Further Explanation of Simulation and Results
(They confirm what seemed clear, back when I compared the final county-by-county vote ratios with what I saw on the election-night broadcasts.)
When you look at the large variations in Trump/Biden ratios among WI counties, you would expect to see the running, cumulative T/B ratio frequently go greater than 1.1 (and less than .9 : <1 is a Biden lead, and probably less-frequently because Biden's strength was in the two big counties of Milwaukee and Dane)* ...that is, outside the narrow range of either Trump or Biden having no more than a 5% lead (~.9<T/B<~1.1). The simulation is wholly consistent with and essentially confirms that WIDE spread, not the narrow spread over the duration of the returns seen in the broadcast results. Put another way, T/B was (at least on the Mrs. Greenspan broadcast, see screen captures) confined to <1.1 and always positive (Trump leading)—until just before the end when it flipped to less than 1 (T/B = .98, the final ratio for every simulated sequence, i.e. the Biden win ratio). Again, this is NOT an assertion that Trump was the actual winner. This is just the simulation showing what is basically obvious: that the tight range of the Wisconsin T/B ratio throughout the evening was extremely improbable, given the wide variations in T/B over the counties.
This is an assertion that there was ample statistical basis for a "conspiracy theory" or whatever you want to call monkeying around with votes as shown in the broadcast WI election returns. This project is about actual analysis and actual debate—things long absent from the DC-NY nucleus of government-media group think (and very much including the courts) with its unthinking jargonistic patter of "baseless" this or that.
The final vote ("canvassing," checking the votes, by the WI Election Commission) gave the county-by-county results shown on the Commission website. My simulation is simply a demonstration (basically a simple Monte Carlo simulation) generating random sequences of the 72 counties and summing the votes (as a cumulative T/B ratio) as each county "reports" in the simulated sequence. No random sequence I've seen (and I doubt you will find one) shows the nearly-constant, tight T/B spread of the Wisconsin results that were broadcast. This strongly suggests that the tight T/B spread was so anomalous as to be inexplicable, unless the vote-reporting was garbage, about which I speculated.
To address a "sampled data" issue that some might mention. Yes, the election-night data shown were sampled at 26, 32, 35, 43 65, 72% reported votes (and then many samples >90%); a more thorough model could take samples from each random sequence at roughly those percentages (and the javascript is there for you to modify or pipe into a sampling or other post-processing program). However, I will state without simulation or proof: that if you run the simulation (refresh the page) a few dozen times, you will be sufficiently convinced that no sequence will yield samples like those broadcast. The wide T/B spreads that occur in the results won't amazingly narrow for those (26...72%) samples or any set of samples. (Also, making an analytical—meaning non-simulated—calculation of any of this would be very involved (for me, anyway) and would require many assumptions that could be challenged. The advantage of a computer simulation—with a few reasonable assumptions—is you can run it many times and see whether any of the results come close to what you observe. None of the simulated county-reporting sequences comes close to what was on TV/internet.)
So yes indeed, there was solid justification for seeing fishiness in the hardly-changing Wisconsin election returns displayed on TV/internet (and yes, the likely explanation is they were garbage, see below). The broadcast Wisconsin returns seemed exceedingly improbable, so it was very reasonable that people were left wondering what was going on. You can read On Wisconsin/Off Wisconsin here (same link as those up the page).
So, what's the explanation? According to a person I've spoken with at the WI Elections Commission (quite a few times about ballot access, but only two times about the election results), there is no central vote summation done in Wisconsin on election night. (Only a week or more later does the Commission check ("canvass") the results.) On election night, the precincts (within a county) transmit ("modem in" I was told) their results to the county election authority (not to a Wisconsin State authority) for each county's totaling of votes. An individual county displays its final summation (and maybe intermediate summations too, I don't know) on its website. So, the results are "diced" (quasi-veg-o-matically) among 72 counties. (And it's "quasi-veg-o" because the results are not diced into 72 counties; the results begin and end in each of 72 counties on election night; there is no Wisconsin-provided total vote-count until weeks afterwards.) That means that news reporters (such as they are) would have had to check 72 county websites simultaneously for an "accurate" report. One reporter could not do this, obviously. It would take, for example, eight reporters each "simultaneously" (or close to it) refreshing nine county websites to get "snapshots," and then adding up the results.
According to the person at the Commission, the Commission emphasizes that the results on election night are "unofficial." I said that the broadcast results seem to be garbage, then. The response was that some people call them that. I was also told that the news outlets tended to report the Associated Press' numbers, since AP (I was told) had reporters checking the Wisconsin county websites. I would be somewhat surprised if these AP reporters did a methodical check (such as 8 reporters each checking 9 sites nearly-simultaneously) because the county-by-county website check is so kludgy, that most people—unless they were intensely-focused on Wisconsin—would just say "forget this," and just check on the big counties or do some other makeshift workaround. (And you couldn't blame them for getting frustrated because Wisconsin makes complicated what other states make simple.) I suggested that the Commission might put out a very clear press release prior to election night (and repeat it often), stating that election-night results are very-likely garbage, not simply "unofficial." In fairness, the Commission person told me that they did release several announcements emphasizing the "unofficial results" prior to election night.
If the media nowadays did a little research and were actually concerned with accuracy, they would have put up a big red flag on the Wisconsin results during election night or even prior to election night. So you should keep this in mind when you watch next week's Wisconsin election returns: that they may well be garbage. (And any real journalists in the media can confirm what I found out just by talking with the Commission.)
So this is a long way of showing that the Wisconsin election-night, broadcast returns were garbage. Big deal, you say. Yes, it is a big deal when the media were showing obvious garbage on their election-night screens while their (Mrs. Greenspan) "pundits" were jabbering on with their "analyses." And then later, the media unify their chorus about the "baseless LIE" and "conspiracy theories" when their non-researching, know-nothing "reporters" don't ask obvious questions about what their own broadcasts showed; and then they don't go to obvious sources (the WI Commission) to get answers.
There was ample basis for questioning the bizarre Wisconsin results, which would naturally lead to questions about similar late reversals in Michigan and Pennsylvania for a Biden win (though I have not looked at those results, so I'm making no assertion about them other than that they reversed late in the game). If the media had real reporters and "pundits" who had any ability to observe, analyze, or just ask obvious questions, then they could have "fact checked" what they showed on their own screens and have headed off at least some suspicions. Instead, they quote polysci-JD lawyers and judges, whose analytical abilities are nil, as demonstrated by my extensive, and still growing, body of evidence.
And I'm also going to re-grind my axe: that the media—as exemplified, in particular, by this Daniel Dale, the CNN fact checker who doesn't know the difference between a fact and an opinion—have become an Orwell-Huxley-Kafka amalgam that uses the word, "LIE," in a repetitive chant to stifle any questioning of media-deemed "valid" government information. In particular, the media's skewering Congressional objections (heaven forfend that our representatives would exercise their 1st Amendment speech on the floor of Congress and utter that dirty word, "object") to vote counts that seemed strangely improbable—which my simulation shows they were, in fact, in WI; and I welcome debate on this. And I made the small effort it took to confirm with the WI Election Commission that their distributed counting process (of independent county tallies) is essentially destined to yield squirrelly results if reported real-time. (In fairness to the Commission, this distributed method is mandated by statute, I was told.)
So I (and others) was completely justified in doubting the Wisconsin results, which doubts I documented. And BTW, if you can't handle an objection to your assertion (it's called real debate in a real democracy), then move to Russia, where showing any doubt about the Ukraine "special operation" gets you 15 years.
Now here's a bona fide conspiracy theory, which I know is a fact. When I was campaigning in New Hampshire, I searched high and low but couldn't find a Walmart that had Pepperidge Farm Double Nantucket cookies ("Double Nantuckets" in addict shorthand), of which I have left a long evidence trail of purchases—for the FBI, cookie-monster division. Here's a buy just a couple of months after the Covid lockdown, not in New Hampshire: $2.44, the price it had been for a long time, and I have taken data for years.
More recently, the delayed effect of the Greenspan Socialist Transform's inflation fuel, sparked by Covid and WW III, is evident at $3.72:
I thought that maybe New Hampshire had instituted a zero-Nantucket-cookie retaliatory measure for the tax that was the subject of the titanic state-battle and my NH v. Mass blog entry. But now, Double Nantuckets appear to be totally gone from all Walmarts since my 10/1/22 Last Nantucket Waltz (and I go to a wide sampling of Walmarts per unit time). Just today, in my addict's desperation, I tapped on the image in the Walmart app, and I saw this Holy Cow, $12.95 shocker:
My conspiracy-theory instincts tell me that Liz Warren is behind this. I think the recent decision out of the 5th Circuit (Texas, among the other basically-conservative states down yonder) federal appeals court ( U.S. consumer protection watchdog's funding unconstitutional, court rules ) set Liz on a vengeful rage. I think that Liz is not only getting even for this decision (I haven't had time to read the decision yet, so my crisp analysis extends only to the Double Nantucket crisis), but I think she may be secretly funding the consumer protection agency out of the windfall profits from sky-high Double Nantuckets. And maybe DeSantis is skimming some Nantucket profits for his Vineyard project. I have declared my Senate campaign against Liz for 2024, and I'll be researching Nantucketgate for what may be a sure-fire campaign issue against her.
I'll add one especially-personal postscript about New Hampshire. Driving over the little bridge into Laconia, over the narrow neck of Lake Winnisquam—and looking at the placid boats with their fishing lines out in the sun-dappled water—I heard my 1966 words: "Hey Dad, we can stop here and go fishing, right?" You can travel the world over, and you won't find a more idyllic scene, for a fishinaholic kid—or most anyone. And the Portsmouth coast brought back more of the same.
But for now, take heed this election night that the simulation tells the tale: chances are good that the broadcast election returns from Wisconsin will be garbage. So find some Double Nantuckets, if you can, to calm yourself.