Thank you for responding, but I am little disappointed that some were unable to respond professionally. Here are some points though:
1. This is not a discussion about conspiracy theories or politics. If you dismiss it as such, you were simply being unprofessional to the point of being silly.
2. This statement is confused and confusing, as Republicans and Military usually dominate by mail voting (before the pandemic) and the statement ignores the fact that Trump was leading in the swing states by the time the polls closed on the West coast: " The main flaw is that you are making the erroneous assumption that a non-random order (or even a random order) of counting votes calls into question the total number of votes for a candidate. There is simply no such connection. It is well known that party affiliations differ between in-person voting and vote-by-mail and that vote counting is done in batches and certainly not in random order. That doesn't change whoever got the most votes."
Moreover, the statement is completely false as to the trend line that is created for each candidate as the majority of precincts are reported in a county and then tabulated for a state. If Trump was leading all evening with a 52 to 48 majority, and then a group of bells so upset have trump falling behind in a 4 to 96 trend you can see that the two data sets are of different origin. And this is what happened in many locations; just the right number of votes in the right trend appeared to reverse Trump's lead. Not too many votes. Not too little votes. But just enough votes to look like his lead had been reversed. A split half test of data reliability would reveal that data in the county has been tampered with elevated to not have the same origin. Of course there were many reports of fraudulent transformations done on votes but these cannot be verified because the data is not completely available.
3. Please share the many logic flaws that you can see. As I mentioned in my post, I welcomed all the thoughts about the fraudulent election. If you look at the difficulty flipping all those states from Trump to Biden you can see that the fraud is obvious. However, the exact method of the fraud is far more subtle. If you were a casino manager, you would throw out a person who was able to run the table like that." David Wagner: Your "analysis" has so many logic flaws that it's hard to know where to start."
4. This statement contains a series of myths about the election including the myth that most of the legal cases were rejected: Most cases we're not really even considered because of lack of standing and all the events of fraud were considered to be local and not part of a pattern of overall fraud. Furthermore, the term debunked is being misused: Judges do not debunk things. Judges simply weigh the case put in front of them to the best of their ability and either grant injunction or put the case on the docket. Of course, No voting data or witnesses are actually examined until the court case and subpoenas begin. And judges do make mistakes that have to be brought to another court on appeal. And we don't exactly know the demographic make up of voters who voted in person and those who voted by mail. Hence, the statement urges us not to repeat false hoods yet the entire statement is replete with falsehoods, ironically: " <span;Most Democrats voted by mail, and many more Republicans voted on election day, as they were encouraged to unsafely do. Many States count the election day votes first. (That makes the "90%" closer to 100%.) Therefore it was expected that Trump would have more votes counted first no matter how badly he lost. Even if Trump had lost all 50 States and DC, he would still be expected to have done well in the same day voting. (Also, the "...<span;>hundreds of affidavits of fraudulent activities..." were repeatedly debunked in court. Also State Secretary's of State were convinced, and others needed for certification. Let's not keep repeating falsehoods.) "
5. This is a very interesting point about independence and causation. It is important to note that positive correlation it is not causation, as we spend a great deal of time getting people to understand when they take statistics classes. Free will voting behaviors by millions of individuals in each state would be the definition of independence. Of course, there could be many positive or negative correlations, but the fact of the votes were cast by people who were not in communication with each other would be deemed independent action. "Multiplication occurs only when events are independent. And it is completely unreasonable to assume that political events in different states are independent of each other when states are part of a national political system. Voting patterns were positively correlated in the 2016 election, which is why pollsters who thought the chance of Trump winning tiny were gravely wrong. And they were positively correlated in the 2020 election."
6. As you well know, the multiplication rule of probability requires an assumption of the odds of an event occurring repeatedly, so 50-50 is a necessary assumption: "Nor is there any reason to simply assume a 50-50 chance of each side winning, pulled out of thin air."
7. This is a funny but hasty dismissal of the argument: 'Mark Twain said "Most people use statistics the way a drunkard uses a lamppost, more for support than illumination."I regret to say that this seems to be an example of such use. Unrealistic assumptions, especially ones incorrectly asserted to be mathematical laws, lead to unrealistic conclusions.'
8. Just to be clear about what is being compared... We are not comparing presidential elections; we are comparing the candidate going in the lead in several states with most of the polls having been closed for several hours and then watch that lead being exactly reversed. I watched the election results all night on CNN, and remarked to myself this has got to be fraud: how can Trump's lead those states be exactly reversed. The problem is remarkably similar to trying to flip an fair independent coin and get ten tails: " If you will forgive a follow-up email, a good way to evalute a statistical model is to compare it to empirical data and see how well the model fits. The relevant empirical data would be every historical presidential election ranked by imbalance in number of states won. I couldn't find that data in a quick Web search. But Wikipedia has a table of the 59 presidential elections ranked by imbalance in electoral votes, and I believe this will be a sufficiently close proxy to show the qualitative properties.
Joe Biden's 306 to 232 electoral vote victory ranked 46 of the 59 (where 1 is the most even and 59 the most lopsided), putting it in the top 22% of most lopsided Presidential elections. Above average, yes, but based on the empirical distribution far from highly improbable. Incidentally Trump's 538 to 304 victory over Hilary Clinton ranked 47, one notch more imbalanced.
In general, the empirical distribution is radically different from what a binomial model with 50% probability would predict. Only 2 of the 59 elections - both George Washington's - were even. And most importantly, imbalances that the model epuld predict as highly improbable are common in the empirical distribution. This is strong evidence that something is wrong with the model.
Statisticians are empiricists, not pure mathematicians, driven by data, not solely by assumptions and mathematical laws. Any useful model ought to fit past events in the class, at least reasonably. This one doesn't.
The use of proxy data means this is not a definitive result. Perhap the correct data table, presidential elections ranked by number of states won, is qualitatively radically different from the proxy data table, presidential elections by number of electoral votes won. But I doubt it."
Again, thank you for all your responses and I'm doubly appreciative for the intelligent responses... ;)
Best wishes,
Dave
Dr. David A. Wagner, MBA, CADC-II, NCAC-I, ICADC, S.A.P., D.B.A., Ph.D.
Faculty, Trident University International
Faculty, World English Institute
Faculty & Mission Team, Vietnam Bible Institute
Bible Translation Ambassador, Wycliffe Associates
Fellow, Academy of Marketing Science
Member, National Association of Cognitive-Behavioral Therapists
Member, American Statistics Association
Principal Data Scientist & Investigator,
http://www.SEMStatistics.com