Election 2020: what the data tell us

The presidential election has been transformed by data. From key swing states and early voting to voter suppression and possible election-night chaos, Elliott Morris, our data journalist and election guru, discusses his polling predictions and answers your questions about the 2020 race for the White House.

00:00 - An unprecedented election
00:35 - Why should we trust election predictions?
02:03 - How will the pandemic affect election predictions?
03:04 - Voter turnout in swing states
03:55 - Why we don’t use early voting data in our forecast
05:14 - Mail-in ballot rejection rates
07:38 - How long can we expect to wait for the results?
08:42 - Could the results flip after all mail-in ballots are counted?
09:39 - Will Americans accept the election result?

Further reading:

[TheEconomist]

On Wednesday November 4th—the day after the polls close—we will be hosting a discussion with Zanny Minton Beddoes, our editor-in-chief, and John Prideaux, our US editor. Tune in to hear their analysis of what happened in America’s election and insights into what might come next.

[DavidWilson-smym]

Economist, you NEED to buy Elliot some bookshelves. ASAP!

[earthwormjum]

Can we please get a Kickstarter campaign going to get that guy some bookshelves!!

[Tomiotano]

If only journalists were like this we would be all better

[bradywade]

I appreciate the more in depth look at things rather than just saying Trump is losing. It actually shows us aspects of polling that have changed since 2016. Not a bad media video.

[badlydrawnturtle]

The easy summary for why predicting this election is so difficult:
Predicting election outcomes relies on modeling, which in turn relies on data input from previous election cycles. This means a prediction is, in a sense, the "average" of previous elections. This in turn means that a built-in assumption of an election prediction is that it's an average, normal, typical election. This falls apart when it's apparent that the election cycle we are currently going through is not average, normal, or typical. The election models are, by their nature, operating on a false assumption of normalcy.

[jasonsteinway]

*Obama/Biden handled the Ebola virus, the Swine flu, the Republican Recession, Hurricane Sandy, and the Zika virus, while mr Trump said "I don't take responsibility" on 3/13/2020* -- *Just as he evaded serving in the military, mr Trump tries to evade responsibility for his failing to keep America safe*

[QuestionEverythingButWHY]

“Politicians and diapers must be changed often, and for the same reason.”
― Mark Twain

[Africanknight]

No matter the cause. People, just get out and VOTE IN PERSON. Bring your ID. And make sure you’re off that day and bring your folding chair. If you can shop for groceries then you can vote 🗳.

[bakersmileyface]

It's rare I like one of these videos, but this guy sold it to me. The dude is straight unbiased facts

[misterfunnybones]

If asked if you're a racist, most people will vehemently deny it, but not at the ballot box.

[FramesFTW]

No matter the results of the election, we need Elliot Morris and Nate Silver to get together and talk about polling in this election and the future of polling.

[benwhitnell]

The biggest problem to me is that most people don’t understand odds and models.

[LeilaLandOfficial]

Who else is here on election night? Desperate that "Biden's Robust Lead" will be true...

[dohyawatchin]

They are not coming for me, they are coming for you!

I’m just standing in their way.

~Donald J Trump

[AjitB]

Problem with polls are, people are not admitting they're voting for trump

[abcdLeeXY]

I hear it on the television everyday “your vote matter”
No, it depends where you are voting from and for. In a true democracy every vote should matter but nope America implement the electoral college system to make certain demographic less impactful with their vote.

[stevec.]

wonder if the polling system includes how many people registered to a party for the 1st time.

[howardj.anguilar]

The U.S. elections seems to have had a positive effect of risk sentiment, driving traders away big time from the U.S. dollar. Appreciations to Anmol Singh NYSE day trader.

[QuestionEverythingButWHY]

"Forecasting is the art of saying what will happen, and then explaining why it didn't! "
--Anonymous