2543
Comments (114)
sorted by:
You're viewing a single comment thread. View all comments, or full comment thread.
7
Shinydenim 7 points ago +7 / -0

Not exactly. In the area of machine learning the problem of overfitting or underfitting a test set to a training set occurs when the resolution of the parameters defined in a line of best fit (regression analysis) is either too fine or too sparse. While saying that a 6th order polynomial assisted an algorithm to predict the vote magick on Nov 3 SEEMS arbitrary, is in fact a testable and reproducible stipulation.

13
Hambonemcgee 13 points ago +13 / -0

Instead of using advanced mathematics, perhaps we can have one vote=one vote, maybe counted by humans and utilizing voter ID. If it’s complicated, it’s unnecessary and will lead to fraud. Voting is an extremely simple process.

4
krzyzowiec 4 points ago +4 / -0

The crux of the scam seems to be that they relied on total number of registrations as a pool to draw fraudulent votes from.

So voter ID is irrelevant because they are doing it algorithmically using people who are already verified, but may not have even showed up to vote.

That’s why they are desperate to have everyone registered, especially illegals. They don’t even need you to go to the polls, they just want you to exist in their database as a pawn to be used.

I think this would be easily detected in a manual recount because the backup paper ballots for these voters simply would not exist. That’s why they are shredding them.

We already have the tools to detect this fraud, the problem is corrupt people are preventing them from being used or ignoring the results.

4
Matcat58 4 points ago +4 / -0

Obstructing investigation into blatant voter fraud is tantamount to treason. And should be met summarily with the same punishment.