this post was submitted on 17 Feb 2025
753 points (92.3% liked)
Microblog Memes
6649 readers
2392 users here now
A place to share screenshots of Microblog posts, whether from Mastodon, tumblr, ~~Twitter~~ X, KBin, Threads or elsewhere.
Created as an evolution of White People Twitter and other tweet-capture subreddits.
Rules:
- Please put at least one word relevant to the post in the post title.
- Be nice.
- No advertising, brand promotion or guerilla marketing.
- Posters are encouraged to link to the toot or tweet etc in the description of posts.
Related communities:
founded 2 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
What is this website? I couldn't find any link to their board directly, but only through a search engine. Their board is just named with three first names and there is broad statements made about them all being passionate, data analysts, bla bla.
There is no specific "CV" for them. E.g. something like "X studied computer science and worked as a data analyst for ten years".
And the plots where they claim "suspicious" patterns looks like any aggregate. As the total number of votes go up, each machine is more likely to get towards the overall turnout, so your distribution peaks gets higher and your scatter plot scatters less. You see the same pattern with the Election day machines. Just that they only go up until 125 votes, rather than 250 or more total votes. So the spread remains stronger. Also the number of machines for early day voting is 964, whereas there was 3,116 machines for election day voting. This is another basic truth of statistics. As N goes up, the shape of your distribution gets more uniform.
So what do we see? We see exactly what is to be expected with a higher number of votes per machine. That the distribution gets narrower. And we see what is to be expected with a higher number of machines. That the distribution between machines gets more evenly.
Ignoring third party votes, this is a classical binomial distribution and you can test all of these effects easily by making your own "draw n out of N" tests.
https://en.wikipedia.org/wiki/Binomial_distribution
The only argument is that there is a higher result for Trump with early voting as opposed to election day voting. And that can needs to be analyzed in the context of demographics and other factors. For instance people who can take off work for election day voting have more white collar jobs and are demographically more inclined to vote Democrats. But of course accounting for these factors is not part of this "analysis".
I am not saying that manipulations are out of the question. But these people are clearly trying to bamboozle you with deliberate misinterpretations of statistics.
It's been ran by a few different independent orgs. I believe it all began from
Then some independent data scientists got in touch with him and ElectionTruthalliance and SmartElections was borne out of it.
I haven't verified the data in-depth, but AFAIK there are inconsistencies like this pattern only appearing on a certain brand of tabulator. I'm pretty sure you're missing something either way or someone else would've as easily refuted it. There is a lot more information posted on https://old.reddit.com/r/somethingiswrong2024/ and https://www.reddit.com/r/Verify2024/
There is a recent AMA from the SmartElection founder here with some more details:
https://old.reddit.com/user/Filmmaker_Lulu/comments/1hjjyx8/i_am_the_cofounder_and_executive_director_of/
even the exact method has been uncovered.
The duty to warn letter was debunked as junk
Debunked how?
Not sure how you could rationalise all those shown breaches.
I fucking hate when people just say "that's bullshit" without actually showing you where to find the information to prove it.