Trump Tweeted that Biden Won

Yesterday, Donald Trump sent out this tweet:

He won because the Election was Rigged. NO VOTE WATCHERS OR OBSERVERS allowed, vote tabulated by a Radical Left privately owned company, Dominion, with a bad reputation & bum equipment that couldn’t even qualify for Texas (which I won by a lot!), the Fake & Silent Media, & more! https://t.co/Exb3C1mAPg

— Donald J. Trump (@realDonaldTrump) November 15, 2020

Note the first two words: "He won." That suggests, for example, that Joe Biden won. Implicitly that also suggests that Trump lost because presidential elections don't have two winners, just one winner and one loser. The comment about Dominion is something Trump saw on the Internet. There is not a shred of evidence backing up that claim.

Not surprisingly, the website of Dominion Voting Systems pushed back hard at these allegations. Here is the top of its front page:

Of course the fact that Dominion claims everything is fine and dandy means almost nothing. They would have reacted the same way even if they were guilty as hell. But before anyone should give Trump's assertion any credibility at all, he needs to show some evidence. "I saw it on the Internet" is not evidence.

After a little while, Trump apparently realized what he had said, so he made clear that he will not concede:

He only won in the eyes of the FAKE NEWS MEDIA. I concede NOTHING! We have a long way to go. This was a RIGGED ELECTION!

— Donald J. Trump (@realDonaldTrump) November 15, 2020

Ron Klain, Biden's chief-of-staff-designate, summed up Biden's response to the tweets by telling NBC's Chuck Todd yesterday on "Meet the Press": "Donald Trump's Twitter feed doesn't make Joe Biden president or not president. The American people did that." Klain also expressed his hopes for a seamless transition, although the Trump appointee running the General Services Administration, Emily Murphy, refuses to release the resources federal law provides to an incoming administration. It is unlikely that Trump will ever concede defeat and Murphy will probably hold out as long as she can. However, Georgia, Michigan, and Pennsylvania will certify their results within the next 10 days. After that, it will be next to impossible to maintain that Trump won. And all states must announce who won by Dec. 8 so the electors can plan their trips to their respective state capitals on Dec. 14 to cast their electoral votes. (V)

Trump Is Doubling Down on Legal Action

Building a team of rivals worked well for Abraham Lincoln, but that doesn't mean it is a solution to all problems. In fact, having two captains on a ship is frequently a very bad idea, especially in a storm. Donald Trump apparently hasn't gotten the memo, though.

Back in June, the Trump campaign began assembling a legal team to dispute states he lost by narrow margins. The effort was led by Citizens United President David Bossie, who has plenty of contacts in conservative circles. He made good progress at putting together top conservative lawyers to contest states where needed. After the election, their focus was Arizona, Michigan, and Pennsylvania. If Trump had just let Bossie run the legal show, Bossie's legal team presumably would not have carried the day, given the facts on the ground, but it would have at least put the best possible case forward.

Friday, Trump screwed the pooch. He appointed his TV lawyer, Rudy Giuliani, to run the legal team. That's the same Rudy Giuliani who staged a major campaign event last week in the parking lot of a landscaping business adjacent to an adult bookstore and a crematorium. Now Giuliani and Bossie are going to be running competing legal efforts. And boy will they ever be rivals.

Although Bossie doesn't have a law degree—in fact, he is a college dropout—he understands enough about politics to know how to develop an overall strategy and to figure out the best approach to the upcoming court cases. Giuliani, who has a law degree and served two terms as mayor of New York City, is basically a loose cannon. Bossie understands that arguing that Democratic voters in Philadelphia got more time to "cure" sick ballots than rural voters in the "Alabama" part of the state, might possibly sway a judge if he can prove it. He also knows that in other states he might need different arguments. One size does not fit all. Giuliani thinks that pitching weird conspiracy theories he found on the Internet, which have no evidence to back them up, is the way to go. One of his favorites, already parroted by Trump (see above) is that Dominion Voting Systems altered vote counts to favor Biden in multiple states. If he had some evidence to that effect, it could impress a judge. Saying: "I saw it on the Internet," is rather less likely to succeed. Bossie and Giuliani are going to be at war with each other at least until Dec. 14, when the electors vote, and maybe beyond.

This battle couldn't have come at a worse moment for Trump. The various law firms that Bossie had lined up are getting skittish. In Arizona, the law firm of Snell & Wilmer withdrew from the Arizona case, sometimes called "Sharpiegate." It alleged that Republican voters were given Sharpie marking pens that would render their ballots invalid. In Pennsylvania, the law firm of Porter Wright Morris & Arthur withdrew from the lawsuit Trump had filed to contest the results there. Another big firm, Jones Day, said it wouldn't take on any new lawsuits for Trump because so many of its lawyers were very unhappy with it trying to undermine democracy.

With law firms dropping like flies and Captain Ahab fighting Captain Queeg for control of the legal ship, the chances of Trump getting the courts to overturn the election are pretty close to zero. (V)

Trump is Setting Booby Traps for Biden

Donald Trump may be fulminating against the election results, but he is acting as though he knows he lost. He is going to use his final 2 months in office to create regulations to help favored industries and do it in such a way as it is difficult for Joe Biden to undo them. Essentially, he is filling the government with booby traps for Biden.

Currently the Office of Management and Budget is reviewing 145 rules that it hopes to enact before Jan. 20. OMB officials are in a big hurry, but they also are being careful because technical errors could give Biden the opportunity to have a judge throw a rule out. It is not likely that all the proposed rules make it before the deadline. The process of enacting new rules during a lame-duck presidency is known as "midnight rule-making."

The rules involve items like the length of foreign student visas, restrictions on the EPA's use of scientific research, and rules about when a company can classify the people who work for it as independent contractors rather than employees. There are also many rules about worker health and safety. For example, one favor Trump wants to give the poultry industry is an increase in the line speed for chicken processing. Worker groups fear this will lead to more accidents for people who work in the processing plants.

Another of Trump's goals is to allow oil companies to drill in the Arctic National Wildlife Refuge and other federal lands. To achieve this goal he wants to issue drilling permits before leaving office. Once oil companies have the permits and start drilling, it will be far more difficult for Biden to ban drilling—after all, the companies have official permission to do so. That can't be revoked without a good reason and companies can fight revocation in court. (V)

Why Didn't Biden Do Better in Cities?

Short answer: because not enough urban dwellers voted for him. Long answer: Historically, Democrats wipe out Republicans in big cities and Republicans return the favor in the suburbs. That didn't happen this time. Biden made serious inroads in the suburbs but, in a development that has been less noticed, Trump did better in cities than he did in 2016. In particular, Trump did better in white working-class precincts in big cities as well as in Black and Latino precincts. This should be a warning signal to Democrats about the future.

Former Pennsylvania governor Ed Rendell (D) said: "There was a backlash, I think nationwide, among people who were afraid of the riots and afraid of what they saw and afraid of this 'defund the police' stuff, which, although we tried to debunk, we didn't debunk enough." The result was Trump getting more votes in Philadelphia this time, in part from white voters who do not like people burning and looting stores, no matter how angry they are about something.

In Northeast Philadelphia, where many police officers and firefighters live, there were signs everywhere reading: "Back the Blue." There were also blue lights on porches. The slogan "defund the police" did not win a lot of votes there.

In the leafy parts of Milwaukee, Democrats did better, but elsewhere Trump did better. Trump's team went into Black areas with an anti-Latino message and into Latino areas with an anti-Black message. That may have worked. States like Pennsylvania and Wisconsin were close and if this effect had been a little stronger, Trump might have won them.

A debate is breaking out among Democrats about what to do about the cities. Did Democrats veer too far to the left, or is the problem not having enough organization there, especially in minority areas? In some places, Democratic campaign workers didn't show up on the ground until very late. Maybe COVID-19 was the culprit, but voters don't think: "I know Democrats didn't reach out to me at all, but I understand it was on account of COVID-19, so I'll vote for them anyway."

The picture for the Democrats wasn't entirely gloomy, though. In parts of America where the economy is strong and jobs are plentiful, Joe Biden won. In places that have fallen on hard times, Donald Trump won. For example, in Phoenix, Fort Worth, and Jacksonville, all of which are thriving, Biden did well. In Miami-Dade and Osceola Counties, which have high unemployment, Biden didn't do so well. In short, while race and education played a role in voters' decisions, the economy also played a big role.

The changing economy definitely affects politics. Geographic areas dependent on knowledge workers and a digital economy are becoming more Democratic. Areas that rely on manufacturing, construction, and energy are becoming more Republican, so it is not only the urban/rural divide. Even within urban areas, a lot depends on what kind of economy the area has. Studies show that when an area loses jobs, the voters become more Republican. When it gains jobs, it becomes more Democratic. In one study, three variables best explain the voting shifts from 2016 to 2020: education, race, and county-level economic performance. When people feel left behind economically and threatened by (nonwhite) immigrants, they vote Republican, despite the fact that Republicans rarely do anything for these people other than small tax cuts attached to bills providing massive tax cuts for rich people. But as long as the voters haven't figured this out, the GOP will continue trying to nurture their grievances. (V)

Georgia's Recount Is 30% Done and Nothing Has Changed

Georgia's secretary of state, Brad Raffensperger (R), ordered a full by-hand recount of the presidential election. It is well underway now. In fact, 48 of the state's 159 counties have finished counting and almost nothing has changed. Some counties reported changes of five votes. Four counties reported no change at all. Joe Biden won the state by over 14,000 votes. If Donald Trump gains five votes per county, it is not going to make a lot of difference.

The Trump campaign cited voting irregularities, but has failed to produce evidence for the claim. In one lawsuit, Trump claimed that 60 absentee ballots in Chatham County were improperly counted. The judge wasn't impressed and threw the case out. One of the votes Trump contested was from a James E. Blalock, who died in 2006. Sounded like a strong case for fraud—until a local TV station discovered that it was cast by his widow, who always has used the name Mrs. James E. Blalock. Raffensperger wants the recount to be completed by Friday. (V)

The Battle for the Georgia Suburbs Is On

The elections are over, except for the Georgia recount and the two Senate runoff races there, which will determine control of the Senate. It is Sen. David Perdue (R-GA) vs. Jon Ossoff (D) for one seat and Sen. Kelly Loeffler (R-GA) vs. Raphael Warnock (D) for the other. According to Generra Peck, an adviser to a Republican firm that tracked suburban voter sentiment in Georgia, "style matters." In other words, suburban voters in Georgia fired Donald Trump because they don't like his style. However, she believes they will come back to their real home in the Republican Party in the runoffs because Trump isn't on the ballot and they don't agree with the Democrats on policy.

It's possible. In Fulton County and five other suburban counties around Atlanta, Joe Biden got 50,000 more votes than Ossoff. This does suggest that tens of thousands of suburban voters cast ballots for Biden/Perdue and they could vote for Perdue again.

On the other hand, Bianca Keaton, the Gwinnett County Democratic Party Chairwoman, said: "Persuading people to vote between D and R is not the fight at all. This race is decided on who gets more people to vote." That is almost certainly right. Turnout in special elections and runoffs is usually much lower than in general elections and the party that is more successful in getting its voters to the polls wins. The last runoff in a Georgia Senate race was in 2008, when turnout plummeted 43% between the regular election and the runoff. Of course, control of the Senate wasn't on the line then, as it is now. Also, the amount of money that is going to be spent on these two runoffs will be many multiples of what was spent then. Still, turnout is everything.

Worth noting is that Sen. Susan Collins (R-ME) was the only Republican senator to win an election in a state that Trump lost. In other words, Trump (plus incumbency) was able to carry Republican senators to victory in half a dozen crucial races. This suggests that Republican turnout is high when Trump is on the ballot, otherwise not so much. Will Republican voters in Georgia show up to vote on Jan. 5 without Trump to vote for? But the inverse is also possible: Will Democrats show up when they can't vote to send Trump packing?

Loeffler is already running ads saying that a Warnock victory would "give radicals total control." Apparently she thinks Biden and Warnock are radicals, something that would surprise many Democrats. But calling Democrats radicals, socialists, even communists is the oldest trick in the Republican playbook. Still, it often works. But who says you can't teach an old party new tricks? In addition to these well-worn attacks, Republicans are also flogging the much newer "defund the police" for all it is worth. Warnock responded to that with: "I'm not interested in defunding the police. It's clear that she wants to defund their health care."

Yesterday, Perdue announced that he would not debate Ossoff. Not surprisingly, Ossoff immediately called him a coward. In the other race, Warnock has agreed to a debate, but Loeffler hasn't responded yet. It is unlikely she will agree; neither Republican wants to try to answer uncomfortable questions about who did, or did not, win the presidential election. (V)

Democrats Are about to Have a Civil War

Whenever a political party loses an election, the aftermath is a huge battle about whose fault it was. Make no mistake, the Democrats lost the election. Or at the very least, they didn't win it. Yes, Donald Trump will cease to be president on Jan. 20 at noon, but he blew that all by himself in so many ways. But Democrats lost seats in the House, failed to win half a dozen winnable Senate races, and did a poor job downballot. The consequences are about to be felt.

Politico's Tim Alberta had a long conversation with Rep. Elissa Slotkin (D-MI) and wrote a long piece about it. Slotkin is probably one of the most accessible, open, and informal members of Congress and someone who doesn't have a problem with speaking her mind to reporters. When she is not in D.C., she lives on a farm in a rural area about 40 miles due north of Ann Arbor, MI. She is also typical of the moderate Democrats who knocked off 41 incumbent Republicans in 2018 and gave Democrats control of the House. And she is worried. Very worried.

Slotkin won her race on Nov. 3 by just 4 points—half of what her internal polls had predicted. Trump carried her R+4 district (MI-08). A number of her friends in the House lost. Slotkin is no closet Trump supporter. She is pitiless in listing his flaws: kowtowing to tyrants, abusing his power, dividing the country, and much more. But she said he does one thing right: He doesn't talk down to or demean voters the way Democrats often do. She said Democrats often criticize voters for not being thoughtful enough, for not being conscientious enough, for not being "woke" enough.

She further said that Democrats want to be a big-tent party but have a big barrier to entry: political correctness. She sees the country as being divided into Democrats, who believe they are enlightened, and everyone else, who resent being seen as ignorant, racist, yokels who are in need of cultural guidance. The problem is that there are 70 million of them.

More to the point, the divide she sees in the Democratic Party isn't so much between moderates like herself and progressives, like Rep. Alexandria Ocasio-Cortez (D-NY). It's between politicians in competitive districts (and states) and those in safe ones. She doesn't fault Ocasio-Cortez for pushing the envelope if that's what her constituents want, but she (Slotkin) doesn't want to be attacked for doing what her constituents want, even if that is very different from what members in very blue, very safe districts want. She resents having members from New York or California telling her how to represent her own district.

She likes to remind AOC and her friends in the Squad that Democrats did not win back the House in 2018 by promising to abolish ICE and pass Medicare for All. They won by promising to protect the ACA and to lower drug prices. On the caucus-wide call on Nov. 5, Speaker Nancy Pelosi (D-CA) congratulated her team on their great victory. The mood lasted until Rep. Abigail Spanberger (D-VA), Slotkin's closest friend in Congress (who barely survived), shouted: "If we are classifying Tuesday as a success from a congressional standpoint, we will get f**king torn apart in 2022." Spanberger then castigated progressives for talking about defunding the police and democratic socialism, which Spanberger said cost the party multiple seats in 2020 and will cost them more in 2022. Then the battle broke out. Within 48 hours it was full-blown warfare within the caucus.

Slotkin likes to point out that Republicans stand for something: Donald Trump's view of the world, which, roughly summarized, is to take the country back to 1950. Democrats are fractured and don't have a coherent vision of what they are for. Are they for improving the ACA or abolishing it altogether and replacing it with Medicare for All? She related hearing someone ask a voter what Joe Biden should do in his first 100 days and the answer was "address climate change." Slotkin thinks that climate change is real and important, but if Biden doesn't produce tangible benefits for millions of people in his first 100 days, he's toast. People will tune him out and say he is just another old politician. The battle for Biden's soul and the party's soul is only beginning. (V)

What about 2022?

Let's start with 2020. Democrats lost at least six seats; Republicans gained those six plus that of Rep. Justin Amash (L-MI). At the moment, Democrats have 219 seats, Republicans have 203, and 13 haven't been called yet. So Democrats will have a small majority for the next 2 years. Now let's get back to Abigail Spanberger and Elissa Slotkin and 2022. A total of 235 congressional districts have a PVI of R+x, where x varies from 1 to 33. In contrast, 192 districts are D+y, where y varies from 1 to 44. Eight districts are even. This means that if Democrats want to keep the House in 2022, they have to win at least two dozen Republican districts and probably more after reapportionment (which will give Texas and Florida more seats) depending how bad the gerrymandering is.

Here's a case study. In VA-05 (R+6), the Democratic candidate, Cameron Webb, a young doctor, ran on a platform of improving health care. His opponent was an extreme right winger who beat a slightly less extreme right winger in the Republican primary. The whole Republican campaign was about how Democrats want to defund the police. The Republican won by 6 points. In NY-11 (Staten Island), when Nicole Malliotakis (R) began running "defund the police" ads against Rep. Max Rose (D-NY), Rose dropped 21 points in the polls and lost.

All over the country Republicans aired 70 different ads on the theme of "Democrats want to defund the police." This contributed considerably to the Party's success in picking off Democrats in Republican-leaning districts. The Republicans also ran endless ads about socialism. These ads resonated with voters who had years ago fled from various left-wing regimes in Latin America and cost the Democrats two House seats in Florida.

Democrats can't run on opposing Donald Trump in 2022, but they had better think of something else since the party's brand is mushy. The 2022 House election is not going to be easy for them. Let's look at some history here. Here is how many seats the president's party has won in midterms since World War II:

You will notice that, in all but 3 years, the president's party lost seats. In 1998, the Democrats won seats because voters were angry with the Republicans for impeaching Bill Clinton. In 2002, George Bush was popular because the country came together after 9/11. The only other time the incumbent's party won was in 1962, when the young, charismatic, and popular Jack Kennedy was president and the mood of the country was good, so the Democrats picked up one seat in the House.

Over all the years, the average loss for the president's party in the midterms is 26 seats. This doesn't mean the Democrats are sure to lose seats in 2022, but unless they stop squabbling, come up with a clear message, and support their members in Republican districts, it could be a bloodbath. It's not for nothing that Rep. Liz Cheney (R-WY), said no to a Senate seat she was sure to win and could easily have held for 30 or 40 years. She clearly expects to be elected Speaker of the House on Jan. 3, 2023. (V)

COVID-19 Could Help Biden

COVID-19 is out of control. A month ago there were 45,000 new cases a day. Now there are 150,000 a day and it is growing exponentially, primarily in the Midwest and the South. This means that in 2-3 weeks, the number of deaths will start to skyrocket. So far, nearly 11 million Americans have been infected and 245,000 have died. The economy is in shambles. And Donald Trump is doing nothing except talking about how he won the election. Probably the only other thing he will do between now and Jan. 20 is issue pardons to his cronies and pass regulations to help companies that supported him. Tackling COVID-19 is not likely to be on his agenda going forward.

How bad is it going to be? If infections triple every month, then in 2 months (Jan. 17), they will be up ninefold. Since it is about 150,000 new cases a day right now, a nine-fold increase would put it at 1.3 million cases a day. But before it gets there, governors in many states will shut their states down. So, let's be optimistic and assume the growth rate is merely linear rather than exponential. Then we get this graph, where the dashed brown line is a linear projection out to Inauguration Day.

While better than exponential growth, 450,000 new cases a day is nothing to sneeze at (or cough at). But despite all the misery and death that will bring, there is one bright spot here. If there really are nearly half a million new cases a day along with 3,000 new deaths a day (basically, a 9/11 every day), people will be screaming "Do something!" and the newly inaugurated President Biden will have a strong mandate to do something. Assuming he has plans ready to deal with both the virus and the economy, he could deliver his plans to the House on Jan. 20 and there is a good chance Speaker Nancy Pelosi (D-CA) will ram them through on Jan. 21.

At that point, Majority Leader Mitch McConnell (R-KY) (assuming he is still majority leader), will have a choice: Pass the House bill or stall. If Biden tells people everywhere to tell McConnell and the Republicans in the Senate to pass the House bill and not argue that it should be 10% as big, the pressure will be enormous, especially on the senators up in 2022. Under enormous pressure to do something right now, it is possible that McConnell will feel obligated to bring the bill up for a vote. If so, it could get a few Republican votes and pass.

It is even possible that COVID-19 could affect the two Georgia runoffs. If the number of new cases per day is 300,000 by early January, the Democrats, Jon Ossoff and Raphael Warnock will be making the case that if McConnell continues to run the Senate, he will block anything Biden wants to do and tens of thousands, maybe hundreds of thousands of Americans will die as a result. Sens. Kelly Loeffler (R-GA) and David Perdue (R-GA) will counter with: SOCIALISM!!! But if Georgia voters are desperate enough, their desire to stop COVID-19 could be a factor in how they vote in the two Senate races. (V)

Trump Overperformed the Polls

If you look at our map of Nov. 3, you will see the polls, in the aggregate, got the correct winner in 48 states and D.C. Only in North Carolina and Florida was the prediction wrong. The polls in North Carolina had Joe Biden ahead 49% to 47%. Donald Trump won the state 49.9% to 48.6%, so instead of being +2, Biden was -1.3, an error of 3.3 points. Similarly, the polls had Biden ahead in Florida 49% to 46%. He lost 51.2% to 47.8%, an error of 6 points. These two states aside, even the states that had the right winner were often way off in terms of the margin. We and others have examined this already and are starting to make some headway in figuring out why the polls were so far off.

We have the polling data and now the election results, so let's take a look. The second and third columns below are our polling averages on Nov. 3. The next two are the election results.

State	D polls	R polls		D election	R election		D over	R over
Connecticut	51.0%	26.0%		59.2%	39.2%		8.2%	13.2%
Wyoming	31.0%	59.0%		26.6%	69.9%		-4.4%	10.9%
South Dakota	40.0%	51.0%		35.6%	61.8%		-4.4%	10.8%
West Virginia	38.0%	58.0%		29.7%	68.6%		-8.3%	10.6%
Kentucky	39.0%	52.0%		36.2%	62.1%		-2.8%	10.1%
North Dakota	38.0%	55.0%		31.8%	65.1%		-6.2%	10.1%
Tennessee	42.0%	51.0%		37.4%	60.7%		-4.6%	9.7%
New York	61.0%	33.0%		56.0%	42.6%		-5.0%	9.6%
Washington	57.5%	35.5%		54.3%	44.2%		-3.2%	8.7%
Utah	41.0%	50.5%		37.8%	58.0%		-3.2%	7.5%
Delaware	54.0%	33.0%		58.8%	39.8%		4.8%	6.8%
Oklahoma	37.0%	59.0%		32.3%	65.4%		-4.7%	6.4%
Hawaii	61.0%	28.0%		64.7%	34.3%		3.7%	6.3%
Montana	46.0%	50.5%		40.4%	56.7%		-5.6%	6.2%
Ohio	47.4%	47.4%		45.2%	53.3%		-2.2%	5.9%
Kansas	41.5%	51.0%		41.3%	56.5%		-0.2%	5.5%
Florida	49.3%	45.9%		47.8%	51.2%		-1.5%	5.3%
Iowa	46.5%	48.2%		44.9%	53.2%		-1.6%	5.0%
Missouri	43.0%	52.0%		41.3%	56.9%		-1.7%	4.9%
Wisconsin	52.1%	43.9%		49.5%	48.8%		-2.6%	4.9%
Massachusetts	62.0%	28.0%		65.2%	32.5%		3.2%	4.5%
Michigan	51.5%	43.6%		50.5%	47.9%		-1.0%	4.3%
Alabama	39.0%	58.0%		36.5%	62.1%		-2.5%	4.1%
Indiana	42.0%	53.0%		40.9%	57.0%		-1.1%	4.0%
Mississippi	41.0%	55.0%		39.6%	58.9%		-1.4%	3.9%
California	61.5%	30.5%		63.8%	34.1%		2.3%	3.6%
South Carolina	44.5%	51.5%		43.4%	55.1%		-1.1%	3.6%
Alaska	45.0%	50.0%		42.7%	53.5%		-2.3%	3.5%
Pennsylvania	50.7%	45.4%		49.9%	48.9%		-0.8%	3.5%
New Jersey	56.0%	38.0%		57.0%	41.4%		1.0%	3.4%
Minnesota	51.2%	42.5%		52.5%	45.4%		1.3%	2.9%
Texas	48.0%	49.2%		46.4%	52.1%		-1.6%	2.9%
Illinois	54.5%	38.0%		57.2%	40.8%		2.7%	2.8%
Nevada	49.0%	45.0%		50.1%	47.7%		1.1%	2.7%
New Hampshire	54.0%	42.8%		52.8%	45.5%		-1.2%	2.7%
Virginia	53.0%	41.5%		54.3%	44.2%		1.3%	2.7%
North Carolina	48.9%	47.3%		48.6%	49.9%		-0.3%	2.6%
Arizona	48.4%	46.8%		49.4%	49.1%		1.0%	2.3%
Georgia	49.3%	47.0%		49.5%	49.2%		0.2%	2.2%
Maine	53.5%	41.5%		54.5%	43.5%		1.0%	2.0%
New Mexico	54.0%	42.0%		54.2%	43.6%		0.2%	1.6%
Oregon	56.0%	39.0%		56.6%	40.4%		0.6%	1.4%
Arkansas	33.0%	61.5%		34.6%	62.6%		1.6%	1.1%
Maryland	58.0%	33.0%		65.2%	32.9%		7.2%	-0.1%
Louisiana	36.0%	59.0%		39.9%	58.5%		3.9%	-0.5%
Vermont	56.0%	32.0%		66.0%	30.6%		10.0%	-1.4%
Colorado	53.5%	41.0%		52.3%	39.5%		-1.2%	-1.5%
Average	48.2%	45.4%		47.8%	50.1%		-0.5%	4.7%

The first thing to note is that the polling numbers don't always add to 100% due to people who said they would vote for a third party or were undecided. Usually the sum of D + R is at least over 90%, although in a couple of polls it is way off. Connecticut is a poster child for that, with the sum being only 77%. This result is a single poll from Sacred Heart College, which is not exactly a powerhouse pollster, but it was the only recent poll for Connecticut. Call it a flaky poll and forget it.

The first four columns of numbers are straightforward. The column "D over" is the amount the Democrat did better on Election Day than the polls predicted. In Wyoming, the polls said Joe Biden would get 31.0%. He got only 26.6%, so he underperformed the polls by 4.4%. Wyoming was not one of his better states. So be it. The last column shows that Donald Trump overperformed the polls there by 10.9 points (69.9 - 59). As an example lower down, Biden did 1.5 points worse than the polls in Florida and Trump did 5.3 points better.

There is something strange going on here. Sometimes Biden did worse than the polls, sometimes he did better. In other words, there are plenty of minuses in the D over column—in fact, 28 of them vs. 19 positive numbers (D.C., Idaho, Nebraska, and Rhode Island were never polled, so they are omitted). On the whole, Biden did just slightly worse than the polls. On average, he did 0.5 points worse than the polls predicted. It is not so much.

Now look at the last column. In 43 states, Trump overperformed the polls. We have colored those in red. In four states (three of which are deep blue states) Trump underperformed the polls. These are colored in blue. On the average, Trump did 4.7 points better than the polls predicted.

Whoa! That's huge. So the polls were off not because Biden did worse than predicted. They were off because Trump did so much better than he polled—nearly 5 points better. How come? Pollsters are now suspecting that when they called Democrats, they were mostly very happy to explain at great length why they hated Trump. But when the pollsters called pro-Trump Republicans, they hung up the phone and thus were undersampled. Our data seems to confirm that Trump overperformed his polling rather than Biden underperforming his. No doubt more folks are going to be doing similar analyses soon and it will be interesting to see if they confirm our analysis. (V)

---

If you wish to contact us, please use one of these addresses. For the first two, please include your initials and city.

• questions@electoral-vote.com    For questions about politics, civics, history, etc. to be answered on a Saturday
• comments@electoral-vote.com   For "letters to the editor" for possible publication on a Sunday
• corrections@electoral-vote.com To tell us about typos or factual errors we should fix
• items@electoral-vote.com            For general suggestions, ideas, etc.

To download a poster about the site to hang up, please click here.

---

Email a link to a friend or share:

---The Votemaster and Zenger