Biden got 255,000 ‘excess’ votes in fraud-tainted swing states in 2020, study finds

That isn’t what the paper says at all. It simply states hat Democrats had lost votes and Republicans gained votes except in areas f suspected fraud. So say in district A Democrats out number Republicans 3 to 1, in the vote tally Republican vote increased by an increase increase that lead to a ratio of 2.5 to 1.5. Republicans lost but gained votes.

LTC David Brown I know what the paper says. I also know the evidence does not support the conclusion. This paper has already been deconstructed in both methodology and conclusion by reputable people, and the journal in question is taking some flack for even publishing it. Oh and the author of this study has a history of…dubious relationships with the truth AND was briefly employed by the Trump administration. Not exactly a high confidence interval…

Not that you care, it says what you want to hear and makes you feel good so it must be true

SFC Thomas Foreman - I left a lot of room in my initial discussion about he veracity of the study. Interesting that you added spin that was incorrect to reinforce your opinion.. That is not the inverse claim. Have a nice try.

LTC David Brown - Yeah, I used the term Republican controlled as that is what was stated in Eggers and Grimmer's analysis. I.e. Lott didn't mention that the pairs of uncontested counties he used were Republican but it appears that in actuality they were.

LTC David Brown - I did some looking around and Lott himself discusses that the pairs are really Republican and Democratic counties.

"I classify those counties that Trump carried as Republican counties and Biden's as Democratic ones. Since the turnout change may differ for Democratic and Republican counties, I identify the counties supporting Trump and Biden with two separate variables. When Biden won a county, the values for the Republican variable are zero. Similarly, when Trump won, the values for the Democratic variable are zero."

https://deliverypdf.ssrn.com/delivery.php?ID= [login to see] [login to see] [login to see] [login to see] [login to see] [login to see] [login to see] [login to see] [login to see] [login to see] [login to see] [login to see] [login to see] [login to see] [login to see] [login to see] [login to see] [login to see] [login to see] [login to see] [login to see] [login to see] [login to see] [login to see] 78&EXT=pdf&INDEX=TRUE

SPC Kevin Ford - I also looked, and it is indeed Trump won counties

LTC David Brown -
"Has Lotte responded to Eggers and Grimmer yet?"
Not that I've seen. From what I understand when they critiqued his earlier paper that presented some of the same errors he stated they were right and made corrections, but the new version now makes the same error somewhere else.

CW4 Guy Butler
This is quite a damning paper for Lott. They did find the thing that popped out at me from reading it least week; that is the paper is a very large non sequitur. I.e., fraud is not the only possible explanation of his numbers. However, the paper those guys did goes a lot further in recreating his models, finding his errors and pointing out things where he misreported what he had done. I.e., they point out that the method to derive the numbers themselves was flawed so there is no reason to move onto looking at if the analysis of those numbers was correct.

"To be clear, we do not consider the version of Lott’s (2022) specification without an intercept to be an appropriate way to diagnose electoral fraud: the pattern of absentee and in-person voting could differ across county boundaries (which are consequential for schooling and other important public services) for many reasons other than electoral fraud. But a specification that yields substantively different results depending on an arbitrary coding decision can be immediately disqualified as a method for evaluating electoral fraud."

SPC Kevin Ford Not to mention that this is the second time they’ve pointed out these flaws in his paper.

CW4 Guy Butler - It looks like he fixed what they pointed out in the earlier version but then he went and reintroduced the same exact error somewhere else in the final version. Unfortunately for Lott, his entire conclusion looks like it was dependent on this error.

He also made some pretty sloppy errors like switching signs of numbers to make it look like there was a problem in Biden's favor when it was in Trumps or misrepresenting the formula used.