Hello and welcome to the 538 Politics podcast, I'm Galen Droog. New economic data last week showed the largest ever quarterly contraction of the American economy. In the second quarter, the economy shrank by an annualised thirty three percent.

Now, we've all heard the wisdom that when it comes to elections, quote, it's the economy, stupid. And there are forecast models that try to predict elections based largely solely on the economy. But in an environment like the one we're in, how much can current GDP data really tell us? Also, as we discussed last week with Congressman will hurt, President Trump has suggested delaying the election because of a pandemic. While he can't make that decision, it furthers a theme of the president trying to cast doubt on the results of the election.

Four months, he's baselessly talked about massive fraud related to mail voting and said that this will be, quote, the most corrupt election ever. So what are the consequences of the president's behavior and how inclined are Americans to believe him?

Here with me to discuss, our editor in chief, Nate Silver. Hey, Nate, how's it going?

Hey, Gaylan. How's it going with you? I'm doing well. How's that model coming along? I mean, it's pretty much done. All right. So I have the knowledge of what it says.

I'm not going to tell anybody, though, who the suspense is.

Here with us is senior politics writer Perry Bacon Junior. Hey, Perry. Good to see you, Gail. You too.

And politics editor Sarah Frost and said, hey, Sarah Hagan and Claire is out today.

Let's talk about the economy. And in doing so, let's ask a variation of our favorite question, which is good use of modeling or bad use of modeling.

So there's a segment of the election forecasting world that puts most of its emphasis on the so-called fundamentals, which is to say, the economy and other fundamental metrics of American life as opposed to the polls. Those models got some attention last cycle for predicting a close election between Clinton and Trump. And so with this new economic data that we have, let's talk about how much the economy alone can tell us who will win. So neat. I know that you have gripes about this type of modeling, but I just want to ask you up front, if you plugged negative thirty three percent annual GDP growth into a fundamentals model, what would it tell you about how this fall's election would turn out?

Well, here is the problem. It depends on which model you're talking about. If you use the Alan Abramowitz time for change model, I think it is right, which is the second quarter GDP. That's just a number we got plus approval ratings in a term for incumbency. It predicts Trump to lose the popular vote by around 35 points. And because the model is very competent, it says the range going to be almost for sure somewhere between like 40 and thirty three.

Right. There's no chance that he'll only lose popular vote by 10 points, for example. Literally impossible, according to this model. You know, when you have a model that says something like that, that's a sign the model is badly designed. But let's say you could have a different model. Some models, for example, use third quarter GDP. So fall GDP a little bit closer to the election that's expected to rebound to maybe be plus seven or plus eight percent.

That would predict a Trump big win write. Some models use disposable income because of the Caires bill. Actually, Americans for the time being have more money in their pocket. So if you used second quarter disposable income, you predict a Trump landslide victory. So part of the problem is that there are a lot of ways to measure the economy. We don't have a very large sample of previous elections. And so the models violently disagree with one another about what they actually predict.

Now, people who make those models will go in and change those models, which kind of gets to the question of like, is this even pretending to be objective science at this point or is just kind of reflecting people's subjective Prior's right now? These models are basically crap. All right. They're basically so fair.

They're basically that your judgment is bad use of modeling. But can I ask before we move on and talk about the economy more broadly from the perspective of people who back this kind of modeling, why do it? What's the point in focusing more on the fundamentals than actual election polling?

Because they want to spin straw into gold, they want to pretend like from their perspective, not from your perspective, because they are bored and they study American elections and you like to be empirical. Right. And the fact that we only have like 12 elections with reliable polling data and, you know, 16 elections with reliable economic data, you just keep fiddling around with regression specification until you find something that you think is persuasive. I mean, it's like bad science.

I'm not being empathetic here because, like, I've been railing against these models for years and years. It was actually a moment for me where, like back in 2011 and 2012, there's a lot of debate about the economy is kind of mediocre. What's that mean for President Obama's re-election chances against Mitt Romney? It's just spent like months and months looking at these things. And it was a real eye opening moment for me that how badly these models that based on past data claim to be incredibly precise and predictive, how they completely failed in many cases when applied to new data that wasn't known to the model or at the time.

And in fact, these models literally did worse on real data, a real unknown elections, an actual prediction than if you had just predicted 50 50. So there are is a term called Overfitting, which means that when you have a lot of different ways to look at past data, you can look at 100 different hypotheses. Right. And you kind of pick whichever one of those hypotheses does the best. You're going to kind of wind up with the model that by chance alone happened to work on past data but won't actually apply very well going forward.

Peacocking is a related term.

I'm trying not to get too technical here, but from what I like, it literally it literally is bad science. It literally is bad science. And the performance of not performing. Excuse me, I'm getting a little bit too woak there. Right. The performance record is modeled on real data is not very good. Now look, I think there are better ways to design economic models, but like the notion that, like, you can somehow pinpoint how voters feel about this economy in a pandemic and therefore be extremely confident about the result based on the economy alone.

I mean, it doesn't really add up in part because, like, I'm spewing out all my arguments here, Gaylene, which I should, you know, you've got to save. The other panelists weigh in. I'm going to say I'm going to hold back at this point. But yeah.

All right. Perry and Sarah, when it comes to the, as I mentioned, 90s wisdom of presidential politics and that it's all about the economy, this is extrapolating from this modeling question. To what extent should we buy into that? How much does the economy actually sway? Who wins the election?

One thing that stood out to me is in recent elections, like so many other things, the economy is increasingly polarized. Political science is now showing things like consumer sentiment, which used to mean people were happy with the economy and that that was a good indicator for presidents. No longer seems to have as much of an effect. And something that Michael Tessler wrote for us recently on the site that really stood out to me was, you know, here we are in the middle of a recession, worse than what we saw in 2008.

And yet if you looked at polls in 2008, both Democrats and Republicans said the economy was bad. And in fact, McCain, he got widely attacked for saying, you know, the fundamentals of our economy are strong. People are like, no, we're in a recession. Whereas now if you look at the polls and this was true as of late June, the gap between Democrats and Republicans on how the economy is doing is the widest it's ever been.

Republicans think it's good. Democrats think it's bad. We saw this under Obama. It's gotten worse under Trump. So it makes me question to some extent, if you're partisan lens, is that strong already?

I don't know how much the economy really does impact you at the ballot box.

Can I toss the question back to you? Because I am curious, like when, you know, whatever Clinton adviser in 92 said, is the economy stupid? Were they right even then, was the economy more of a factor in the 80s and 90s? Actually, generally don't know. What I really object to is like the precision with which these models claim to be able to predict elections, right. If you have to choose between hypothesis number one, the economy completely determines elections.

Right. You didn't never have to look at a poll. And about this, number two, that the economy has nothing to do with it. Number two is actually closer to correct statistically if you model this properly. But, you know, in our estimate, the economy explains from our model, which we think avoids most of these problems, though not all. We think it kind of explains about 30 percent of variance and incumbent performance. So 30 percent is a lot.

Right. If you can explain 30 percent of how well the incumbent party does based on the economy, that can tell you quite a bit. But it's also a long way from 100 percent and it leaves plenty of room for candidates and campaigns and and wars and pandemics and things like that to matter. Now, there's a question of like in which period has the economy been relatively more or less predictive? I mean, it has been true that like in kind of the 70s, 80s, 90s, the economy was fairly predictive.

And these models, one of the reasons why I'm skeptical of these models, because it doesn't turn out to be as predictive if you extend the analysis backward when you have a really small sample size. Right. Then you would want to expand it by saying, OK, let's estimate what the economy look like in 1994 and see how the incumbent party did then. Right. We tried to fix and are now back to 1880, which more than doubled sample size.

But the other other question is like, can a normal economic indicator be applied in a pandemic like this one? Like you mentioned, Gailen, we had the worst quarterly GDP number in history. However, if you look at how do voters feel about President Trump's handling of the economy in polls, he actually is even money.

The real Clear Politics average of his economic job ratings say forty eight point six percent approve, forty eight point zero percent. Disapproval is actually technically in positive territory on the economy, which suggests that somehow that GDP figure might not kind of capture how the public thinks about the economy in the middle of a recession. One reason why is that the government has until recently been putting a lot of money in people's pockets. We talked about disposable income. You know, there hasn't been inflation, which happens in some economic crises.

The stock market has been pretty steady. Housing prices, it's actually a little bit unclear. Right. But like people, I think also understand that, like, OK, Trump may have done everything wrong in the pandemic. I think it's very poor handling for his response of covid. And that's one reason why he's a favorite to lose. However, I don't know that they necessarily blame the economy on Trump in particular. They blame the economy on covid.

So it's like the fact that you have to the fact that if you take some of these formulas like the Aronowitz formula and apply it to the real data, it comes up with ridiculous answers is kind of proof at how ridiculous I mean, this exercise is in the first place.

So let's move on from those fundamentals based forecast models, because you have disposed of them thoroughly.

But, you know, when it comes to how Americans are thinking about whether they want to give Trump a second term or not, is the economy the key factor for voters or does the pandemic trump absolutely everything?

I mean, we've already seen that voters have said they would rather use restrictions to help tamp down on the spread of covid even if it will damage the economy. So is this an election where the economy is just going to take a backseat, period?

The American national election studies, like they ask voters each election what the most important issue is facing them and the economy since 08 has led on that metric. Granted, it was more important to more voters in 2008 than it was in twenty sixteen.

And what we've seen in polls that are being conducted now, like the Ania's, of course, is only in the lead up to the election. Right. Is that more Americans are saying I am concerned about health care, which is related to the pandemic if they don't ask about the coronavirus specifically or a few weeks ago it was the protest. Right. That was top of mind for a lot of voters. So I do think that the economy could be, as Nate was saying earlier, too, right.

Like people aren't blaming Trump for the economy, they're blaming covid. So at the ballot box, they could be caring more about the coronavirus than the actual economy, even though they're kind of linked together.

So I would say the economy matters in two ways in the context of this election. The first is that Trump, for most of his last three years, has been saying, look, whatever you think of me, the economy is booming under my leadership. And a lot of Republican voters say that's why they're voting for a lot of Republican legislators and members of Congress say that. So I think when you look at like Trump get forty six percent of the vote in 2016, he's polling in like 40, 40 to something like that.

I do think some of the. Dropoff is probably explained by the fact that the one thing that people said, I'm voting for Trump because of the one thing Trump has been effective at, I would argue, is now been undermined. I think that's one fact, and I think that's a big problem for him. The second is, I think media coverage matters and how people perceive Trump in politics more broadly. And I do think Trump would be covered differently and I would say more positively if unemployment was at four percent as opposed to being in double digits.

So I think the media coverage of Trump has been very critical these last few months that release to the pandemic that relates to the protests, but also translates to the fact that the one thing that everybody cares about the economy that Trump was good at, he no longer looks good at. And I think Trump's reaction to that has been a third issue, which is that I actually think Trump in some ways has over interpreted 2016. He seems to think he won the election because he sort of did this white identity politics stuff.

And I think he won that in part because he's a Republican and Republicans vote for Republicans. And I sort of think he has decided the economy's bad. Let me lean into the identity stuff. And I think if he decided the economy's bad, let me fix that pandemic. Let me say I don't want to raise taxes. I don't I think he's shifted from the economy to the sort of race stuff. And I think he should go for option C.

I think that's right. I mean, it deprives Trump of the most traditional strategy for an incumbent, which is kind of the Morning in America strategy. Right. Or even kind of a modified Trump and version of it where you would sometimes hear from voters.

Yeah, we think don't like the tweeting. And the president's maybe a little racist. Right. But the economy is good and blah, blah, blah. Right. And now you can't say that now you can't like maybe the economy is going great for for a few people. Right. If you own, like, a liquor delivery business or something. Right. But for most people, you cannot be happy with the economy, the state of the country.

I'm just saying, though, that, like, covid is the predicate to to all of this. Right. And there is no kind of consideration of the economy independent of covid, I guess. And to the extent there is, it seems like mildly positive for Trump that people are a bit forgiving based on whatever three years of growth we had before covid before we move on, we've seen a little bit of uptick in terms of Trump's approval recently and even in some battleground states.

Does it look like he hit bottom a couple of weeks ago and things are going better for him now?

Or is that an overinterpretation of a small amount of recent data?

I think it's an overinterpretation of a small amount of recent data. I don't think things have moved that much. With that said, you know, if he was down nine points before eight and a half now or whatever, like I mean, that's a big that's a big margin. And you might expect things to tighten. Right? I mean, you know, one way to when where you can look at this is just to kind of say, OK, well, we don't know what our prior should be about the economy.

Right. Let's just kind of throw our hands up and say we assume the prior is it's 50 50 defaults toward a competitive election in a polarized electorate. Right. I mean, that's not a crazy approach to take and then maybe expect some tightening potentially. We were in a technical debate already, so I don't know how much more technical to get. But I think, you know, some polling averages like Real Clear Politics that don't account for house effects and a mix of pollsters they have can show artificial swings.

Our polling average hopefully avoids that and shows less of that, especially you look at kind of state polls and not just national polls. I mean, the state polls have been, you know, roughly speaking, as bad for Trump as they have, I think, at any point in the campaign. And before we close up on the economy subject, I would like to say if voters are no longer determining who to vote for based on economic growth, great.

The president has very little control over these things. This has always been a dumb part of politics coverage. As George Bush didn't hurt the economy in 92, he lost because the things he had no control over, like the president since, helped set tax policy. So that changes the economy a little bit. In this case, the president is head of the pandemic badly. That's affected the economy. But I'd be much more logical to vote based on whether you want pro-choice or pro-life judges.

And if you think Barack Obama, Donald Trump or John McCain can fix the economy, which they have no actual expertise in for sure.

Yeah, yeah. So in some ways, maybe becoming more partisan and unmooring, your preference for a president based on the economy is a positive effect of an increasingly partisan political world.

Who'd have thought silver linings anyway? Let's move on and talk about the president's attempt to cast doubt on the results of the election.

Last week, President Trump suggested delaying this fall's election, quote, until people can properly, securely and safely vote. He doesn't have the power to do that. Congress does. And according to reporting last week, lawmakers of all political stripes aren't on board. But his suggestion is part of a theme.

For years, the president has cast doubt on the results of elections. He claims, obviously baselessly, that millions of people voted illegally in person in 2016. Now, he baselessly says male voting will lead to, quote, massive fraud. What are the potential consequences for American democracy of President Trump working to delegitimize the results of the election?

I mean, it appears to me President Trump says things all the time that are problematic and misleading and outright false. So I think the question to me becomes not if what Trump says does matter, but whether people follow that, particularly do Republican elected officials in states who determine legislation decide because Trump says mail in ballots are invalid or fraud, that they're going to stop counting them? And I think the one thing that really has an impact over is it appears the person Trump has put in charge of the U.S. mail system is making some changes to it in terms of slowing down mail delivery that might affect the election itself.

And the other factor depends on if voters, particularly Republican voters, come to believe that all mail in votes are fraudulent for whatever reason. That creates an issue of like if a state has slow counting, is the Republican base in this state pressuring them to stop counting? So that matters. So I think it's important to not isolate, Trump says, all kinds of misleading, crazy things all the time. The key question is, do voters believe those things?

And do Republican elected officials act on those things?

Sarah, what do we know about the extent to which voters believe these things are primed to believe them?

A study released earlier this summer by 538 contributor Lead Men with the Voter Study Group looked at this question of authoritarianism and where Republicans and Democrats fall on that question. And one thing they found that was particularly striking was both Republicans and Democrats were open to their preferred presidential candidate, rejecting the legitimacy of the election if they could claim credible evidence of illegal voting or foreign interference. And that touches back to a poll from The Washington Post that came out in 2017 that found that among Republicans, roughly half believed Trump won the popular vote and they would support postponing the 2020 election.

Again, that was 2017. It's been a few years now. But the fact that that voter study group came out this summer and echoed that finding, I thought was particularly chilling. And it cuts both ways for both Democrats and Republicans.

Is it fair to say then that this is a significant threat? If the president were to follow through on, I guess, the extension of this, which is doubting the actual results of the election, maybe making some statement as to his win before the ballots have been counted? Is that a genuine threat to, you know, American democracy or is that alarmist at this point? It feels alarmist, right.

But Trump this is also familiar territory in the sense of how you opened the segment.

Right. Like he did this in 2016, claiming that people had gone in person and voted illegally.

And Gallup recently had polls and granted, this was from 2019.

But showing that, you know, only 40 percent of Americans were confident in the election, again, the majority was not. And in theory, that probably will be at a new high this fall. But I think it goes back, though, to what Perry was talking about earlier.

A lot of this will come down to Will. How do other elected officials respond? How do Republicans handle this if it continues to escalate? And what we saw so far, right, is that most did not stand with the president and his push to postpone the election.

What are the incentives there in terms of how Republicans respond to this? There are some situations in which Republicans push back pretty forcefully on certain things the president does or says pulling troops out of Syria is one example. This is another example. So what are the patterns in terms of when his fellow Republican lawmakers outright reject him? I mean, Trump has not gotten too much support on these sentiments from other Republicans, I don't think you can certainly argue that they have not done enough to push back or denounce him.

I think state level officials, like governors, secretaries of state, probably feel No. One, people generally like to make it easier to vote and they want the option to be able to vote by mail during a pandemic. Number two, there is a fine line between Trump trying to whatever prepare the ground for attacks on democracy and things that will discourage turnout from his own base, which would hurt Republicans up and down the ballot. Right. If you kind of allege that, hey, this thing is rigged, maybe people don't vote at all.

If you say, hey, mail ballots are rigged, maybe people, you know, are counting them to turn out on Election Day, but things can come up on Election Day. So Trump has not necessarily got that much support from other elected officials, but it is kind of a. Always dangerous, says with Trump, it has the makings of a coherent plan, a coherent plan where you create, which is going to happen, I think now a big partisan split and who votes by mail and who doesn't.

And then there are things that you do at a neglect or deliberately to make it harder for the ballots to be delivered to people and or people to get their mail ballots back to their states and or, you know, at the very least create a situation in which vote counting is slow.

Maybe you have a lead at 1:00 a.m. on election night because not all mail ballots will have been counted. Right. And then you kind of cast doubt on the outcome. The casting doubt on the outcome, I think may ultimately not lead to Trump winning or a constitutional crisis or even that. Right. But it is like it could delegitimize the next president. If it is Joe Biden, it could, you know, trigger lawsuits and stuff like that. If things are so slow like we're seeing when it isn't Trump's fault, it's like these primaries in New York where still there are a couple of race there, undeclared.

It's taken months to count ballots. You can imagine if that happens in a swing state, Arizona or Wisconsin or whatnot, how that might affect things. So there are scenarios where it gets messy to the question earlier about like, is it time to panic or is it alarmist? Right. It's one of those things where, like, the more alarmed people are now, then maybe the less risk there is to be alarmed, you know what I mean?

It's almost like covid or something. Right. But more panicked you are, the less things get out of hand right in. Unless the reason there is to panic. There are different groups here, right. Number one, voters need to be prepared for being very diligent about making sure they get their ballots and making sure they postmark them and send them in. Right. Number two, the media needs to be more prepared for the fact that we may not have a neat and tidy election night.

In fact, we probably won't. Although the election is as much of a blowout as polls show it is currently, maybe it would be resolved, but odds are it's going to be a long election week or longer. And number three, kind of state and local officials and Congress also. Right. I don't know what type of leverage Nancy Pelosi has. And she has a lot of leverage because, like, it would really have happened if there were another stimulus package and would probably hurt Trump if there are a lot of leverage to make sure there is enough funding for counting ballot returns and enough funding for the post office, sometimes I think, frankly, Pelosi doesn't use leverage in that way as effectively as like Mitch McConnell might or something.

But the fact people are alert to this, I think is is is less reason to worry. But there is like there is a real plan by which Trump could cause, at the very least, real confusion.

So I would say I am alarmed and I would say two things. The first thing is the president of the United States hinting, suggesting, implying we should change the election date is bad even if no one else follows it. Is this a bad thing? That's alarming in the sense that no other president I can remember would do that of either party. So the second part of that, I think, is really problematic is it looks like our election systems are already not very good pre covid like we had very long lines, blue states, red states, urban areas, rural, I mean, urban areas, long lines were terrible accounting mail in ballots and too slow even before this.

We have lots of people get their mail in ballots rejected because they forgot to sign the third signature requirement and which was not very clear about. So when you add on these existing system, which happen, red states, blue states everywhere, and you in theory in the middle of panic, need a president who's saying we need to work on election infrastructure. We need to like spend money to improve our entire structure because it's already bad. A pandemic makes it worse.

And instead, we have a president who is very much not doing this. So you have a situation where I think my real worry is there's going to be some state where the number of mail in ballots that are rejected or turned in late is higher than the margin of victory of one of the candidates. And that is really worrisome. And I would say that even if Joe Biden won the state or if Trump won that state, you shouldn't have more rejected votes than there are than the margin of victory is.

How is this different from and I ask this intellectually governors moving quickly to postpone primary elections earlier in the year in many cases, which was seen as the responsible thing to do. Well. Constitutionally, the date of the primary elections are not mandated, but no, I don't know that it was a good idea for governors to do that. And I thought there were too many liberals who were not even thinking long term and saying, of course, you have to be moved right now.

How dare Wisconsin make people go out and vote in the middle of this pandemic, right where it's like, again, constitutionally, Trump has no authority to do that. And there's, I don't think any allegation that these dates were moved in a way that contravened the Constitution. But precedent wise, yeah, I thought it was a bad precedent. And the fact that people can't like actually I mean, political commentators don't seem to give one iota of concern about, like, actually being intellectually consistent.

But no, I think that was not great. Yeah. And I think there was there was criticism from the left. I don't know how everybody in the left said that there was criticism from the people who did say moving the date is bad. Trump might take advantage of this. Don't do this. I think there was some under which there was some. I think Ohio, particularly, where the people were like, what do you or Wisconsin? I think both are like, what are you doing here?

I mean, I would say if and I'm going to be honest here, if on November 1st, 2016, Mitt Romney was president and he had said, hey, I think if people have to vote on this day, there's some pandemic this dangerous that it might kill lots of people. I would have heard it differently than Donald Trump. I'm not going to pretend like we live in a different world. John McCain, Mitt Romney, George Bush, Barack Obama, Bill Clinton, John Kerry saying it is different than somebody who violates democratic norms all the time.

I think it's worth noting Donald Trump is different. And like all you know, people should be more skeptical about things he says in this regard. You brought up history there in some signs referring to past presidents. And I want to dig a little bit more into whether we can rely on history at all as a guide for this moment. I know that we have had controversial elections in the past in the United States, probably the most recent one being of the 2000 election and the role that the Supreme Court played.

But in this particular case, how much can we rely on history as a guide for what happens when the legitimacy of election is called into question?

And how much is this simply very different from something that we've experienced as a country before?

There is precedent in the sense that, you know, we had an election during the civil war. Like one of the hallmarks of American democracy, right. Is like the peaceful succession of presidents. Right. In 1876 was a hugely contested one with Samuel Tilden and Rutherford B. Hayes. And essentially, in order for his to win reconstruction ends right in the south, agrees to it because Tilden had been the Democrat. So we have had contentious elections.

What makes this different? I think, again, goes back to what Perry was saying about you could imagine if President Romney was saying or President Clinton, if we have this life threatening pandemic, we should take precautions. That means postponing the election by a month. That would be a different messaging. Right. But at the same time, he's arguing, let's postpone the election. He's saying let's reopen schools. So there's a lack of consistency. I think it goes back to what Nete was saying about it doesn't actually fall within the president's wheelhouse to make this decision need to have Congress or would have to be the courts involved.

And so it's just the way in which he chooses to go about trying to push something that on one hand could have good instincts.

It's a life threatening pandemic, doesn't actually in practice achieve that stated goal.

So I guess, you know, we had this election in 2000 that was very contested. I think afterward in 2001, 2002, there was talk people will believe in the system anymore. They won't trust the Supreme Court. I don't think that was borne out. I think in 2004, we had a traditional election. George Bush won. I don't think people are going around. People hated George Bush who were Democrats, but I don't think they believed he ultimately won in 2004, even 2016.

There's been a Russia investigation, but I'm not sure that Democrats doubt that more people in Wisconsin voted for Donald Trump than voted for Hillary Clinton. Did Russia help? Trump is a contested issue. I don't think people are questioning the actual results of the election that much. So historically, I think we haven't had at least recently, we haven't had real questions about the legitimacy of the election results. I guess my worry about this one is I do think the sort of partisan slash cultural divides in the electorate are deeper than ever, meaning the sort of like the white Christian party versus the people who are not white and Christian is kind of getting much more deeper than ever.

So I do think the two sides really don't want to see the other win. And I think the Democrats feel like the second term of Donald Trump is the end of America. And so I think Joe Biden sort of implies this at times, that Trump is very using very apocalyptic language, as will your way of life will. And in the suburbs, so I do think in that sense, the stakes feel really high for people. I do think the stakes this election feel higher to people than terrorism.

And therefore, I think that the fighting over like the results could be much more divisive and intense if the results are in any way disputed.

Yeah, I mean, it feels like a lot of these things, right, where like directionally there is kind of precedent for Republicans trying to make it harder for people to vote generally. There is precedent for very messy elections from, you know, Samuel Tilden to 2000. I guess I feel like Trump has crossed several lines that hadn't been crossed before. I'm not sure I'm completely confident in that. And I would defer to someone who knew more about kind of the history of contested elections.

I know I don't think there's been any sitting president before who said, hey, the system that got me elected, that system is rigged and you can't trust what's going to happen in the next election. I don't think there's been that line crossed before. I mean, I have to think I think have to decide how serious I think these claims of, like making the post office dysfunctional are right. That would seem like it probably crosses a line, but maybe not.

Right.

I mean, obviously, we had years and years and years in this country where, like, black people were disenfranchised officially but unofficially or free or whatever.

Right. And women couldn't vote in people under 21 couldn't vote. Right. And so, you know, in some sense, the right to vote has been kind of at the center of all of this, potentially. But it does feel like this isn't normal.

That's if that's one way to ask it, to extend this to its potential logical conclusions. What would happen in a scenario where President Trump rejects the apparent outcome of the election and refuses to concede even if he has lost?

I tend to think this is depends a lot on what actually happens if Biden wins every swing state, Georgia and Texas. The media declares Biden obviously one Mitch McConnell starts making plans for the next year. And I think that Trump will be speaking and people will be ignoring him. And on January 20th, he will walk out and he'll just it'll all be a bit of a bunch of bluster. In other words, I think this is very dependent upon what the results are or is it is a down to one state.

If it's like Wisconsin, if if if it's down to like like a two sixty two sixty electoral vote wise and Trump is behind in Wisconsin by 400 votes, I think it's going to be hard to remove him, to be totally honest. And I think there's a problem even if he actually lost. I mean, I think people should be honest about who won the election, but that's going to be more challenging if the results are more clear. I think that is a useful thing to think about.

I totally think like here in Kentucky, for example, where I am, you know, in twenty nineteen, we had a governor's race where Matt Bevin, who was the incumbent Republican governor and, you know, lost by about 5000 votes, he was kind of hinting at a recount. What am I going to do? And one day Mitch McConnell is like, seems to me andI bushier one, you know, and everybody else sort of reacted, oh, this is over in name.

And it's unfortunate that it took Mitch McConnell to say it. But I but I know he did it once. I don't doubt if Trump lost. Clearly, he might consider doing it again. So I think the Republican reaction will be important. And I think telling and I think the media, you know, we work for the American broadcasting companies are worth noting that if the media is fairly unequivocal about seems that Donald Trump lost, here's where things work that I think you'll see polling showing Donald Trump should leave the White House.

And I think public opinion, I'm not convinced that if Donald Trump says I didn't lose and he lost by clear about that, every Republican will be staying with him forever. I think there are plenty of Republicans who believe in the rule of law and believe they can win in twenty, twenty four. So I'm I'm not worried about this. I think people might react in a sort of normal way. If Donald Trump loses, particularly, he's down by eight in the polls.

I think people are anticipating Donald Trump losing. So I think it's not going to be out of nowhere yet.

Stephen Calabrese, am I saying that name right here is a co-founder of the Federalist Society, wrote an op ed about Trump's claim saying, hey, this is not kosher. You know, I've stood by Trump by a lot and this is way over the line, you know, but that's type of person that, like, might actually split from Trump in the event that he tried to on infirm grounds, like say, oh, somehow I won and justices and so forth.

I mean, there is a process for for playing all this out, right? I mean, ultimately, it goes to the House of Representatives and it goes to Electoral College. Right.

You can wind up with crazy scenarios where Nancy Pelosi becomes president because the Senate, the Senate has not been able to appoint a VP and the House, that we will achieve consensus on who they would pick. Adjudicate is a of the Electoral College. Right then in theory, on January 20th, Trump's term is over. It goes to the line of succession and Nancy Pelosi become acting president. So, you know, interesting scenarios about there are scenarios where whoever the Senate confirms as vice president would become president if the House is deadlocked.

So, you know, interesting times. But, yeah, what happens if Trump, like, physically refuses to leave? The White House, even though the process has been carried out and he lost, right, constitutionally, he's no longer president, I think kind of the further you get from like November 3rd to January 20th, the more you get to January 20th, the more there are clear laws on the books about what happens. Right. It's more like if you have a close election and Trump messes with the vote counting, then that creates issues.

Or if he messes with the perception of legitimacy for the president elect Biden or whatever. Right. But those are, I think, two different grades. Right. Like messing with the actual result in such a way that there's evidence that, like making it harder for people to vote was enough to swing the margin in some key states and change the outcome where some 20000 area where stopping counting when there were ballots before some deadline are not counted. And that would change the outcome.

That's a bigger crisis than merely delegitimizing the next president. I mean, they're both still bad, right? But if you actually had some evidence that, like shenanigans affected the result and that's a bigger problem. Or if you have evidence that like, say, some state appoints electors that don't seem to reflect the popular will of that state. Right. That gets to be a pretty big crisis, right? Yeah, it seems like there's plenty of dry tinder here for people who would take advantage of a scenario where there's already low faith in institutions in the United States.

And that brings us to another question which all of you have brought up at some point during this conversation. And that's the role that we all play in this conversation, in the media, in trying to make a potentially confusing and contentious election less. So what can we do either now or on election night to make sure that we reduce the number of potential problems that come with this tricky election?

You know, one thing that both, you know, Attorney General William Barr and Trump are saying is mail in voting leads to widespread voter fraud. There is no evidence of that, as we've said. But the one thing that I think we're losing sight of is this could be an election with unprecedented mail in ballots. Right. Like Nathaniel Rakic had looked at primaries held here during the Soviet era, nearly every state had an increase from their last previous primary in the amount of ballots that were cast remotely.

I think we need to be telling readers in our audience that this could be a messy election, not because there's widespread fraud, but because of the scale by which people are voting remotely and the types of issues that raises.

Because I feel like Trump has made that conversation fraught. You can no longer have it with nuance or intelligently. But it is something that I think we need to be doing more for readers to understand that this this could be messy and not because of fraud.

I guess if I was Lord of the media, I guess I would say we should think about this is like a two month process of the election we're going to have is supposed to election day. I think October is going to be full of voting early voting by mail. So on October is not every election day and that's always been happened. But now we're really going to have a different kind of election process than November will be counting, not November 3rd.

So you're ninety seven person panel in which you have 97 people who are excited to say what the election means, maybe rein that in because we don't know what the election means if half the votes have not been counted yet. So I think this notion that we have to have this big election night that we've all trained ourselves for, I think is going to have to be sort of unwound in a really aggressive way. Election night is no matter what Donald Trump says election night is, you're probably not going to hear a speech from Joe Biden on Election Day or Donald Trump about who won.

And we may have to wait. And, you know, we've had to just that because, I mean, this New York these New York races have taken a long time. I don't know if we need six weeks. I mean, hopefully it'll be sooner than we have been. I think we need it faster than that. But I just think we have to get in the mindset of maybe we don't know who the winning election until December even. I think that's a realistic possibility.

I mean, I think Trump might try to give an election night speech, right, if he's ahead in the popular vote. Sure. And there are a lot of uncolored states, right. I agree with all that. And the media has a big role here. The smaller point, I think, and this also figures into the conversations, you know, we've had as a group as like people I think would behoove themselves to look at the situation in individual states and what the expectations are as far as timing of different votes coming in and individual states.

You know, I know in Florida, for example, often, right when polls close, they report the first tranche of absentee ballots that were mailed in early. Then they report election night and then they report later returned absentee ballots. So that can create interesting patterns, right. If the absentee ballots are very democratic, you might actually have Biden way ahead. Eight o'clock, and then Trump really catches up or maybe overtakes Biden, but then things begin to shift back toward Biden, right.

Something kind of familiar with these scenarios and how they might play out because they can potentially send false signals in multiple directions, too. But it's not like some random patterns. Right. Apparently a little bit different each day. But like there's some predictability, you know, like different tranches of ballots. And when they're counted, we've never, by the way, before had this big a partisan split and intention to vote by mail or not. It's usually been pretty nonpartisan.

And it's possible that some people who say I'm going to vote in person as a Republican are going to say, actually, I usually voted by mail in a state like Arizona. Here's my ballot. Just easier send in or who knows. But like any type of election night needle or forecast, I think is going to be fraught. Yeah, I mean, we deal with these questions very hands on.

Does that mean we're not going to have an election night forecast on their negotiations about that? If it were totally up to me, then we would tread very lightly on that. I think one thing we might try to do is have scenarios. You can game out, OK, if Biden wins this in that state, then how does the math look then? But like without getting in too much trouble with my bosses and their bosses. Right. Like, yeah.

I mean, I think the very worst manifestation of like kind of jumping to conclusions might be if you try to like take all these unknowns for which you have no real precedent and plug them into like some type of model. Now, I do think if like if you literally are The New York Times and you said we literally are going to devote 15 people to this election night or election month project half time from July 1st through November 3rd.

And they're going to research every nuance of how the vote will come in and talk to secretaries of state and elections officials and like plan for 30 contingencies. If you want to put that type of massive scale into it, then then that might be really valuable. But that would be mostly a reporting project and not like a statistical project. Right. You know, Cisco models work when you have some past precedents to rely on. And we can get into big debates about like how many precedents are enough with sample size is enough, right?

If you have a sample size of zero, then statistics doesn't do much good. Right. You need to rely on reporting to understand what to anticipate a new situation. Yeah, I guess so. The thing I might say is news organizations tend to cover what the president says is news and often their first tweet is president says X, It might not be good Election Day to be like Trump's. If Trump says, like if you go to vote in Florida, you will die of covid.

You may not want to, like, do an alert on that and just have it. A transcript. What the president said, I think in this situation where the president seems to have an incentive to create misleading comments, it'll be incumbent upon the news media, Facebook, Twitter, to be very vigilant about making sure he does not sort of alter the electoral process in problematic ways by spreading misinformation.

So fair to say this will be an election like none other. Sorry, sorry to all of our listeners who might have been excited for a regular election night, but I guess in exchange you get potentially weeks of election night coverage. So hopefully you can live with that.

But I think that. Does anyone have any thoughts as we wrap up here?

I mean, every every election, Gailen, every election. And I say how every election gets crazier. How can things get even crazier next election?

And I wish I didn't think that. Fair enough, election month. Here we come. Oh, my gosh, I mean, fortunately, you can't really take a vacation anyway. So I was going to say, don't plan your vacation for November 4th.

You can't go anywhere anyway. Maybe by the time we actually know who's president, then you'll be able to go to more than two other countries as Americans. So, you know, look on the bright side. All right. Indeed, the bright side. Let's leave it there. Thank you. Thank you, Galen. Thank you very Thanksgiving. And thank you, Sarah. Thanks, Galen.

And some housekeeping. Before we go, I wanted to let listeners know that we are not planning to have a second podcast this week unless, of course, there is emergency news because Joe Biden announces his VP pick. But we have a lot of busy weeks ahead of us with our forecast launch and the conventions and so on. So I'm going to be off. But, you know, we will be back in case of emergency. Otherwise, we'll see you next week.

My name is Galen Droog. Tony Chow is in the virtual control room. You can in touch by e-mailing us at podcast at 538 Dotcom. You can also, of course, tweet us with questions or comments. If you're a fan of the show, leave us a reading our review in the Apple podcast store or tell someone about us. Thanks for listening and we'll see you soon.