLefty said:
The more random numbers you generate, the more likely you'll see a pattern.
Click to expand...
Only with randU numbers.
AProudLefty said:
It's like how you see images in snow on TV like Poltergeist. Look up the Infinite Monkey Theorem. It's a fascinating study.
Click to expand...
Irrelevant.
AProudLefty said:
There is a reason why the lotto balls or dice have to be balanced.
The paint on them got weight which can influence the outcomes.
Click to expand...
Neither lotto balls nor dice generate randU numbers. Dice generate paired randR numbers, and lotto balls generate randN numbers. ALL of them are types of random numbers.
 
Into the Night said:
Polls commit a math error.
Either:
1) The 'data' is not published (as is required for a statistical summary).
2) The data is not free of bias (as is required for a statistical summary).
3) The collection method is unknown, since the questions are unknown or the order of appearance of them is unknown or the number of people polled is unknown or the time the poll is unknown.

Polls do not declare and justify a variance either, which is required to calculate the margin of error value (a required number in a summary result).
Click to expand...

So, let's use the upcoming presidential election as an example. Each state will have polling that will attempt to predict, within a margin of error, what the actual voting will be in each state. Obviously there are a number of variables that impact the difference between polling numbers and actual results.

Do you believe that you could throw a dart at a dart board with 0-100 and be as accurate as the polling results.
 
IBDaMann said:
After all this time, you still have nothing intelligent to say?
Click to expand...

I've never heard The Sock say much of anything intelligent.

He has numerous accounts here, a 'Sybil' which very well may be another account on another forum but which he uses here like a 'Harvey', and an account on another forum you and I both frequent.
 
ZenMode said:
Yep. There's basically two options that are being compared to polling. One is throwing darts at a dart board - you can hit the same number twice. The other is picking numbers from one of those bingo, spinning cages - once you pick a number, it can't be picked again. Neither, despite the flaws of polling, makes any sense
Click to expand...

Throwing darts at a dart board is a randU.
Picking bingo numbers or lotto balls is randN.

Polling is just randU. You might as well be throwing darts at a dartboard.

You are intentionally inserting a bias with the dartboard, and the same thing often happens with polls.
 
ZenMode said:
So, let's use the upcoming presidential election as an example. Each state will have polling that will attempt to predict, within a margin of error, what the actual voting will be in each state. Obviously there are a number of variables that impact the difference between polling numbers and actual results.

Do you believe that you could throw a dart at a dart board with 0-100 and be as accurate as the polling results.
Click to expand...

Yes. You are intentionally introducing bias in both cases.
 
Into the Night said:
Yes. You are intentionally introducing bias in both cases.
Click to expand...

I just want to make sure we are on the same page. For the next 100 presidential elections, referencing polling in each state and taking an average of all of the polling, you believe that you could be as accurate as the average pulling by throwing darts at a dart board and using those numbers?
 
Into the Night said:
Only with randU numbers.

Irrelevant.

Neither lotto balls nor dice generate randU numbers. Dice generate paired randR numbers, and lotto balls generate randN numbers. ALL of them are types of random numbers.
Click to expand...

No such thing as RandU and RandN numbers. Made up terminologies.
 
ZenMode said:
I just want to make sure we are on the same page. For the next 100 presidential elections, referencing polling in each state and taking an average of all of the polling, you believe that you could be as accurate as the average pulling by throwing darts at a dart board and using those numbers?
Click to expand...

Yes. You are intentionally introducing bias in those cases.
 
Into the Night said:
Yes. You are intentionally introducing bias in those cases.
Click to expand...

Yes? No. Even before you start, you know that certain numbers are basically impossible when you have two candidates. Do you think Biden or Trump would ever have 0-25% of the polling vote or 80-100% of the polling vote?
 
You must log in or register to reply here.
Back
Top