A Matter of Probabilities

Was it Michigan that had a big chunk of votes, 136,000, that came in where every. single. one. was for Biden and none for Trump or any of the other candidates on that ballot? An excuse I heard given was they were mail in ballots and Democrats were more likely to vote by mail than Republicans.

Really? That’s the excuse?

Well, as a physicist I do know a thing or two about numbers and probability. Let’s say that about 40 percent of the electorate were Trump voters (steel man argument–assume that he really, truly would lose that race without shenanigans). Now, let’s further say that Trump voters are 1000 times less likely than Biden voters to vote by mail. And let’s just discount the minor party candidates, again, steelmanning as including them would actually weaken the explanation for the big bolus of Biden (only) ballots. That means that any random mail ballot would be 99,96% likely to be a Biden vote and only 0.04% likely to be a Trump vote.

Probabilities multiply. The odds of a fair flip of a fair coin coming up heads two times in a row is 1/4 ((1/2)*(1/2)). Three times in a row 1/8 ((1/2)*(1/2)*(1/2)). And so on. So we can do the same with chances of Biden votes in a row. One vote for Biden: 99.96% Two in a row: 99.92% Three 99.88%. Ten: 99.60% (See, the “explainer” says, “it’s not at all unlikely”)

But we go on. 100? 96.08% 500? 81.87% 1000? 67.03% (Um, things are getting a little bit shakier here, but still more likely than not).

5000? 13.53% (Okay, it could happen.)
10,000? 1.83% (Well, longshots _do_ come in sometimes.)
20,000? 0.03% (Um, well, surely something like that could happen _once_ in the history of voting. Right?)
50,000? .000000205% (It’s a big universe. Somewhere, sometime, a longshot like that just _has_ to come in.)
100,000? 4.21E-16% (Anyone want to try defending this one?)
136,000? 2.34E-22%

To give you an idea of just how preposterous that number is, there are only about 1E21 stars in the Universe.

If they had bothered to actually mix it up a bit they might have been able to claim these were real results. But, no, every single ballot for their preferred candidate in a chunk big enough as to produce utterly ridiculous results. But they didn’t bother to disquise it because they figured nobody would hold them accountable for it.

But, in this case, people did. There was a lot of “whoa! Are you kidding me” at the jump in the count for Biden with no increase for anyone else. And so we got “Oh, it was a data entry error. Our bad. Here. Fixed.”

Do they actually think that makes it better? Seriously? First off, how does a “data entry error” record 136,000 votes for one candidate and none for any others? How does a person “fat finger” such a number? Especially when counting was supposed to be close for the night anyway? But, okay, data entry error (remember: steelmanning). So, if it’s possible to enter 136,000 votes for one candidate by “accident” in a situation where no other votes were being recorded, how can we trust that other “errors” weren’t and aren’t being made where they are not so obvious, where other, real votes are going in so that a bunch of votes entered by “accident” for one candidate is not so clearly spurious? How many of the other votes recorded were also “data entry errors”?

How can we trust the results when they can make that kind of “mistake”? And is Michigan uniquely bad? Can someone explain why Michigan would be so very much worse than all of the remaining 49 States? Because if there isn’t some reason to demonstrate that Michigan is uniquely bad then we can’t trust any of the others either.

And isn’t it strange that these errors all go one way. The probability equation is a bit more complicated than that used up at the beginning of this post, but the general gist is the same. The probability that these “errors” going one way by a large preponderance happens by chance becomes vanishingly small.

There’s an old saw: Once is chance. Twice is coincidence. Three times is enemy action. And we’re well, well past “enemy action.

Are We the People going to let them get away with it?

Well They Seem Nice

Short one today. Saw this on FaceBook.

Reeducation camps “like China”. Banning voting for people over political views. Elimination being “over-egging the pudding.” (You’ve heard the term “damning with faint praise”? Well, this is the converse.)

And they have the gall, the gall, to say “Trumpers are not Americans.”
If these folk actually _win_ the election we’ll be armpits deep in blood before we’re through. It’s no longer a matter of “Oh, we lost this time. We’ll try to figure out how to win next time.” It’s a matter of “we lost to folk who want to destroy us and our line, render us extinct.”

That never ends well. Crack open a history book.