Saturday, January 09, 2021

Voter Fraud Conspiracy Nonsense Debunking Roundup

There have been many claims of voter fraud or other misconduct in the 2020 election.  Most of these claims simply do not stand up to scrutiny, and most of the people passing on these claims seem not to be checking their accuracy.  A catalogue of unverified allegations appears at https://hereistheevidence.com/.  I have been trying to check these allegations, and here is what I have found about some common claims.  Please click the links for more detailed explanations.

VOTING MACHINES

Did Vote Counting Machines Flip Votes from Trump to Biden?

There are all sorts of wild claims about Dominion Voting Systems, from who owns it to how its program in written.  It is not necessary to analyze the machines themselves to know whether they produced an accurate count.  The states that Trump contested use paper ballots, which can be recounted if there is any question about the machine vote count.  The Trump campaign asked for recounts in Milwaukee and Dane County in Wisconsin, which use a different vendor.  The Wisconsin recounts found no vote-flipping.  In Georgia, a hand recount was conducted automatically; it was not requested by the Trump campaign.  The recount found some human error; it did not find any vote-flipping.  In Michigan, Trump improved from 2016 in counties that use Dominion machines, and declined in counties that use other vendors.

Did Dominion Flip Votes in Antrim County?

There was an error in the initially reported result in Antrim County, MI, which the clerk has attributed to a problem with updating the software.  The magnitude of this error makes it unlikely that fraud could explain it.  A hand recount verified that the corrected machine count was accurate.

What about Hammer and Scorecard?

That story is a hoax.

ELECTION DATA

Anomalies in the reported election results could provide evidence of fraud, or at least suggest areas that merit further investigation.  However, many such claims don't stand up to scrutiny.

Does High Voter Turnout Prove Voter Fraud?

Voter turnout was high, but not unreasonably so.  Reasons for high turnout include population growth, more absentee/mail voting, and intense support and opposition to Trump.  Turnout was up in every state, and many safe states had big turnout increases.

Turnout in Detroit was over 100%!

President Trump claimed that Detroit had "far more votes than people".  Russell Ramsland claimed that turnout in Detroit was 139%.  The turnout was actually 51%.  Similar claims about other jurisdictions are also false.

Did Trump lose due to surges of turnout in Detroit, Philadelphia, and other big cities?

Biden got 1000 FEWER votes than Hillary did in 2016 and Trump got 5000 MORE votes than he got four years ago in Detroit.  Biden's margin in Philadelphia also declined compared to Hillary in 2016.  Biden won by improving his margins in upscale suburbs in Michigan and Pennsylvania.  Biden did improve in other cities like Atlanta and Milwaukee, but the improvements were not only in swing states, as some claimed.

Do US Senate election results prove tens of thousands of Biden-only ballots?

It is perfectly normal for some voters to only vote in the most high-profile race.  This has often happened in the past, and it happened more in Michigan in the past than in 2020.

How could Trump lose when he improved with minorities?

There is solid statistical evidence that Trump had a small improvement with blacks and a significant improvement with Hispanics.  Trump lost because he lost ground with white voters, particularly college-educated voters.  With all the discussion of minority voters, it is easy to forget that the majority has more votes than the minority.

Did Battleground States Stop Counting Votes Before the Results Swung to Biden?

North Carolina (which Trump won) stopped counting because it ran out of votes to count.  Some localities did break for the night.  Michigan, Wisconsin, and Philadelphia did not stop counting.

Why were the late election results bad for Trump?

In many states, including Michigan, Pennsylvania, Georgia, and Wisconsin, Trump led early, but Biden gained until winning the election later.  There is a clear reason for this.  These states all counted election day votes first and absentee votes later.  Rs disproportionately voted on election day due to concerns about the security of mailed votes, while Ds disproportionately voted absentee due to concerns about COVID.  Thus it is no surprise that Trump won election day votes while Biden won absentee votes.  Note that Ohio counted absentee votes first, and Trump trailed early but won late.

STATISTICAL ANALYSES

Charles Cicchetti found the chance of Biden winning Pennsylvania, Michigan, Georgia, Wisconsin after Trump's early lead is less than one in a quadrillion!

For this calculation to be correct, votes would have to be counted in random order, so that a vote counted late was no more likely to be for Biden than a vote counted early.  However, this is false, since in these four states, absentee votes were counted after election day votes.  Biden did better in absentee votes than in election day votes.  Matt Parker and Robert VerBruggen have longer explanations.

Does Benford's Law prove that Biden's vote totals were fraudulent?

Benford's Law is a tool that can provide evidence of fraud in some data sets.  Some precinct data seems to violate this law.  To be applicable, the data set must follow a "power law", and must vary over several orders of magnitude.  Neither assumption is likely to hold for election data, so Benford's Law does not apply.  Matt Parker has a longer explanation.  Also, it is unlikely that Benford's law would catch most types of voter fraud, unless someone were making up vote totals from scratch.

Has Dr. Shiva Proven Michigan Voter Fraud?

Dr. Shiva made a serious mathematical error in his analysis and his conclusion is based on a false assumption.  The effect he found for Trump votes applies similarly to Biden votes.

"Mathematician" Bobby Piton was kicked off of Twitter after presenting evidence of fraud in Arizona!

It isn't clear whether Piton was ever kicked off Twitter, but he is there now.  Piton is not a mathematician, he is a financial adviser.  He describes his approach on his website:

Bobby has read well in excess of a million pages over his career and has extensively studied physics, quantum mechanics, mathematics, economics, trading, portfolio construction, model development, asset valuation, and alpha generation to develop and refine his methodology.

Piton's presentation in Arizona is on Youtube and the associated write-up is on his website.  Good luck following his argument.  He does compare (19:30) the turnout for the 1998 (governor) election to the 2020 presidential election.  Midterm elections almost always have lower turnout than presidential elections, so this is not a relevant comparison.

ISOLATED CLAIMS

Many election observers have signed affidavits concerning their experiences on election night.  It is often noted that these are sworn under penalty of perjury, and therefore they are proof of fraud.  Realistically, it is unlikely that anyone would be charged with perjury, which would require proving beyond a reasonable doubt that they are lying.  That is not to say that most of them are lying.  Some are clearly just misunderstandings, many don't actually allege fraud, and many are impossible to check.  Some specific claims are addressed below.  Recall that Trump improved in Detroit and Philadelphia relative to 2016, which is hard to square with massive fraud in those cities.

Detroit's election workers input a false birthdate of 1/1/1900 for absentee ballots!

Yes, they did.  Detroit's election software required a birthdate to input the ballots, which election workers would not have.  The city clerk instructed workers to input that date as a placeholder so that the real birthdates could be filled in later.  Note that it would not make sense for anyone committing fraud to claim that a voter was 119 years old!

Election workers ran the same stack of ballots through the machine multiple times!

Each precinct has a list of everyone who voted, called a poll book.  If many ballots are run through a machine repeatedly, this will create a mismatch between the number of votes and the number of recorded voters in the poll book.  While many of Detroit's precincts did have mismatches, they were generally small, not hundreds or thousands of votes.

Almost half or 48% of the 134 counting boards were individually off by plus or minus four votes or fewer each, according to results certified by the county's bipartisan board of canvassers. Another 39 boards, or 29% of them, were in balance, while 31 boards, or 23% of the total, were off by five or more votes.

The total "difference in absentee ballots tabulated and names in poll books in Detroit was 150".

In Georgia, were there problems with secret vote counting, illegal voters, signature matching, etc?

Gabriel Sterling, a Republican who is Georgia's voting system implementation manager, debunked these claims in a press conference.

HISTORICAL COMPARISONS

While these do not directly prove voter fraud, they are supposed to show that something strange happened.  In many cases, there are perfectly reasonable explanations.

Was There are Red Wave Downballot in 2020?

The election results for president and congress were quite consistent.  Rs picked up house seats relative to 2018, not relative to 2016.

Do Incumbent Presidents Who Increase Their Votes Always Win?

There are few elections where incumbent presidents lose, so there are few relevant elections to consider.  Also, this anomaly has happened before.

Can a President Lose While His Party Gains House Seats?

This is another claim which applies to only a few elections.  It has also happened before.

How could Trump lose when he won most bellwether counties?

A bellwether county is a county which always, or almost almost always voted for the winner in recent elections.  They are not crystal balls, they are statistical anomalies--counties that happen to swing the same way as the nation when party coalitions change.  These counties are mostly rural, and Trump did well in them in both 2016 and 2020.  Biden won by improving in large suburban counties, not rural areas.

ELECTION LAWS AND LAWSUITS

The Trump campaign and supporters have filed around 60 lawsuits challenging the election results.  They have lost all of them, with one minor exception.

Did any judges rule on evidence of voter fraud, or were all the suits dismissed for lack of standing?

The suits not filed by the Trump campaign were generally dismissed for lack of standing.  This is something that the plaintiffs should have considered before filing, as not just anyone can challenge the results of an election.  However, some judges did rule on the merits of the claims.  Note that some of the more wild claims of fraud were not actually made in court, where lawyers can be disbarred for lying.

In Wisconsin, a Trump-appointed judge allowed the Trump campaign to present evidence.  However, they declined to present any evidence ("stipulated [the] set of facts") and only contested election law issues.  A Republican judge in Pennsylvania dismissed a Trump campaign lawsuit, saying "this Court has been presented with strained legal arguments without merit and speculative accusations, unpled in the operative complaint and unsupported by evidence."  The Nevada Supreme Court (which is ideologically mixed), similarly upheld the dismissal of a Trump campaign lawsuit on the merits.  In Michigan, the Trump campaign dropped its lawsuit before its claims could be heard.

Were election laws changed before the election?

Some legislatures and judges changed laws before the election.  The reason cited was generally the COVID pandemic.  It is likely that some judges were hoping to help Ds with their rulings.  This help could mean either encouraging more Ds to legally vote, or making it easier for fraud to be committed.  Some of these decisions were challenged before the election, and some were overturned, but others were not challenged before the election.

After the election, the Trump campaign filed challenges to many of these decisions.  Their complaints in Pennsylvania and Wisconsin seemed to have some merit.  The problem with this is that there is no way to know whether these decisions changed the outcome of the election.  So if the judges' decisions were illegitimate, what is the remedy?  The Trump campaign was asking for hundreds of thousands of votes to be thrown out, or even the entire election.  But this would disenfranchise millions of legitimate voters, including those who voted in good faith based on the rules presented at the time.  This would be a bigger injustice than the original decisions.  Many rulings have cited the doctrine of "laches" to toss out Trump campaign lawsuits.  Essentially, this means they waited too long to file.  The issues could have been resolved before the election with far less disruption, so waiting until after the election to file is too late.

Can state legislatures appoint the electors themselves?

The Constitution gives states the power to decide how presidential electors are chosen. Since the early 1800s, all states have passed laws to choose presidential electors by popular election.  A state legislature could theoretically repeal this law and choose the electors itself for a future election, but none have tried to do so since this would be very unpopular.  A legislature cannot unilaterally appoint its own slate of electors in defiance of the laws that are already on the books.  It is very dubious whether it could repeal the law for an election that has already taken place.  In any case, no legislature attempted to appoint its own electors.

Why was the Texas lawsuit thrown out by the Supreme Court?

In the Texas lawsuit, a group of states was attempting to get the Supreme Court to change their election laws.  This is a violation of federalism, as states can make and interpret their own laws unless there is a violation of federal law involved.  The case was basically a political stunt, and all nine justices viewed it as such.

ABSENTEE BALLOTS

The COVID pandemic led to a surge of absentee/mail voting.  Absentee voting can be more vulnerable to fraud, since the person who votes is not actually seen casting the vote.  However, we have a list of the people who voted, or at least the name they gave, so if there were many ineligible or fake voters, it should be possible to identify them.

Were there many dead people, minors, felons, illegals among the voters?

This has been claimed by Matt Braynard of the "Voter Integrity Project".  It isn't entirely clear what methodology he used to conclude this, but it seems likely that he compared lists of voters to lists of people who are prohibited from voting and looked for matching names.  The problem with this is that there are many people with the same name, so this will lead to many false matches.  This seems to be the case in Georgia, where a D state representative was able to discredit many of the names cited by Braynard.  The voter lists should certainly be investigated by authorities, but claims of hundreds of thousands of illegal votes should not be endorsed without proof.  A Georgia audit of 15000 signatures found only two with problems.

Did Steven Crowder find Massive Voter Fraud in Detroit?

Crowder simply misunderstood the way that Detroit reports absentee ballots.

PLAUSIBLE ARGUMENTS FOR VOTER FRAUD

There are a few arguments for voter fraud that seem plausible to me.  That does not mean that they are definitely true, but at least I don't see any obvious flaw in them.

Were there differences in the absentee ballot voting rates on the borders of counties?

John Lott argues that there are differences in the absentee ballot voting rates in precincts on opposite sides of county lines, where one of the counties is an urban center controlled by Ds.  He says that differences do not exist for in-person voting, and there was no such difference in 2016.  This analysis could be vulnerable to cherry-picking, so it would be good to see it checked with a larger data set.

Did more than 1700 Georgians vote twice in 2020 elections?

It appears so.  These voters apparently voted both in person and by absentee ballot.  It is important to recognize, however, that the number 1700 combines multiple elections.  For the November election, the total claimed is "at least 400".

Thursday, December 31, 2020

January 2021 Judiciary News

Leave no vacancy unfilled (until January 20).

Nominations, Hearings, Confirmations:

Post-election nominations:  Ed Whelan refutes the claim that there is a long tradition of not confirming nominees of an outgoing president.  The last time an outgoing president had a senate run by his party was 1980, when Jimmy Carter nominated Stephen Breyer to the 1st Circuit after the election.

Obstruction:  Carrie Severino reviews the tactics D senators used to try to obstruct President Trump's judicial nominees.  R senators should keep this in mind.

Durbin:  Senator Dick Durbin will become the Democrat on the Senate Judiciary committee after Senate Ds passed a rule allowing him to hold the position and remain D whip.  Some progressive groups preferred Sheldon Whitehouse get the job.

Biden:  This article on Biden's approach to judicial nominations says that many D appointed judges are planning to take senior status next year, but the Georgia runoffs may affect their timing.  Several R appointed judges confirm that they were approached about retiring under Trump.  John Jones (MD-PA), Michael Kanne (7th Circuit), and Leslie Southwick (5th Circuit) all said that they have no immediate plans to retire, but would not rule out retiring under a D president.

Biden:  Various interest groups are demanding more 'diversity' among Biden's judicial nominees.  Not ideological diversity, obviously.

Biden:  Harsh Voruganti of the Vetting Room identifies four judges as possible Biden Supreme Court nominees.  They are Ketanji Brown Jackson (D-DC), Leondra Kruger (California Supreme Court), Sri Srinivasan (DC Circuit), and Paul Watford (9th Circuit). 

Senate Judiciary Committee hearings:
TBD

Confirmations:
TBD

The Federal Judiciary:

Supreme Court:  The Supreme Court unanimously dismissed a lawsuit by the Texas Attorney General attempting to overturn the election results in four states.  Andrew McCarthy explains why the suit was frivolous.

Trump:  Conservative judges, including those appointed by President Trump, have rejected his post-election lawsuits.  This refutes the claims that they are partisans who rule for policies they personally prefer.

Alito:  Justice Alito recently gave a speech to the Federalist Society.  He criticized the court's lack of vigilance in protecting religious liberty and the Second Amendment.  He implicitly criticized Chief Justice Roberts on those issues.

Breyer:  Justice Stephen Breyer says he will retire "eventually".

Federal Circuit:  The Federal Circuit never got a Trump nominee.  This circuit is less partisan, as it deals with patent law and some suits against the government.  The article is mainly concerned with the race of potential nominees, not their qualifications.

2nd Circuit:  Senior Judge Ralph Winter died on December 8 at age 85.  Reagan appointed him to the 2nd Circuit in 1981.  He took senior status in 2000.

5th Circuit:  Senior Judge Thomas Morrow Reavley died on December 1 at age 99.  He was on the Texas Supreme Court (1968-1977), and Carter appointed him to the 5th Circuit in 1979.  He took senior status in 1990.  He married fellow 5th Circuit judge and Carter appointee Carolyn Dineen King in 2004.

5th Circuit:  Trump's three appointees to the 5th Circuit representing Texas, Don Willett, James Ho, and Andy Oldham, are among the most conservative judges on the appeals courts.

Senior Status declarations:
WD-AR: Paul K. Holmes III (Obama) 11/10

State Supreme Courts:

Alaska:  Alaska Supreme Court Chief Justice Joel Bolger will retire in June 2021.  His replacement will be appointed by Governor Mike Dunleavy.  In July 2020, he appointed Dario Borghesan, who is (apparently) the only conservative on the five-member court.

Georgia:  Governor Brian Kemp announced the appointment of Fulton County Superior Court Judge Shawn Ellen LaGrua to serve on the Supreme Court of Georgia on December 1.  She was previously Inspector General for the Georgia Secretary of State’s Office, DeKalb County Solicitor General, and a prosecutor in Atlanta.  She replaces Justice Keith Blackwell, who retired in November.

Illinois:  Judge Robert Carter of the 13th Judicial Circuit Court was selected by the Justices of the Illinois Supreme Court to fill the district 3 seat of Justice Thomas Kilbride, who lost a retention election.  Carter was elected as a D.  He is 74, and has pledged not to seek election in 2022.

New Mexico:  Governor Michelle Lujan Grisham appointed Julie Vargas to the Supreme Court, filling the seat of Justice Judith Nakamura, who retired.  Vargas was a judge of the New Mexico Court of Appeals since her election in 2016.

North Carolina:  Supreme Court Chief Justice Cheri Beasley conceded the election to Republican Justice Paul Newby.  Rs won all three races in 2020.  The court's breakdown is now 4 D, 3 R.

Ohio:  R Ohio Supreme Court Justices Sharon Kennedy and Pat DeWine are considering running for Chief Justice.  Chief Justice Maureen O’Connor is age-limited in 2022.  Kennedy's term ends in 2026, while DeWine's term ends in 2022. D justices Michael Donnelly and Jennifer Brunner are rumored to be interested in the seat.

Rhode Island:  Governor Gina Raimondo named Melissa Long and senator Erin Lynch Prata to the Rhode Island Supreme Court.  Long is a Superior Court Judge since 2017, and was previously deputy secretary of state.  Prata's nomination was controversial due to a state ethics law that requires a one year hiatus before a sitting legislator can take a state job.  However, the state Ethics Commission voted 5-2, against the advice of its lawyers, to allow Lynch Prata to apply without waiting a year.  Long will succeed Supreme Court Justice Francis Flaherty, who retired December 31, and Prata will succeed Justice Gilbert Indeglia, who retired in June.  The Rhode Island Supreme Court had all R appointees, despite not having an R governor since 2010.

Texas:  Governor Greg Abbott appointed Houston judge Jesse McClure to the Texas Court of Criminal Appeals, where he will be the only black judge.  McClure was appointed to his current job in 2019 but lost election in 2020.  He replaces Judge Michael Keasler, age 78, who is age-limited.

Wisconsin:  Justice Brian Hagedorn of the Wisconsin Supreme Court has ruled differently than his three R colleagues on several high-profile issues.  Depending who you ask, he is either following the law in contrast to his more partisan colleagues, or engaged in a Roberts-style effort to appear above the fray.

Numbers and Trivia:

Party Line Votes:  Only 17 of President Trump's 54 appeals court judges (31%) got any support from Ds on the Senate Judiciary Committee.  Perhaps President Biden's nominees will receive similar treatment from Rs on the Senate Judiciary Committee.

Chief Judges:  The Presidents who appointed chief judges of the 13 appeals courts are Clinton (4, 6, 9), W (1, 2, 3, 5, 7, 8, 10, 11, Fed), and Obama (DC).  There are four chief judges that will change in 2021.  They are expected to be
6th Circuit (May 23) R. Guy Cole (Clinton) -> Jeffrey Sutton (W)
Federal Circuit (May 24) Sharon Proust (W) -> Kimberly Ann Moore (W)
9th Circuit (August 14) Sidney Thomas (Clinton) -> Mary Murguia (Obama)
3rd Circuit (December 4) Brooks Smith (W) -> Michael Chagares (W)
There could be more, and the dates could be sooner, if any chief judge steps down early.

History:

Senior status:  Harsh Voruganti of the Vetting Room reviews the history of taking senior status upon confirmation of a successor, an increasingly common practice. 

Resources:
Bench Memos (National Review)
The Vetting Room

Wednesday, December 23, 2020

Did Dominion Flip Votes in Antrim County?

Since election night 2020, controversy has swirled around Antrim County in northwest Lower Peninsula.  The reported vote totals showed Joe Biden handily winning the county, and many people (including me) immediately realized that this could not be correct.  The county clerk retracted the results and ordered a retabulation, which produced a reasonable result.  (Trump's margin declined by 233 votes compared to 2016.)

This launched a conspiracy theory about Dominion Voting Systems, a manufacturer of voting machines and vote counting machines.  Antrim County Clerk Sheryl Guy (R) has explained the problem as follows:
Guy has said Biden appeared to be winning in early, unofficial county results because of a problem that developed when she attempted to update Election Source software on isolated tabulators. Because the update was not applied to all tabulators, results were transposed as they were pulled from the tabulators into the county's main voting software.
I don't understand exactly how this happened, but this sounds like the sort of thing that could happen if a program is not written to account for user errors.

There are all sorts of wild claims about Dominion, from who owns it to how its program in written.  I won't try to evaluate these claims.  Perhaps their software is badly designed or has vulnerabilities.  Even if this were the case, it would not mean that the count for this election were necessarily wrong.  Fortunately, it is not necessary to analyze the machines themselves to know whether they produced an accurate count.

It doesn't make much sense to claim that the initial vote count was the result of vote-flipping.  If the expected margin is off by 40% or so, this will certainly attract attention, and likely lead to a recount, which is what happened.  Vote-flipping could only change the margins by a single digit percentage and remain plausible.

Furthermore, the vote totals in Michigan do not support the claim that Dominion was flipping votes.  I analyzed Michigan' vote totals by county, comparing counties that use Dominion to those that don't.  I found that Trump actually improved from 2016 in counties that use Dominion machines and declined in counties that don't use Dominion machines.  The data provide no evidence to support the theory that Dominion machines flipped votes from Trump to Biden.

A report by Russell Ramsland of the "Allied Security Operations Group" claims that Dominion machines are "intentionally and purposefully designed with inherent errors to create systemic fraud and influence election results."  The report was commissioned due to a (supposedly) unrelated lawsuit over a local marijuana ordinance.  Judge Kevin Elsenheimer, an R former state Republican rep, ordered the release of the report.

The report was disputed by several experts.  One is Christopher Krebs, the first director of the Cybersecurity and Infrastructure Security Agency.  Krebs was appointed by President Trump in 2018, and was fired after he disputed Trump's claims of fraud.  He said that ASOG "misinterpreted what it saw in the computer coding and used it to "spin" allegations that voting machines couldn't be trusted."
"I'm seeing these reports that are factually inaccurate continue to be promoted. That's what rumor control is all about. That's what I'm continuing to do today, based on my experience and understanding and how the systems work," Krebs said. "We have to stop this. It's undermining confidence in democracy."
I won't try to evaluate the report itself, since it is outside my area of expertise.  However, I can consider the credibility of the person who wrote the report.  Russell Ramsland has previously filed affidavits in lawsuits alleging fraud in the 2020 presidential election.  One affidavit claimed that various townships in Michigan had over 100% turnout.  The townships were actually in Minnesota, not Michigan, and turnout was not over 100%.  Another affidavit claims that turnout in detroit was 139%, when it was actually 51%.  It also claims false turnout rates for many other jurisdictions in Michigan.

All the controversy led the Michigan Bureau of Elections and Antrim County to conduct a hand recount in Antrim County.  It found a 12 vote gain for Trump, which is well within the range of error that you would expect when a few people don't mark the ovals clearly.  Thus the corrected vote count was accurate.  There was no vote-flipping in Antrim County.

This raises the question of why the Trump Campaign never asked for a recount in Antrim County.  Indeed, they never asked for a hand recount anywhere that uses Dominion machines.  They did ask for recounts in Milwaukee and Dane County in Wisconsin, which use a different vendor.  The Wisconsin recounts found no vote-flipping.  In Georgia, a hand recount was conducted automatically; it was not requested by the Trump campaign.  The recount found some human error; it did not find any vote-flipping.

So why didn't the Trump campaign request any hand recounts in jurisdictions that use Dominion vote counting machines?  The inescapable conclusion is that they don't believe that vote-flipping happened.  Is it possible that the Trump campaign would like people to believe that the election was stolen, and it doesn't want to produce evidence contrary to that belief?

Wednesday, December 16, 2020

Can a President Lose While His Party Gains House Seats?

Steve Deace recently claimed (1:32) that in the past 100 years before 2020, only one incumbent president lost reelection while his party gained house seats.  This claim is true, but highly misleading.  Only four incumbent presidents have lost since 1916--Hoover, Carter, HW Bush, and Trump.  (Ford, while technically an incumbent, was never elected.)  In 2 of those 4 elections (1992 and 2020), the president's party gained house seats.  The 1932 and 1980 elections were landslides, so it is not surprising that the president's party lost house seats then.  (In 1976, the Ds gained just 1 house seat, so this doesn't add much to the pattern.)

So the thing that almost never happens actually happened 2 out of 4 times!  Something that happens half the time is not surprising!  This is the sort of factoid that is cited as evidence of election fraud, but doesn't prove anything.  It sounds impressive due to the length of time (100 years!) but there are actually only 4 relevant elections, so the denominator of the fraction is small.

It should also be noted that whether a party gains or loses house seats depends greatly on the previous (midterm) election.  In 2018, Rs lost 42 house seats, so it is not surprising that Rs gained some in 2020.

Sunday, December 13, 2020

Was There are Red Wave Downballot in 2020?

Some people are arguing that the 2020 election results were a "red wave" aside from the presidential election, and this indicates voter fraud in the presidential election.

Even if this were true, it would not prove voter fraud.  It could just be that some segment of normally R voters refused to vote for the president, but still voted R downballot.

But was 2020 a red wave?  In the senate, Rs lost two seats (AZ, CO) and picked up one (AL) for a net loss of one (pending the outcome of the Georgia runoffs).  The results of the senate races and the presidential race were the same in all but one state.  The one exception was Maine, where Susan Collins is a popular moderate incumbent who regularly outperforms the top of the ticket.

Comparing R senate candidates' margins to Trump doesn't reveal an obvious pattern.  Trump did better in some states, while R senate candidates did better in others.  Patrick Ruffini made a map illustrating the differences.

In particular:

Trump did better than R incumbents in: AZ, ID, IA, KY, MS, MT, OK, SC
R incumbents did better than Trump in: AK, AR, CO, GA, LA, ME, NE, NC, SD, TX, WV
Trump did better than R challengers in: AL, DE, KS, MA, NH, NJ, OR, RI, VA
R challengers did better than Trump in: IL, MI, MN, NM, TN, WY

In the US house of representatives, Rs picked up 14 seats (CA-21, CA-25, CA-39, CA-45, FL-26, FL-27, IA-1, IA-2, MN-7, NM-2, NY-11, OK-5, SC-1, UT-4), and lost 3 seats (GA-7, NC-2, NC-6), none incumbents, for a net gain of 11 seats (with NY-22 still undecided).  It appears that Biden won 224 seats and congressional Ds won 222 seats.  There were some Trump/D seats and some Biden/R seats.  Left Coast Libertarian at RRH Elections notes that:

Joe Biden got 80,048,633 votes. Congressional Democrats got 76,298,374.

Donald Trump got 73,902,347 votes and Republicans got 72,542,282.

So Democrats got 95% of Biden’s vote total and Republicans got 98% of Trump’s vote total. If we average them Democrats underperformed by about 1.1 million votes. That’s a really small number of votes. It shouldn’t be unexpected that one party would do slightly better.

So how is it possible that Rs picked up seats?  Rs picked up seats RELATIVE TO 2018.  In 2018, Rs lost 42 seats from 2016.  Thus Rs lost house seats since 2016, the last time Trump was on the ballot.  Note that all but two seats that Rs picked up in 2020 were seats that Rs lost in 2018.

Some have noted that Rs won all tossup house races (according to the New York Times).  But this just indicates that the election raters did a bad job, likely due to relying on bad polls.  The conservative-leaning RRH Elections did much better, though still leaning too D.  My own ratings of Michigan elections were very good, as I did not rely much on polls.

Rs picked up two state house chambers, both in New Hampshire.  This was apparently due to the popularity of incumbent R governor Chris Sununu, who campaigned heavily for R state legislative candidates.  Rs flipped a modest number of state legislative seats, further picking up rural seats in states like KY and WV, and flipping back some seats that were lost in 2018.

Overall, the 2020 election results were slightly worse than in 2016, but significantly better than in 2018.  There was no inconsistency between the presidential election and downballot results in 2020.

Thursday, December 10, 2020

Do Incumbent Presidents Who Increase Their Votes Always Win?

One claim going around is that for the past 150 years, no incumbent president has increased his raw vote total and lost (and this suggests voter fraud).

In the past 150 years, there were 38 presidential elections.  In 19 of those elections, there was no president running for reelection (an incumbent who was not elected cannot run for reelection).  Since the claim is only about incumbent presidents, we have to discard those.

That leaves 19 elections with an incumbent running for reelection.  Incumbent presidents won 12 times.  Since the claim is about incumbents who lost reelection, we have to discard these also.

In the past 150 years, only seven incumbent presidents have lost reelection in the general election.  (This does not count Gerald Ford, who was not elected initially.)  Presidents Taft, Hoover, Carter, and George HW Bush all lost in landslides.  The claim is only really relevant to incumbent presidents who lost narrowly, which includes Grover Cleveland, Benjamin Harrison and Trump.  A trend that only holds for three elections isn't much of a trend.

There is one more problem.  THE CLAIM IS FALSE.  Grover Cleveland got 4,914,482 votes in 1884, when he won, and 5,534,488 votes in 1888, when he lost.  Sadly, it seems that people don't bother to check claims like this before passing them on.

Tuesday, December 08, 2020

Polling: What Went Wrong?

People who want to know the future have many tools.  There are crystal balls, fortune cookies, horoscopes, and just a bit more credible, public opinion polls.  The 2020 election featured a spectacular failure of the polling industry.

RealClearPolitics catalogues all major (non-candidate) polls.  They also average the results of all recent polls, which effectively increases the sample size and reduces the margin of error.  Their national polling average was Biden +7.2, while the popular vote was Biden +4.5, an error of 2.7%.  The results in many state polls were even worse.  In Ohio, the error favored Biden by 7.2%.  In Iowa, it was 6.2%.  In Wisconsin, it was 6%.  Overall, Trump outperformed the polls in 33 states.  Curiously, Biden did overperform the polls in a few states, including Minnesota.

A review of how polling works is in order.  It is neither possible nor practical to ask everyone in a population their opinion.  Thus a pollster seeks a sample of the population and tries to infer the views of the whole population based on the sample.  For this to work, the sample should be randomly selected, that is, every possible sample should be equally likely to be picked.  A randomly selected sample is unlikely to match the population exactly, but there are well-understood mathematical laws that describe how far from correct the results are likely to be.  This kind of error is known as sampling error--error caused by a sample not matching a population.  The margin of error is the margin on either side of the estimate that we can be 95% confident contains the true value.

In practice, however, the sample a pollster obtains is not random.  Nobody can be forced to participate in a poll, and if even one person declines, the sample is not random.  It used to be the case that most people answered the phone and talked to pollsters.  But over time, response rates declined due to telemarketers, robocalls, answering machines, caller ID, and cell phones.  Now response rates typically range between 1% and 5% of people called.  The sample a pollster obtains is usually wildly unrepresentative of the population.

How do pollsters deal with this?  They ask respondents various demographic information (race, sex, political party, education).  Then they weight the results so they match the presumed demographic breakdown of the electorate.  But they don't actually know this breakdown.  Pollsters make an educated guess based on demographics of past elections (which can be estimated, but not known exactly) and their beliefs about what the electorate will look like.

Essentially this makes the poll itself an educated guess.  Educated guesses are often close to accurate, and they are more accurate than the sort of wishful thinking that predominates among political ideologues.  But educated guesses can be wrong, sometimes wildly so.

This cycle, it appears that many Trump supporters didn't answer their phones, or refused to participate in polls.  This skewed the samples, even with the adjustments that pollsters made.  But why did Trump supporters refuse to talk to pollsters?  Some may simply hate the media.  Others may be concerned about admitting their views, even in a supposedly anonymous poll.

This phenomenon is known as social desirability bias.  One previous example of this is the Bradley effect, in which voters were supposedly more likely to say they would vote for a black candidate than to actually do so.  This effect remains controversial, however.

Researchers try to account for social desirability bias in various ways.  One way is to ask people what their friends or neighbors think about the election.  This is the methodology used by Trafalgar, a pollster who found much better results for President Trump than other pollsters.  The problem of social desirability bias applies to issue polls, as well.

Polls can be useful when appropriate precautions are taken, but other indicators of public sentiment should not be ignored.

Thursday, December 03, 2020

Does High Voter Turnout Prove Voter Fraud?

One argument that there was mass voter fraud in the 2020 election is that the overall number of votes is too high.  The argument often includes the fact that Joe Biden got 80 million votes in 2020, while Barack Obama got 69 million votes in 2008.  How can this be explained?

1. Population growth.  The US population is not constant.  In 2008, there were 304 million people; in 2020 there are 331 million, according to the census bureau.  That's 27 million more people.  The popular vote went from 131 million to 159 million over the same time period.  That's an increase of 28 million votes.  Turnout was slightly higher than 2008, but not much.  Population growth almost entirely explains the change.

2. Vote by mail.  Due to COVID, many states loosened their rules and promoted voting by mail.  Some Secretaries of State (including SOS Benson) mailed absentee ballot applications to everyone.  Their goal was to increase turnout (and help Ds), and it worked.

3. Likeability.  Who do voters dislike more, Hillary Clinton or Joe Biden?  Most likely, Hillary.  Some D-leaning voters refused to vote for Hillary but decided Biden was good enough.

4.  Trump.  The election was a referendum on Trump.  Many people strongly support the president, and many strongly oppose him.  Biden was a vessel for the latter group.  Even aside from COVID, there was no reason for anyone to go to his rallies.

By the way, there is nothing unusual about this historically.  In 2000, Al Gore got 51 million votes (48.4%).  Four years later, John Kerry got 59 million votes (48.3%).  Kerry didn't have that many enthusiastic supporters, but the left hated George W. Bush.  Ralph Nader got 2.7% in 2000, but only 0.4% in 2004.  Many third party voters saw the election as a referendum on the incumbent.  It appears the same happened this year.

Presumably, proponents of the high turnout theory believe that millions of illegal votes were added for Biden.  Absentee ballot fraud seems to be the most common theory for how this could happen.  Breaking down the turnout increase by state casts further doubt on this theory.

Turnout was up in every state from 2016 to 2020.  The turnout percentage hit a 40-year high in 44 states.  Biden 2020 had a higher raw vote total over Hillary 2016 in every state that has completed its vote count.  Trump's margin declined in at least 35 states, and increased in only five (AR, FL, HI, NV, UT), with 10 left to finalize their results.  If the turnout increase was due to fraud, the fraud occurred all across the county, including in states that were not at all competitive.

I calculated the percentage increases in turnout by state (CA, IL, MS, NJ, NY not included due to lack of finalized vote totals).  The largest percentage increase was in Hawaii, up 34% from 2016.  Hawaii is a deep blue state with no competitive statewide races.  The second largest increase (32%) was in Utah, a deep red state with no competitive statewide races, and only one competitive congressional race.  The top ten percentage increases are 

HI 34%
UT 32%
AZ 30%
TX 26%
ID 26%
NV 25%
WA 23%
GA 21%
TN 21%
MT 21%

Only three of these states (AZ, NV, GA) were swing states (four if you count Texas).  The state with the smallest increase was Louisiana, which was dealing with the aftermath of a hurricane.

Were Ds conducting massive voter fraud operations in safe red states?  It seems unlikely.  Note that this does not mean that there was no absentee vote fraud, only that a turnout increase, by itself, does not prove it.

Monday, November 30, 2020

December 2020 Judiciary News

Leave no vacancy unfilled.

Nominations, Hearings, Confirmations:

Lame Duck:  Senator Feinstein has called on Republicans to stop processing judicial nominations during the lame duck session.  They declined.

Biden:  Dana Remus will be Biden’s White House counsel.  She was a clerk for Justice Alito, and supported the nominations of fellow clerks Michael Park and Andy Oldham.  A fellow clerk says that Remus “never held herself out as especially ideological”.

Biden:  President-elect Biden's team is supposedly vetting potential judicial nominees.  It isn't clear for which positions, since there are very few vacancies at present.  They are supposedly looking for candidates with "demographic diversity, but also different backgrounds", so finding candidates who fit their criteria could slow the process.

Biden:  How will senate Rs treat Biden's judicial nominees?  It is expected that there will be some resistance, but it isn't clear how much.

Feingold:  The leftist American Constitution Society, run by former Senator Russ Feingold, is sending a list of prospective judicial nominees to Biden.  It remains to be seen whether their list will be put in the shredder by Biden or McConnell.

Feinstein:  Senator Dianne Feinstein will step down as top Democrat on the Senate Judiciary committee.  She faced criticism from the left of her handling of the Kavanaugh and Barrett nominations.  It isn't clear whether she stepped down voluntarily or was forced out.

Durbin:  Senator Dick Durbin will seek to become the Democrat on the Senate Judiciary committee.  He seems likely to get the job.  Some progressive groups are unhappy, and would prefer Sheldon Whitehouse get the job.

New Nominations:
TBD

Senate Judiciary Committee hearings:
December 3 (business): Thomas Kirsch (7th Circuit) and four district court nominees are likely to be held over.

Confirmations:
TBD

The Federal Judiciary:

7th Circuit:  Judge Joel Flaum took senior status on November 30 at age 84.  He was appointed to ND-IL by Ford in 1974 and to the 7th Circuit by Reagan in 1983.  He has a moderate record.  He was the longest-serving active appeals court judge.  That honor will now go to Pauline Newman of the Federal Circuit.

9th Circuit:  Many liberal judges have been waiting for the end of Trump's presidency to take senior status.  A 9th Circuit judge says “I anticipate quite a few people doing things to enter senior status,” but “they might want to wait for a Democratic Senate, although I don’t know whether that ever will happen.”  Another judge suggests that they should make taking senior status contingent on confirmation of a successor.

D-MD:  Judge Richard Bennett of the U.S. District Court for the District of Maryland is taking senior status upon confirmation of his successor. He was appointed by George W. Bush in 2003.

State Supreme Courts:

Retention Elections: 
Alaska:  Liberal Susan Carney won 63%.
Florida: Conservative Carlos Muniz won 66%.
Illinois:  Justice Thomas Kilbride got only 56%, failing to win meet the 60% threshold.  The other justices will choose a replacement for the next two years.

Multi-candidate elections:
Illinois:  David Overstreet (R) won 63% against Judy Cates (D) for an open R-held seat.
Kentucky:  Conservative Circuit Judge Robert Conley defeated D state rep Chris Harris 55% to 45%.
Louisiana:  Judge Jay McCallum (R) won the district 4 seat with 57% over an (apparently) less conservative R.  In district 7, Judge Piper Griffin and 4th Circuit Court of Appeal Judge Terri Love (both D) head to a runoff on December 5.
Michigan:  Chief Justice Bridget Mary McCormack (D) was easily reelected with 32%.  Liberal D attorney Elizabeth Welch won the second seat with 20%.  R prosecutor Mary Kelly was third with 17%.  The court now is 4D, 3R.
Mississippi:  Conservative Justice Kenny Griffis won 51% over liberal Judge Latrice Westbrooks.  Justice Josiah Coleman was also reelected with 63%.
Montana:  R-leaning Justice Laurie McKinnon defeated trial lawyer Mike Black 57-43.
Nevada:  R-supported Eighth Judicial District Court judge Douglas Herndon beat D state rep Ozzie Fumo 47-36 (with the rest going to none of the above).
North Carolina:  Court of Appeals judge Phil Berger Jr. (R) defeated Lucy N. Inman (D) with 51%, and former state senator Tamara Barringer (R) defeated Justice Mark A. Davis (D) with 51%.  Justice Paul Martin Newby (R) leads Chief Justice Cheri Beasley (D) by 433 votes as a recount continues.
Ohio:  Liberal D former SOS Jennifer Brunner won 55% against Justice Judi French (R).  Justice Sharon Kennedy (R) won 55% against Judge John O’Donnell (D).  The Ohio Supreme Court will be 4R, 3D.
Texas:  All R incumbents were reelected with 53% to 55% of the vote.

New Hampshire: Republican Governor Chris Sununu was reelected, and Rs won a majority on the state Executive Council, which will allow the NH Supreme Court vacancy from 2019 to finally be filled.

Colorado:  Governor Jared Polis appointed Maria Berkenkotter to the Colorado Supreme Court.  She replaces retiring Chief Justice Nathan Coats, the last Republican-appointed justice on the court.  Justice Brian Boatright will be the new chief justice.

Kansas:  Governor Laura Kelly appointed Judge Melissa Standridge to the Kansas Supreme Court, filling the vacancy left by former Justice Carol Beier.  Standridge served on the Kansas Court of Appeals since 2008.  She also served as chambers counsel to U.S. District Magistrate Judge David Waxse and in private practice.  This is Kelly’s third appointment.

Massachusetts:  Governor Charlie Baker appointed Kimberly Budd as chief justice of the court, replacing Chief Justice Ralph Gants, who died at age 65.  She has been a justice since 2016.  Baker nominated Boston Municipal Court Judge Serge Georges Jr. to replace Budd.  He was nominated to the lower court by Deval Patrick.  Baker also nominated Appeals Court Associate Justice Dalila Argaez Wendlandt to fill the seat of Justice Barbara Lenk, who is age-limited in December.  She clerked for Judge John Walker Jr. (2nd Circuit), and was appointed to the appeals court by Baker in 2017.

Numbers and Trivia:

Circuit Judges:  The new circuit justice assignments for Supreme Court justices have been announced.  They hear emergency appeals from lower courts in these circuits.
1 Breyer
2 Sotomayor
3, 5 Alito
4, DC Roberts
6, 8 Kavanaugh
7 Barrett
9 Kagan
10 Gorsuch
11 Thomas
My prediction from October was only off on the 8th Circuit.

History:

Kennedy:  Ilya Shapiro reviews the history of Anthony Kennedy's confirmation and his jurisprudence on the court.

Resources:
Bench Memos (National Review)
The Vetting Room
Twitter: FedJudges Senate Cloakroom
Senate Judiciary Committee
ABA Judicial Ratings
Wikipedia: Trump Judges US Appeals Courts
Senior Status Spreadsheet
Future Judicial Vacancies
BostonPatriot diaries: History Trump DC-5 6-11 9th
Ballotpedia--State Supreme Court Vacancies Elections
The Supreme Courts
2020: March April May June July August September October Elections November

Judicial Nomination Strategies

Assuming that we end up with President Biden and an R majority senate, what strategy should R senators pursue with regard to Biden's judicial nominees?  There are several possibilities.

1. Maximum Resistance.  Don't confirm any Biden nominees.

2. One-for-One.  Confirm one conservative and one liberal for a pair of vacancies.

3. Status Quo.  Confirm liberals to replace liberals and conservatives to replace conservatives.

4. Tit-for-Tat.  Treat Biden nominees the way that D senators treated Trump nominees.  D senators who acted reasonably (IL, MI, CT) get their choices, those who didn't, don't.

5. Back to Normal.  Confirm most Biden nominees, rejecting a few who are too extreme.

Strategy 1 offers the biggest upside and the biggest downside, depending which party gets control of both the presidency and the senate first.  Strategy 2 would likely result in a net gain for conservatives, since more liberals are likely to retire under Biden than conservatives.  Strategies 2-5 at least offer the advantage of filling seats that could go to more leftist nominees if Ds take the senate.

Strategies 2 and 3 could turn into strategy 1 if Biden doesn't go along.  But they at least allow R senators to say they offered a compromise.  Obama did agree to a one-for-one deal in Georgia in 2014 when Ds still controlled the senate, so Biden might agree to something similar.

Personally, I'm leaning toward strategy 2 for appeals court nominees.

Friday, November 27, 2020

2020 Michigan Local Election Results

There were many local election results of interest in Michigan.  Rs picked up some seats in areas where President Trump did well, but there were also missed opportunities due to local GOPs not contesting seats.

Allegan:  Rs won all six countywide offices and won 7/7 county commission seats, all uncontested.

Bay:  What a failure by the Bay County GOP.  Rs won all state elections in Bay County, but the Bay GOP did not contest six of eight countywide offices.  Popular D county executive (and former congressman) Jim Barcia was reelected.  R Michael Rivard did defeat incumbent D drain commissioner (and former state rep) Joe Rivet.  Rs contested only two of seven county commission seats, and won one of them.

Berrien:  Rs won all seven countywide races.  Rs won 10/12 county commission seats.  The district 5 (St. Joseph) seat resulted in a tie, which was won by the R drawing lots.

Calhoun:  Rs won all five countywide offices and 5/7 county commission seats.

Eaton:  Rs won 3 of 5 countywide offices, prosecutor, clerk, and drain commissioner.  Former state senator Rick Jones lost for sheriff by 2.6%, and an R lost for treasurer by 4.5%.  Rs won 9/15 county commission elections, picking up three seats.  An R won district 10 by 7 votes.

Genesee:  Ds won all six countywide offices.  Rs won 2/9 county commission seats, holding district 6 (Fenton) and picking up district 7 (North), but losing district 9 (East) by 2%.  Rs also came within 5% in district 5 (Grand Blanc) and 2% in district 8 (West).  Thus Rs were close to winning a majority on the Genesee County Commission.

Grand Traverse:  Rs won all six countywide offices.  Rs won 5/7 county commission seats.

Jackson:  Rs won all six countywide offices and 7/9 county commission seats.

Kalamazoo:  Rs won only the unpaid surveyor office of the six countywide offices.  R Treasurer Mary Balkema narrowly lost to underqualified D Thomas Whitener.  Rs won 4/11 county commission seats, picking up districts 6 and 9.  In district 6 (Cooper, Richland, Ross), former commissioner (2002-16) Jeff Heppler won 56% over a D appointed by the D-majority board when an R resigned. District 9 (Texas), vacated by Christine Morse to run for state rep, was won by the man she beat in 2018, former state senator (1994-2002) and commissioner (2014-18) Dale Shugars with 50.2%.  Rs got about 48% in districts 5 and 11.

Kent:  Rs won all five countywide offices, though the drain commissioner race had only a 3% margin.  State senator Peter MacGregor was elected treasurer by a 4.4% margin; there will be a special election to fill his seat.  Rs won 11/19 county commission seats.

Lenawee:  Rs won all eight countywide offices and 8/9 county commission seats.

Ingham:  Ds won all six countywide offices and Rs won 3/14 county commision seats.

Leelanau:  Rs won all seven countywide offices in a county that President Trump lost.  Rs won 4/7 county commission seats.  District 1 was won by only 2 votes.

Livingston:  Rs won all six countywide offices and 9/9 county commission seats.

Macomb:  Rs won 4 of 5 countywide offices, only losing the sheriff race.  State senator Pete Lucido won 52% for prosecutor.  There will need to be a special election to fill his seat.  Former state rep. Tony Forlini beat incumbent clerk Fred Miller 51-49.  Incumbents Larry Rocca and Candace Miller were reelected as treasurer and public works commissioner.  Rs won 7/13 county commission seats, beating incumbents in districts 5 (Sterling Heights) and 10 (St. Clair Shores, Harrison Township).  Rs also came close in districts 9 and 12.  Rs will control redistricting.

Marquette:  Ds won all seven countywide offices, and 6/6 county commission seats.

Monroe:  Rs won 3 of 6 countywide races, prosecutor, clerk, and sheriff.  The R barely lost for sheriff.  Rs did not contest the drain commissioner or surveyor.  Rs won 7/9 county commission seats.  Former R state senator Randy Richardville beat former D state rep Bill LaVoy in district 5.

Muskegon:  John James won this historically D county.  Rs contested four of six countywide offices, but won none.  The drain commissioner race was lost by only 1.4%.  Rs won 5/9 county commission seats, picking up district 9 (Whitehall), and losing district 3 (Muskegon Township) by only 3.5%.

Oakland:  Rs won only one countywide office.  Mike Bouchard got 54% for sheriff.  Mike Kowall got 44% for county executive.  Joe Kent got 46% for Treasurer.  Rs won 10/21 county commission seats.  Klint Kesto lost district 5 (West Bloomfield) by 606 votes (1.7%).  Adam Kochenderfer won after a "technical glitch" initially reported that he had lost.  Rs have lost control of redistricting (which is controlled by the commission), which is bad news for Oakland County Rs.

Ottawa:  Rs won all five countywide offices and 10/11 county commission seats.

Saginaw:  Donald Trump won Saginaw County in 2016, but the Saginaw GOP only contested one countywide race, and lost it.  Rs won 5/11 county commission seats, picking up district 7 (southeast), but losing district 3 (Saginaw Township).

St. Clair:  Rs won 6/7 county commission seats.

Van Buren:  Rs won all 7 countywide offices uncontested.  Rs won 6/7 county commission seats.

Wayne:  Rs won only county commission district 9 (Livonia/Northville).  District 15 (Downriver) was lost by 11%, so this district may be worth targeting in the future.

Washtenaw:  Ds won all countywide races and 9/9 county commission districts.  District 3 was a close 3.2% loss.

Monday, November 23, 2020

Did Steven Crowder find Massive Voter Fraud in Detroit?

Conservative commentator Steven Crowder claims to have found "173K Anonymous Votes" in the Detroit election returns.

Crowder is a good guy, but he doesn't know what he's talking about.  The "precincts" with no registered voters are for absentee votes (that's what AV means).  Some jurisdictions put their absentees in separate "precincts", which often combine several real precincts together.  (Kalamazoo County used to do this.)  Note that Detroit has 503 real precincts, but only 134 AV precincts.  It's  a dumb system, but that's how Detroit does it.

Does anyone think local Republicans or the Trump campaign wouldn't have complained about fake precincts with thousands of votes if there was something here?  They didn't, because they know how Detroit reports election results.

Recall that Detroit had 1000 FEWER votes for Joe Biden than for Hillary Clinton in 2016, which is hard to square with massive voter fraud there.

Has Dr. Shiva Proven Michigan Voter Fraud?

A recent youtube video by Dr. Shiva Ayyadurai claims that precinct-level data in Michigan reveals voter fraud.  Dr. Shiva has a Ph.D. in biological engineering from MIT.  He has twice run for senate in Massachusetts as an independent, once after losing the Republican primary.  He also has some "interesting" views on other topics, as detailed on his Wikipedia page.

   

My favorite part of the video is when Dr. Shiva says that Michigan has "around 86 counties".  Why not just say 83, the actual number?

Dr. Shiva analyzed data from the four largest counties in Michigan (Wayne, Oakland, Macomb, and Kent).  He made scatterplots of the precinct data for each county, comparing the straight ticket percentage for Republicans to the difference between Trump percentage and R straight ticket percentage.

On the Oakland County plot, he drew a horizontal line segment, followed by a downward sloping line.  While this seems to fit the Oakland data, it doesn't fit the Kent or Macomb data any better than just a regression line.  The plotted lines slope downward.  He claimed that this is proof of votes being taken away from Trump.

Note that the three counties Dr. Shiva accuses of fraud use three different voting vendors.  His video isn't focused on Dominion Voting Systems--his theory would implicate all three vendors.

Matt Parker, a math popularizer, responded to Dr. Shiva.

He points out a serious mathematical error.  Dr. Shiva subtracts two percentages that don't come from the same population!  This results in nonsensical data.  As Parker points out, this subtraction isn't really necessary to make the point Dr. Shiva wants to make.  You can just find a correlation between R   straight party percentage and Trump percentage.  I did this for Kalamazoo County.


The correlation is quite strong, as we would expect.  Dr. Shiva is asserting that the proportions of Republican straight ticket voters and Trump voters among non-straight ticket voters should be about the same.  That is, the slope of the regression line should be (approximately) 1.  However, we find that the slope is actually less than 1.  Dr. Shiva claims that this is evidence of voter fraud--that an algorithm in the vote counting machine has flipped votes from Trump to Biden (while presumably not flipping straight ticket votes).  But is this assumption correct?

Parker points out that this assumption can be checked by plotting straight ticket D votes against Biden votes.  If Trump votes were flipped to Biden, those votes should show up in this plot, right?  But the slope of the regression line for this plot is also less than 1.


This disproves the vote-flipping theory.  But what is causing the slopes of both these plots to be less than 1?  Parker didn't try to answer this, but there is a reasonable explanation.  When a precinct leans heavily to one party, there will be more people who defect to the minority party than the reverse because there are more people in the majority party to defect.

Consider an example.  Say a precinct has 100 voters, 80 typically vote R and 20 typically vote D.  Say half of the voters of each party vote straight ticket, 40 R and 10 D.  Then the R percentage of straight ticket votes is 40/(40+10) = 80%.  Now say 10% of each party's non-straight ticket voters defect to the other party's candidate.  That would be 8 R and 2 D voters.  Then the non-straight ticket voters break down 40-8+2 = 34 R and 10-2+8 = 16 D.  The R percentage of non-straight ticket voters is 34/50 = 68% R, less than the 80% R straight ticket voters.

Feel free to construct your own example.  As long as the proportion defecting from each party is the same, you will find the same effect.  Thus there is a reasonable explanation for the fact that non-straight ticket votes are less partisan than straight ticket votes.  Dr. Shiva is mistaken.

Thursday, November 19, 2020

How Trump Lost Michigan

Four years ago, I wrote a long analysis of How Trump won Michigan.  The key points of that analysis are

  • Trump did much better than other republicans in rural and downscale (working class/union) areas like Downriver, Macomb, suburban Genesee, and the northeastern Lower Peninsula.
  • Trump did poorly in upscale (highly educated) suburbs in Oakland, Kent, and Kalamazoo counties.
  • Turnout in Detroit was relatively poor, providing democrats with a relatively small absolute margin there.
Trump's margin of victory in Michigan in 2016 was 10,704 votes.  At the moment, the unofficial totals show Joe Biden with 2,804,039 votes and Trump with 2,649,852 votes.  Biden's margin over Trump is 154,187 votes.  This analysis will consider county vote totals to determine where this shift occurred, and whether there is any reason to be suspicious of the official totals.

County Margin Changes

Trump's performance by county is illustrated in the map below.


Data at the SOS website is broken down by county.  Subtracting 2016 and 2020 margins, we find the biggest (approximate) changes in margin in the following counties.

-54200 Oakland
-32500 Wayne
-30900 Kent
-23000 Washtenaw
-11300 Ingham
-10700 Kalamazoo
  -9400 Macomb
  -7300 Ottawa
  -4600 Grand Traverse

The totals are broadly consistent with the results of the last two elections.  Counties with many white leftists, particularly those featuring large universities (Washtenaw, Ingham, Kalamazoo) saw large increases in D turnout.  Upscale suburbs continued to trend away from Trump.  This is particularly the case in Dutch Reformed Christian areas (Kent, Ottawa), which voted for Ted Cruz in the 2016 primary.

Trump also improved his margin in many smaller counties.  The largest gains were about 3300 each in Monroe and St. Clair, and 3100 in Montcalm.

Local Voter Fraud?

Of course, this is taking the unofficial vote totals for granted.  Since election day, there have been many claims of fraud, and allegations that the election was stolen.  Specific allegations include that late-arriving absentee ballots were back-dated so that they could be counted, and that some poll workers encouraged voters to support Biden.  These allegations are plausible (though unproven) and deserve further investigation.

We should be careful to distinguish between voter fraud that happens a few votes at a time, and mass voter fraud (adding, deleting, or changing thousands of votes at a time).  Only the latter could have swung the election by a large enough margin to give the state to Biden.

Allegations of fraud have centered around the city of Detroit and the counting of absentee ballots at the TCF Center (formerly Cobo Center).  This is understandable, given Detroit's long history of incompetence (or worse) at running elections.  To get a handle on what happened in Detroit, consider the raw totals and percentages for GOP and D candidates for the past six elections.

Year – GOP (percent) and Dem (percent):
2000 – 15688 (5.22%), 282111 (93.88%)
2004 – 19343 (5.93%), 305258 (93.65%)
2008 –   8888 (2.65%), 325534 (96.99%)
2012 –   6018 (2.09%), 281743 (97.63%)
2016 –   7682 (3.11%), 234871 (94.95%)
2020 – 12654 (5.08%), 233908 (94.00%) 

Note that Trump got 5000 MORE votes than he got four years ago, consistent with a modest improvement with black voters nationwide.  Biden got 1000 FEWER votes than Hillary did in 2016.  The data provide no evidence to support mass voter fraud in Detroit.

How then did Biden gain 32000 votes in Wayne County versus 2016?  Trump lost ground in the upscale suburbs of western Wayne County.  For example, Trump's margin declined in the following jurisdictions.
-7500 Canton Township
-5200 Livonia
-2000 Northville Township
-1800 Plymouth Township

Similar patterns occurred elsewhere in the state.  In Kent County, Trump lost 13000 votes in Grand Rapids, and lost the suburbs of Kentwood, East Grand Rapids, Wyoming, and Grand Rapids Township.

Trump's biggest relative declines from his 2016 percentages were in West Michigan, Oakland, and Washtenaw.  Trump's biggest relative increases from his 2016 percentages were in the northern lower peninsula and Wayne County.

Dominion Voting Systems

If fraud in one or two jurisdictions cannot explain the swing to Biden, could fraud have been widely distributed across the state?  Such theories revolve around Dominion Voting Systems, a manufacturer of voting machines and vote counting machines.

On election night, something went wrong with the vote totals in Antrim County in northwest Lower Peninsula.  The reported totals showed Joe Biden handily winning the county, and many people (including me) immediately realized that this could not be correct.  The county clerk retracted the results and ordered a recount, which produced a reasonable result.  (Trump's margin declined by 233 votes compared to 2016.)

It still isn't clear what happened in Antrim County.  The clerk blamed a software error.  The Secretary of State blamed human error (i.e., the clerk).  If Dominion software produced a false result in Antrim, could it have done so elsewhere?

In theory, vote counting software could be programmed to flip a small fraction of votes from one candidate to another, so that the fraud would be difficult to detect.  Some commentators have claimed that Dominion did this.  This cannot be what happened in Antrim, as the results there were absurd, and quickly corrected.  Did Dominion flip votes in other counties?  Dominion machines are used in 65 Michigan counties, and the others are split between two other vendors.  The state of Michigan helpfully maps the vendors by county on a website.  I broke down the presidential vote totals by counties that use Dominion machines and those that don't.


In 2020, the totals in counties that use Dominion machines are
Trump 1410197, Biden 1459230 (49.1% Trump two-party)
In 2016, the totals in the same counties were
Trump 1237598, Clinton 1335566 (48.1% Trump two-party)
In 2020, the totals in counties that don't use Dominion machines are
Trump 1209676, Biden 1212807 (49.9% Trump two-party)
In 2016, the totals in the same counties were
Trump 1069867, Clinton 1056032 (50.3% Trump two-party)

Thus Trump actually improved from 2016 in counties that use Dominion machines and declined in counties that don't use Dominion machines.  The data provide no evidence to support the theory that Dominion machines flipped votes from Trump to Biden.  Indeed, some of the counties where Trump declined the most (Oakland, Macomb, Ottawa, Kalamazoo, Grand Traverse, Livingston) don't use Dominion machines.

Of course, a conspiracy theorist could double down and claim that all three vendor's machines were compromised, not just Dominion.  There is a simple way to test this theory.  Remember that Michigan doesn't have voting machines, it has vote counting machines.  There are paper ballots that can be recounted when desired, as was done in Antrim County.  As far as I know, the Trump campaign has not requested a recount in Michigan.

Comparison to US Senate Results

Another argument casting doubt on the validity of the election results concerns the results in the US Senate race.  The unofficial vote totals give Gary Peters 2,734,559 votes, and John James 2,642,221 votes, a margin of 92,338.  The distribution of votes across the state is similar to the Presidential race, but the relative differences are noteworthy.


Some observers have noticed the vote totals for Trump and James were quite close (only 7631 apart), while the vote totals for Biden and Peters were about 70000 apart.  Indeed, there were about 58000 more votes case in the presidential race than the senate race.  This has lead some to claim that 70000 ballots were cast only for Biden, and that this is proof of voter fraud.

There are several problems with this theory.
1. There is no proof that dropoff is due to votes only for Biden.  Voters could have skipped the senate race but voted in other races downballot.
2. There is nothing unusual about dropoff.  In 2012 (the last year before 2020 with both presidential and senate elections), there were 78043 more votes for president than for senate.  In 2008, there was a dropoff of 153156 between these races.  There is typically more dropoff further downballot, as some uninformed voters choose not to vote in races when they are unfamiliar with the candidates.
3. Someone committing mass fraud would presumably want Ds to win downballot, and they could just as easily vote straight ticket D to do this.
4. There was essentially no dropoff in Detroit (note that 77% of votes in Detroit were straight ticket D).
5. Trump and James did better in different regions of the state.  The map below uses Orange for Trump performing better than James, and blue for James performing better.


Trump performed better in downscale areas, while James performed better in traditional GOP areas in West Michigan.  Note that James won Kent, Muskegon, and Leelanau Counties, which Trump lost.  James got 11000 more votes than Trump in Kent County and 8700 more in Oakland, while Trump got 12000 more votes than James in Macomb County.  James did essentially the same as Trump in Detroit, gaining only 80 more votes there.

Conclusion

There is nothing in the data to indicate mass voter fraud in Michigan of the size that would have to exist to change the result of the Presidential election.  More than 150,000 fraudulent votes should be noticeable in the raw election results, but they are not there.  If there is any doubt about the reliability of the vote counting machines, a manual recount would resolve it.