4 ms·

There are some convincing arguments made by data scientists examining time-series data. Most of the results are suppressed from Google, if you try searching "Ly

There are some convincing arguments made by data scientists examining time-series data. Most of the results are suppressed from Google, if you try searching "Lynda McLaughlin from the Data Integrity Group" you can find more info using Yandex or Bing. > According to their analysis, which was published online as early as Dec. 24, time-series election data shows Trump’s votes were decrementing in various counties instead of increasing as they do normally. > At least 30,593 votes were allegedly removed from Trump as result, including 17,650 votes in Dougherty County, 7,008 votes in Dodge County, and 5,935 votes in Putnam County. This is by simple analysis of time-series data that was posted to public sources like the New York Times. There was an open-source repository on Github with the analysis, but it has since been removed. I regret not downloading a copy. [1] https://www.americanthinker.com/blog/2021/01/theres_evidence_of_vanished_votes_in_pennsylvania.html https://www.americanthinker.com/blog/2021/01/theres_evidence...

Bishop_ 6y ago
The source you're citing is citing another source, which is behind a paywall. I'd like to look at the time series data as well, but it looks like either the "Data Integrity Group" didn't provide or The Epoch Times felt it unnecessary to share the data in the article. > The Epoch Times was unable to verify the allegation independently. [0] https://web.archive.org/web/20210104154823/https://www.theepochtimes.com/georgia-election-data-shows-17650-votes-switched-from-trump-to-biden-data-scientists_3640670.html https://web.archive.org/web/20210104154823/https://www.theep...
sb52191 6y ago
Two things that I'd need explained to trust this take: "Often, the data are in the hands of Democrat politicians and bureaucrats who refuse to make it available." 1) The Georgia SOS in charge of elections is a Republican, so why would it benefit him to work against the Republican Nominee 2) If the data is being held by Democrats who refuse to make it available, what "data" did The Epoch Times analyze that showed votes being removed? In my experience, stuff like this often boils down to people's lack of understanding of systems or practices and they read into "anomalies" that might look weird to an uneducated outsider, but for someone that knows what they're talking about, isn't sinister...
the_drunkard 6y ago
i'm curious if this phenomenon has existed during previous elections. for example, is this a normal event when compared to previous elections or is this type of activity indeed suspicious. from the article you linked: > The one thing the time-trend line should never do is drop, which would mean that votes are disappearing from Trump's tally. > That's exactly what happened. In at least 15 counties, according to the Epoch Times story, Trump lost in-person votes, while he lost absentee votes in three other countries. The total result was staggering: > At least 432,116 votes — 213,707 election day votes and 218,409 absentee votes — were removed in total.
- snakeboy 6y ago
  This is one fallacy I saw over and over and over again back in November. Taking 10 counties in Pennsylvania and saying "Doesn't that look suspicious to you??" It's all meaningless junk stats without a random sample of other counties around the country and from the last few elections for comparison.
pwinnski 6y ago
This is just bad, bad, bad math. Try running the exact same analysis for Biden votes. Turns out that drops too. "The one thing they should never do is drop" turns out to be completely incorrect! https://youtu.be/aokNwKx7gM8 https://youtu.be/aokNwKx7gM8
- igammarays 6y ago
  That linked video is refuting a completely different claim by a different person. When counting votes, numbers should never go down.
  - snakeboy 6y ago
    But honestly how reliable do you expect the data reports of individual counties? As I said elsewhere, stuff like this is mostly meaningless without comparing to a random sample of other counties, preferably from other elections. Of course as another commenter said, we can't find the data they're using, so theres is an entirely worthless analysis.

Bishop_ 6y ago

The source you're citing is citing another source, which is behind a paywall. I'd like to look at the time series data as well, but it looks like either the "Data Integrity Group" didn't provide or The Epoch Times felt it unnecessary to share the data in the article. > The Epoch Times was unable to verify the allegation independently. [0] https://web.archive.org/web/20210104154823/https://www.theepochtimes.com/georgia-election-data-shows-17650-votes-switched-from-trump-to-biden-data-scientists_3640670.html https://web.archive.org/web/20210104154823/https://www.theep...

sb52191 6y ago

Two things that I'd need explained to trust this take: "Often, the data are in the hands of Democrat politicians and bureaucrats who refuse to make it available." 1) The Georgia SOS in charge of elections is a Republican, so why would it benefit him to work against the Republican Nominee 2) If the data is being held by Democrats who refuse to make it available, what "data" did The Epoch Times analyze that showed votes being removed? In my experience, stuff like this often boils down to people's lack of understanding of systems or practices and they read into "anomalies" that might look weird to an uneducated outsider, but for someone that knows what they're talking about, isn't sinister...

the_drunkard 6y ago

i'm curious if this phenomenon has existed during previous elections. for example, is this a normal event when compared to previous elections or is this type of activity indeed suspicious. from the article you linked: > The one thing the time-trend line should never do is drop, which would mean that votes are disappearing from Trump's tally. > That's exactly what happened. In at least 15 counties, according to the Epoch Times story, Trump lost in-person votes, while he lost absentee votes in three other countries. The total result was staggering: > At least 432,116 votes — 213,707 election day votes and 218,409 absentee votes — were removed in total.

snakeboy 6y ago
This is one fallacy I saw over and over and over again back in November. Taking 10 counties in Pennsylvania and saying "Doesn't that look suspicious to you??" It's all meaningless junk stats without a random sample of other counties around the country and from the last few elections for comparison.

pwinnski 6y ago

This is just bad, bad, bad math. Try running the exact same analysis for Biden votes. Turns out that drops too. "The one thing they should never do is drop" turns out to be completely incorrect! https://youtu.be/aokNwKx7gM8 https://youtu.be/aokNwKx7gM8

igammarays 6y ago
That linked video is refuting a completely different claim by a different person. When counting votes, numbers should never go down.
- snakeboy 6y ago
  But honestly how reliable do you expect the data reports of individual counties? As I said elsewhere, stuff like this is mostly meaningless without comparing to a random sample of other counties, preferably from other elections. Of course as another commenter said, we can't find the data they're using, so theres is an entirely worthless analysis.