Explained: Red Mirages, Blue Shifts, and Data Dumps in the 2020 Election

[Z.W. Wolf]

Rand Paul has tweeted a link to a site which alleges statistical anomalies
https://www.foxbusiness.com/politic...in-states-where-trump-lost-calls-out-big-tech




The site. It dates back to Nov. 24.

https://votepatternanalysis.substack.com/p/voting-anomalies-2020/comments

One doesn't even need to look at the "analysis" It's once again based on a false premise that a statistical analysis needs to done at all. We have the primary records. And it's disingenuous to pretend we don't. This is just one more case of pointing to mail ballot tallies being released in large batches (the "data dumps") and calling it mysterious and suspicious.

This was all predicted before the election. The mail ballots in some states were not counted until election day (per laws passed by Republican state legislatures). It takes time to physically process mail ballots, so it was expected the tallies would come in later.

We even had names for this before the election.
"The Red Mirage" - the count of in-person ballots cast on the day of the election will come in first and these will favor Trump. It will look as if he has an unsurmountable lead. But this will be nothing more than a mirage (illusion).

"The Blue Shift" - The mail ballots are going through a process of signature verification, and then the ballots are laboriously removed from the envelopes. It takes time even with a machines to open the envelops. In some cases, notably in some jurisdictions' in Wisconsin, mail ballot tallies will not be released until all mail ballots are counted. So there will be a sudden large shift in relative vote counts, because of large batches of Biden-leaning mail ballots being reported. The Red Mirage will disappear.

Mon November 2, 2020
https://www.cnn.com/2020/10/30/politics/red-blue-mirage-election-results/index.html

Likely shift from red to blue
Some people call this the "red mirage" or the "blue shift," where early results favor Trump but later ballots even things out and might even put Biden ahead once all the results are tallied. This won't be a sign of fraud or irregularities. Rather, it's just a reflection of how states count votes. Some states process early ballots first, and will report those early in the night, while others save them for last. Here is a breakdown of what to watch for in the pivotal states.

Content from external source

Sunday November 1, 2020
https://www.reuters.com/article/us-...ware-of-early-u-s-election-wins-idUSKBN27H1A6

Don’t be fooled, voting experts and academics say. Early vote counts in the most competitive, battleground states can be particularly misleading this election because of the surge in mail-in or absentee ballots, and the different ways that they are processed.

The states that count mail-in votes before Election Day are likely to give Biden an early lead, since opinion polls and early voting data suggest those ballots favor the Democrat. Conversely, the states that do not tally mail-in votes until Nov. 3 will likely swing initially for Trump.

These so-called red or blue mirages will disappear as more ballots are counted, though experts say it may take days or even weeks to process the huge number of mail-in ballots, spurred by voters seeking to avoid crowded polling stations because of the COVID-19 pandemic.

“Be patient,” said Gerry Cohen, a member of the Wake County Board of Elections in North Carolina. “You need to count all of the votes, and that’s going to take some time.”

Content from external source

 

[Z.W. Wolf]

https://www.msn.com/en-us/news/poli...ps-quest-to-overturn-the-election/ar-BB1bs02h

[Before the election] Trump also was given several presentations by his campaign advisers about the likely surge in mail-in ballots — in part because many Americans felt safer during the pandemic voting by mail than in person — and was told they would go overwhelmingly against him, according to a former campaign official.

Content from external source

 

[Mick West]

The Wikipedia article on this is a good read, with historical examples

Article:

The phenomenon was first identified by Edward Foley of Ohio State University in 2013. He found that Democratic candidates are significantly more likely to gain votes during the "canvass" period, which is the votes counted after election night. This asymmetry did not always exist; in the 20th century, as recently as the 1996 United States presidential election, Republicans and Democrats were both able to cut their opponents' leads during the canvass period. Foley conjectured that the 2002 passage of the Help America Vote Act accelerated the pronounced asymmetry of the blue shift phenomenon, because it required states to allow provisional ballots to be cast. He later found that the variation in the size of the blue shift is positively associated with the number of provisional ballots and the Democratic partisanship of the state in question. The growth in the persistent blue-shifted overtime vote began with the 2004 United States presidential election. However, Foley has stated that political scientists have not fully "pinned down causality" of this phenomenon. Foley did not find that mail-in or absentee votes favored either party.

Source: https://en.wikipedia.org/wiki/Blue_shift_(politics)

The current situation was anticipated by many people.

Article:

Foley expressed concern that this phenomenon, along with difficulties in conducting an election during a pandemic, could lead to "a perfect storm" in the 2020 United States presidential election. The blue shift could decide the election. This concern was particularly pronounced due to the fact that incumbent president Donald Trump did not state before the election whether he would accept election results. Given Trump's repeated attacks on mail balloting, nonpartisan experts warned that he might plan to build up a lead among ballots cast on Election Day, claim victory, and then say, "stop counting ballots because all those absentee ballots are illegitimate," thus disallowing the likely blue shift.

Source: https://en.wikipedia.org/wiki/Blue_shift_(politics)

 

[Mauro]

The current situation was anticipated by many people.

Article:

Foley expressed concern that this phenomenon, along with difficulties in conducting an election during a pandemic, could lead to "a perfect storm" in the 2020 United States presidential election. The blue shift could decide the election. This concern was particularly pronounced due to the fact that incumbent president Donald Trump did not state before the election whether he would accept election results. Given Trump's repeated attacks on mail balloting, nonpartisan experts warned that he might plan to build up a lead among ballots cast on Election Day, claim victory, and then say, "stop counting ballots because all those absentee ballots are illegitimate," thus disallowing the likely blue shift.

Source: https://en.wikipedia.org/wiki/Blue_shift_(politics)

Wow, that's what I call a prophecy!

 

[Mendel]

Oh, the paper that published the research cited in that Wikipedia article is even more predictive:

Article:

Explaining the Blue Shift in Election Canvassing​

Edward B. Foley, Ohio State University
Charles Stewart III, Massachusetts Institute of Technology

[..]
Normatively, a growing gap between initial and final vote counts — and a gap that is skewed in one direction — raises the possibility that losing candidates and their supporters may increasingly regard the vote count as rigged. The public is more likely to perceive, rightly or wrongly, that ballots counted for the first time after Election Day are more susceptible to partisan manipulation than ballots counted on Election Day, with this perception stronger if these overtime ballots tilt more favorably toward one party and diverge from the Election Day count. Should this occur, it has obvious implications for the legitimacy of American national elections.

The notion of candidates harvesting extra votes during the canvass is unpalatable to begin with, but it becomes especially problematic if this harvesting process is perceived as systematically one-sided. Even if every additional ballot counted during the canvass is an indisputably valid vote that deserves to be counted, the public will perceive this harvesting process in the aggregate (not ballot-by-ballot). From this aggregate perspective it may start to look as if, when an election goes into extra innings, one of the two teams is given extra at-bats.

Source: https://papers.ssrn.com/sol3/Delivery.cfm/SSRN_ID2659239_code1609318.pdf?abstractid=2653456&mirid=1

2020 is special because of the pandemic, which produced many more mail-in votes than usual; and the partisan split over the seriousness of the pandemic and the "fraud" rhetoric of the Trump campaign meant that this year, absentee ballots also skewed predominantly Democrat. And then Republican legislatures blocked initiatives to have absentee ballots counted early in Michigan, Wisconsin and Pennsylvania:

Article:

The reasons so many Biden-friendly mail ballots were counted so late in states like Michigan, Pennsylvania and Wisconsin are actually rather simple. One reason is that populous urban areas (which lean strongly blue) have more votes, which means it takes longer to count them. Another is that Trump spent months raising concerns about mail-in balloting, leading to an extraordinary partisan imbalance in which voters embraced them.

But the biggest reason for the lag is that Republicans allowed it to happen. Despite the urging of state election officials, the GOP-controlled legislatures in those three key states all declined to let mail ballots to be counted earlier — unlike the vast majority of states — despite knowing the number of mail ballots would dwarf any previous election.

Source: https://www.washingtonpost.com/politics/2020/11/05/trump-blames-democrats-late-counted-mail-ballots-he-should-blame-gop/

These Republicans make it look as if creating as big of a blue shift as possible was their strategy.

This year, we have two very prominent issues of "alternative facts" put to the test:
1) "Cornovirus is a hoax" vs. "Covid-19 is a serious epidemic"
2) "The election is rigged" vs. "Biden won the election fairly"

The first issue has been playing out all year, with accounts on social media of Covid-deniers falling sick and dying; and even the White House contracting the virus in larger numbers than some small countries; and nurses in overworked ICUs all across the USA talking about the reality of this epidemic.

The second issue is being decided in the courts in a heavily litigated election that find the proponents of election fraud short of evidence. (I saw a youtube clip yesterday titled "Giuliani drops bombshell evidence" that was 95% rhetoric with no evidence mentioned.)

I hope that both issues will illustrate to many people that truth is often not a matter of opinion, and that denying the truth can cost lives.
Unfortunately, this insight comes too late to save us from the climate catastrophe.

 

[Z.W. Wolf]

From a presentation recorded last week in the White House Diplomatic Room, and released today. No news media present.

Leading up to the presidential election, we were warned that we should not declare a premature victory. My opponent was told to stay away from the election. Don’t campaign. They were acting like they already knew what the outcome was going to be. They had it covered. And perhaps they did.

Wisconsin. We are leading by a lot. And then at 3:42 in the morning there was this. It was a massive dump of votes. Mostly Biden; almost all Biden. I went from leading by a lot to losing by a little.

Ultimately I am prepared to accept any accurate election result, and I hope that Joe Biden is as well.

Content from external source

This was debunked by AP on November 4, 2020

https://apnews.com/article/fact-checking-afs:Content:9647421250

CLAIM: Graph shows that over 100,000 votes for Joe Biden in Wisconsin “came out of nowhere.”

AP’S ASSESSMENT: Missing context. Milwaukee didn’t finish counting absentee ballots until 3 a.m. The outstanding ballots at that point -- including 169,000 from Milwaukee -- overwhelmingly broke for the Democrat.

Most of Wisconsin’s largest cities, including Milwaukee and Green Bay, count absentee ballots in one centralized location. The count of Milwaukee’s absentee ballots, for example, began Tuesday morning and was live-streamed for people to view throughout the day on YouTube. But Wisconsin law requires that the results of those absentee votes be reported all at once, Wisconsin Elections Commission Administrator Meagan Wolfe explained Wednesday.

“With that you’re going to get the result sets all as one, with all the absentees,” Wolfe said when asked about the overnight spike in vote totals.

According to The Associated Press, in Milwaukee, the state’s most populous area, the city didn’t finish counting its roughly 169,000 absentees until around 3 a.m. When they did, Milwaukee police escorted the city’s elections director from a central counting location to the county courthouse to deliver thumb drives with the data, and it was immediately folded into the overall count.

Content from external source

 

[Mendel]

They were acting like they already knew what the outcome was going to be.

Content from external source

Trump has "one of the great memories of all time" , surely he recalls what the polls were like leading into the election?

 

[Tim Chow]

I'm having a conversation with a friend who pointed me to Rand Paul's claim, and Metabunk was my first go-to website for information. The information about Milwaukee submitting all its votes at once is the most useful thing I've gleaned from this thread. However, as I understand the election fraud claims, it's not about "blue shift" per se. One of the main claims is that the sudden spikes appear only in battleground states. The proposed fraud mechanism is that fraudsters planned ahead of time to inject huge bundles of fake votes into the system, by exploiting weaknesses in the mail-in ballot system, but they only bothered to do this in states where the race was predicted to be close.

So my first question is, can similar spikes be observed in non-battleground states? I have had trouble finding the answer to this question in the usual "debunking" sources. It would be helpful if the answer were yes.

A second question is, how many of the allegedly suspicious spikes can be identified in the same way that the Milwaukee spike was? That is, can we point to each of the largest spikes and say, "That spike is Milwaukee; that spike is [Municipality X]; that spike is [Municipality Y], etc."? I recognize that this might not be possible, but if it is possible, I think that would be useful information for debunking purposes.

 

[Z.W. Wolf]

The proposed fraud mechanism is that fraudsters planned ahead of time to inject huge bundles of fake votes into the system, by exploiting weaknesses in the mail-in ballot system, but they only bothered to do this in states where the race was predicted to be close.

The narrative being that The Bad Guys have exploited this weakness: They could wait to see exactly when huge numbers of Biden votes would need to be injected, and could inject just enough to give Biden the win.

But... how were they are able to both produce the fake votes and to hide the evidence that they are fake? There's no evidence that this happened. When it comes to proving that there were fraudulent votes in this election, you're still at square one

That there was a spike led people to be suspicious. Now that the reason the spike happened has been explained... there still remains a residue of general suspicion

A general impression of facile corruption leads people to think something like, "If they can inject the votes at the right time, they can produce the votes." A non sequitur of the crudest kind.

The spike in Wisconsin was caused by a law which the Republican legislature put in place. But a spike itself is not very important. The important question is this: Is the proportion in that spike of Biden votes to Trump votes credible?

Yes it is. These ballots have gone through two "filters." Mail ballots favor Biden and these areas favor Biden.

A somewhat similar argument is more common: It is not credible that Biden conveniently got these large margins in Swing States only, because he underperformed Clinton everywhere else.

The first fallacy here is that "Swing State" is defined as Swing state that Biden won. Biden lost the swing states Florida, North Carolina, South Carolina, Ohio and Texas. They are not counted as a Swing State in this theory.

The claim in general was debunked weeks ago:

https://www.usatoday.com/story/news...ed-clinton-most-major-metro-areas/6349084002/

The claim: Biden underperformed Hillary Clinton in every major metro area except four in swing states run by Democrats, where the vote exceeded the number of registered voters

Content from external source

​

As states work to certify their vote counts, some are still searching for ways that President-elect Joe Biden’s win might not be legitimate. Claims that vote counts could not be accurate due to the sheer number of votes Biden received are circulating online.

Rasmussen Reports, a conservative-leaning pollster, shared an article by right-wing media group Gateway Pundit on Twitter, which was screenshotted and shared to Instagram and Facebook. Rasmussen’s tweet pulls a quote from the article.

“Biden underperformed Hillary Clinton in every major metro area around the country, save for Milwaukee, Detroit, Atlanta and Philadelphia ... in these big cities in swing states run by Democrats ... the vote even exceeded the number of registered voters,” the tweet reads.
Biden received significantly more votes than Clinton in most major metro areas.


USA TODAY aggregated official 2016 voting data and unofficial 2020 voting data from the country’s 10 largest cities — or the counties in which they’re located, if data by city was not available — plus Detroit, Milwaukee and Fulton County, where Atlanta is located. Biden received more votes than Clinton in 10 of those 13 locations. All data was aggregated Nov. 18.

According to unofficial counts, Biden outperformed Clinton in:

Chicago by 19,613 votes,
Philadelphia by 20,150 votes
Milwaukee by 6,004 votes.


He also outperformed the 2016 presidential candidate in the counties that include:

• Houston (+210,279 votes)
• San Antonio (+128,902 votes)
• San Diego (+226,849 votes)
• Dallas (+137,496 votes)
• San Jose (+103,574 votes)
• Phoenix (+337,867 votes)
• and Atlanta (+84,093 votes) are located, according to their unofficial counts.

The three cities where Biden appears to have underperformed Clinton are New York, Los Angeles and Detroit, one of the cities in which the post lists Biden as allegedly overperforming.

Content from external source

Trump lost Texas but won Georgia, yet they show a similar urban versus rural pattern. But Texas is not counted as a Swing State in this conspiracy theory. If Texas had gone all the way for Biden, then suddenly it would be counted as a swing state. This is Alice in Wonderland logic.

https://www.texastribune.org/2020/11/06/texas-trump-biden-counties-rural-suburban-city/

Four big blue counties: The counties that are home to Texas’ four biggest cities — Houston’s Harris County, San Antonio’s Bexar County, Dallas County and Austin’s Travis County — reliably go for Democrats. They saw their combined number of registered voters increase by 630,796 since 2016.

Content from external source

https://www.brookings.edu/research/bidens-victory-came-from-the-suburbs/




https://www.brookings.edu/research/bidens-victory-came-from-the-suburbs/

 

[Tim Chow]

Z. W. Wolf: Thanks for the information. I looked carefully but it doesn't seem that any of the data you presented answers my two questions. Did I miss something?

 

[Mendel]

So my first question is, can similar spikes be observed in non-battleground states? I have had trouble finding the answer to this question in the usual "debunking" sources. It would be helpful if the answer were yes.

Article:

In 17 states, statutes or 2020-specific rules allow for absentee/mail-in counting to begin before Election Day. In 16 states, statutes or 2020-specific rules allow for absentee/mail-in counting to begin on Election Day before polls close. In 17 states, statutes or 2020-specific rules say absentee/mail-in ballots cannot be counted until after polls close on Election Day or have that effect.

Source: https://ballotpedia.org/When_states_can_begin_processing_and_counting_absentee/mail-in_ballots,_2020

The article lists the states, take your pick, avoiding the first group.

 

[Mick West]

So my first question is, can similar spikes be observed in non-battleground states? I have had trouble finding the answer to this question in the usual "debunking" sources. It would be helpful if the answer were yes.

It's going to depend on if there's a highly polarized county or city that has a counting center that reports in large batches. I've not looked at all the states. Here's Alabama

There are smaller spikes, and one that's 75% trump.

 

[Z.W. Wolf]

So my first question is, can similar spikes be observed in non-battleground states? I have had trouble finding the answer to this question in the usual "debunking" sources. It would be helpful if the answer were yes.[/ex]or debunking purposes.

Nothing as dramatic as in Wisconsin that I can find. This was due to an extraordinary state law and is fully explained. There were lesser bumps such as in Michigan, due to individual jurisdictions reporting large numbers of mail ballots.
see: https://www.metabunk.org/threads/ex...-michigan-bump-its-detroit.11460/#post-242781

All have been fully explained. Mail ballots are labor intensive to process and then count. It takes time to count them and report them. There is no mystery. Anyone who is still making an argument that there is a mystery to this is ignorant of the facts or is being disingenuous.

Naïve people who are simply repeating some version of the "mysterious late votes" story may be ignorant of the facts. As for the "experts" who are still making a formal argument that Blue Shifts are evidence of fraud; it's not credible that they are still ignorant of this common knowledge.

A second question is, how many of the allegedly suspicious spikes can be identified in the same way that the Milwaukee spike was? That is, can we point to each of the largest spikes and say, "That spike is Milwaukee; that spike is [Municipality X]; that spike is [Municipality Y], etc."? I recognize that this might not be possible, but if it is possible, I think that would be useful information for debunking purposes.

Answer: All of them.



Let's reframe the question to ask where there were Blue Shifts and where there were Red Shifts.

States where there was a dramatic RED shift



Reason: State law allowed mail ballots to be counted before election day. Mail ballots favored Biden, and these were reported as soon as polls closed. Significant Blue Mirage. It looked to many (including me) that Ohio was going to go to Biden. But that was a Blue Mirage.

There were States where there was a more ephemeral Blue Mirage, such as Kansas.

Reason. Ditto



States with Blue Shift

Reason: Mail ballots were not allowed to be counted before election day. Mail Ballots favored Biden. They are labor intensive to count, so take a long time to report.

Very early on, Dems were getting nervous about Virginia but Virginia counted mail ballots quicker than other states, so the Blue Shift happened relatively early on, on election night.

There's also a distinction between processing the mail ballots and counting them. (In some states there was also a distinction between mail ballots and absentee ballots.)

Minnesota wasn't able to start counting mail ballots until the polls closed but counted them relatively quickly because the mail ballots could be processed starting Oct. 27.

Virginia: The date processing was allowed to begin is "unspecified" in my source, but they could be counted before polls closed..

Pennsylvania was able to start counting when polls opened but counted very slowly due to the fact that the ballots were not allowed to be processed before that time. So Pennsylvania ended up being the state that called the election.





List of when mail ballots could be processed/counted:

https://ballotpedia.org/When_states_can_begin_processing_and_counting_absentee/mail-in_ballots,_2020

 

[Rory]

Matt Parker has posted a new video looking into Republican claims - shared by Fox News and included in the lawsuit - that the chancs of Biden turning around the deficit in Georgia, etc was "one in quadrillion":

Source: https://www.youtube.com/watch?v=ua5aOFi-DKs&ab_channel=Stand-upMaths

Red mirage and blue shift are mentioned, with a nice basketball analogy to explain the mistake of treating the two populations the same.

 

[Z.W. Wolf]

I thought they were treating the vote as a purely random sequence - like a series of coin flips. Ridiculous. We've all been through election reporting before... waiting for various wards/districts/cities to report. This district is heavily Republican... this one traditionally votes Democratic...

But... it's even more ridiculous than that. The analysis assumes that the election day vote population is biased exactly the same way as the mail ballot vote population. As if the mail ballot vote would also be biased toward Trump. This is fraud; there's no other reasonable conclusion.

And he starts with the same point I made. Why do we have to have a statistical analysis - a guess about the odds of something happening - when we have the primary records?

It's like having video of a traffic accident.. then having someone trying to prove that it didn't happen with a statistical analysis.

Just look at the video!

 

[Mendel]

And he starts with the same point I made. Why do we have to have a statistical analysis - a guess about the odds of something happening - when we have the primary records?

That's the problem in a nutshell.

The Trumpers thought that Trump would win the election again, based on the fact that Trump had those big rallies and Biden didn't, that Biden had been talked down to them ("senile"), that they thought Trump was doing a good job, god was on his side, and QAnon as well.

The "statistical" analysis is a way to invoke mathematical authority to say, "here's what we thought ought to have happened", to feel justified in having an expectation for something that did not happen.

The "righteous" lost, and that's always hard to bear, especially if you've put your trust in a leader who told you he couldn't lose.
(Of course, there's also a "righteous" fringe on the left, but they're more accustomed to losing elections.)

For a Trumper, to understand why Trump lost, it may be more helpful to spend an hour on Youtube looking at Project Lincoln videos, produced by Republicans who don't support Trump, rather than thinking about statistics.

 

[Mendel]

There are no "statistical anomalies" in the vote shifts in the "battleground" states if you compare them to other states:

Article:

What we should do, though, is consider the broader context for Navarro’s claims. He focuses on the six states that have been targeted by Trump since the election. We’ve labeled them as the “irregularity” states below, to translate them to Navarro’s document, but one could also call them “swing states” — except that Michigan wasn’t really all that close.
Anyway, the point is that, if these were states where something demonstrably unusual happened, if there was some exceptional fraud at play, they would look different from other states. But, as we’ve repeatedly demonstrated, they don’t.

But there’s nothing exceptional about the “swing states.” There’s no obvious pattern showing that particular places overdelivered for Biden in any particular way.

Source: https://www.washingtonpost.com/politics/2020/12/18/this-might-be-most-embarrassing-document-created-by-white-house-staffer/

 

[Jeff Semenak]

That's the problem in a nutshell.

The Trumpers thought that Trump would win the election again, based on the fact that Trump had those big rallies and Biden didn't, that Biden had been talked down to them ("senile"), that they thought Trump was doing a good job, god was on his side, and QAnon as well.

The "statistical" analysis is a way to invoke mathematical authority to say, "here's what we thought ought to have happened", to feel justified in having an expectation for something that did not happen.

The "righteous" lost, and that's always hard to bear, especially if you've put your trust in a leader who told you he couldn't lose.
(Of course, there's also a "righteous" fringe on the left, but they're more accustomed to losing elections.)

For a Trumper, to understand why Trump lost, it may be more helpful to spend an hour on Youtube looking at Project Lincoln videos, produced by Republicans who don't support Trump, rather than thinking about statistics.

Why, would you think those 2 groups, share any commonalities? So, a Trumper (Your words) should listen to someone opposed to their Candidate for advice? Why, would you think that? President Trumps supporters, wouldn't listen to the Democrats either. Did large numbers of Democrats, ask the Trump Campaign for clarity on their choice?

 

[Mendel]

So, a Trumper (Your words) should listen to someone opposed to their Candidate for advice?

Not for advice, no.

But if they want to learn why there were Republicans who didn't vote for Trump without it being fraud, it'd be smart to listen to those Republicans, right? It would explain why Biden got more votes than they expected. It's research that leads to a good answer.

 

[Mendel]

Running past my feed today, I'm posting it here because I was advising going to the Lincoln Project to find out why conservatives were voting Biden:

 

[deirdre]

"Battle not with with monsters, lest you become a monster"

It's so sad that a temporary President can manipulate so many (on both sides) to become just like him.

 

[Z.W. Wolf]

I disagree. I moved my reason for disagreeing over to the thread where it's on topic.

Here: https://www.metabunk.org/threads/de...-change-the-election.11494/page-2#post-243696