Skip to Content
US Videos

Where to Turn When the Political Climate Heats Up

Amy Walter of The Cook Political Report shared her insights at our Morningstar Investment Conference.

Amy Walter: Good morning, everybody. Thank you so much for having me here. Still getting used to life in the virtual world. It would be so much more exciting to see everybody in person. But this is working out quite well. And I hope that we can spend some time here talking about where we are politically and get some questions from you all at the end. My goal here is to provide basically a big-picture analysis of where the race stands, the political climate, and the race for president and Congress stand, as we are now coming in just under two months until the election.

It's always easier to start, though, back at the beginning of the year than where we are at this exact moment, because I think that tells us so much about how we got to where we are right now. What's remarkable about this political environment that we sit in is not that it is unpredictable and chaotic but that it has been, really, for the last four years. We've gone through an incredible number of issues and been through--it wasn't even a year ago that we went through an impeachment of the President of the United States.

And yet through it all, what we have seen is that opinions about the president, opinions about this race have remained pretty stable. In fact, opinions of this president haven't changed all that much since he was first sworn in as president. This is a president who, over the course of his first term, has seen opinions that very rarely--they're in a very tight window--somewhere between at a high mark about 47%, 48%, and at its lowest around 38%. Why does that matter? Well, because as we went through the years, I said through these last three years, all kinds of up and downs, good economy, but some bad lows for the president, especially on issues that he wanted to see push through like the healthcare legislation. The party lost control of Congress in 2018 and then, of course, the impeachment. But even through all of this, his ceiling stayed pretty steady, and his floor stayed pretty steady.

Then came COVID. And of course, it looked as if the world was going to change overnight. And it did change overnight, just in terms of the way that we are all doing the things that we normally used to do. Then, on the heels of that, came the killing of George Floyd. Both of these were crises--and the protests that followed those--those were the first real crises faced by this president. And of course a pandemic, nothing like we've seen in 100 years, is the sort of testing of a leader that comes with the job of being president.

And it was in this moment where the president did seem like he was able to maybe bring in some of those voters who had never thought as highly of him even during the good times. He got a little bit of a bump at the very beginning of the pandemic. And then from April all the way through until now, opinions of how the president's handling this crisis have been mostly negative. About 60% of voters disapprove of the way the president handled or continues to handle the COVID pandemic. Similar numbers of voters disapprove of the way he handled the protests in the wake of the George Floyd killing.

And so what we found over the course of the summer was a president whose pretty steady floor looked like maybe it wasn't so steady anymore, maybe it had some cracks, maybe he was going to fall into a deep trough. Other presidents have. Not that long ago, George H.W. Bush, Jimmy Carter, even George W. Bush both saw their approval ratings at their lowest marks fall into the 30s and, in some cases, the high 20s.

I remember over the course of the late summer talking with a bunch of even Republican strategists--were polling all over the country and they were seeing the president's numbers down everywhere, really red areas, rural areas, suburban areas. And, as I said, they were worried that maybe he had truly broken through that floor. The president was trailing in the middle of the summer Joe Biden by about nine points.

Since then, though, it looks as if his floor is pretty steady. He has now closed the gap a little bit with Joe Biden, it's somewhere between a six-, seven-, eight-point race at this point, if you look at the averages of all the national polls out there. It is still not a great place to be for a sitting incumbent less than two months away from the election, but it is a better place than where he was in July. But, as I said, here we are this close to the election down by this many points in approval rating, somewhere around 42% or 43%, and two thirds of the country saying that they think things are headed in the wrong direction, they're off on the wrong track. All of these numbers are very, very dangerous. We've never seen a president reelected with those three numbers combined or certainly just one of those three, would mark a real, real challenge in coming back and winning reelection.

Now, I also appreciate, though, that there may be a lot of folks listening here who say, "Well, didn't things look like this in 2016? That things looked really, really bad for Donald Trump, that Donald Trump was never ahead in the polling, and yet he came out and he won." Obviously, he didn't carry the popular vote, but he won the Electoral College. So should we trust this data? Should we trust historical norms? Isn't he a different kind of politician? He's been able to defy traditional political gravity.

And it's a fair question. And I just want to address that part right now, too. The first and most important thing to remember about this year compared to 2016 is that Donald Trump is no longer the challenger. He is the incumbent president. And with that comes the responsibility in the minds of voters for handling the issues that are in front of him. This is why, in many ways, it's so much more fun to be a challenger in a difficult political environment--because you don't get any of the blame and you can cast all of it. It's great to be an incumbent in good times. It's hard to be an incumbent in bad times. Even when it's not your fault, you take the blame.

In this case, trying to push that blame onto others is not going to work so well for a sitting president. And it's been clear, as I said, where his approval ratings still aren't moving up, that voters have a pretty fixed opinion about this president. And, in fact, I would argue that this race is entirely about President Trump.

The second, and this goes to the issue both of the polls and also what makes this race different than 2016, is that, fundamentally, Donald Trump was hoping and Republicans were hoping to run against a different kind of Democrat. If they were going to run the 2016 race, well, they wanted to--once again and win by a narrow margin--they wanted to run up against somebody like they did in 2016, who had their own political baggage, who also had very high unfavorable ratings. And, obviously, Joe Biden is not the same as Hillary Clinton. He does not have the high unfavorable ratings. The intensity of dislike for Joe Biden is nowhere near the intensity of dislike for Hillary Clinton. And more important and more related to 2020 is that Joe Biden also isn't Bernie Sanders.

Look, lots of Republicans, including the Trump campaign but all the way down to congressional candidates--Republican congressional candidates--were really hoping--and at one point, earlier this year, it looked likely--that the race would be between Donald Trump and Bernie Sanders. And that the race could be a referendum, not on just Donald Trump and who he is and how he's behaved as president, but a race between a candidate that may not be everybody's cup of tea, a president who's not everybody's cup of tea, and somebody who comes from a much farther left wing of the party than so many voters would be comfortable with.

A referendum on socialism on Medicare for all was the kind of campaign that many Republicans were hoping to run. They're still trying to do that. But it just doesn't stick to Joe Biden like it would, obviously, with Bernie Sanders, who ran on this. And also just because Joe Biden is a well-known quantity. People may not know him intimately. They know he was vice president. They know he's been in Washington for a long time. Trying to turn him into a wild-eyed left-wing socialist--it's just not really sticking. And we've seen that the president has been trying, and his allies have been trying, to make that case. But Joe Biden's favorite overall favorable ratings continue to stay in a pretty decent place. He's not the most liked politician in history, but he's not as disliked as the president nor as Hillary Clinton was.

And finally, the other big difference between where we were in 2016 and where we are now: Voters are much more committed to their candidate than they were in 2016. In some ways, that makes sense. It was an open seat. This wasn't an incumbent president running for reelection. But there were also a lot of voters in 2016 who disliked the two choices that they had. Remember, the Hillary Clinton versus Donald Trump campaign was a campaign between two of the most disliked politicians in modern American history. This year, we have two candidates. While they are--again, their approval ratings are nothing to write home about--both of them are better liked today than the two candidates were in 2016.

That includes Donald Trump. I think you can account for Donald Trump's improving image to the fact that he's got that support. Remember, I talked about that floor that he has. He's got stronger support among Republicans than he did back in 2016, when there was still in a number of Republicans, including a lot of conservative Republicans, evangelical Republicans, who didn't know if they really believed that Donald Trump was going to deliver on their issues. Now, they see that he has, so he has that stronger base of support.

But why is it important that these two candidates are better liked and better defined? Well, there are fewer people now that say that they're undecided or that they're thinking about voting for a third-party candidate. At this point in the 2016 election, about 17% of voters said that they were either undecided or they're thinking about voting for a third-party candidate. Today, that number is 6%. What that means is that there's less volatility that's possible in this race. So you don't see the sort of spikiness that you saw in 2016. And as we get closer to the election, the more [inaudible 00:12:54] candidates are, the less likely they say they are to switch from one candidate to the other or vote for a third-party candidate, the harder it is for Donald Trump to catch up.

What he needs to do right now is to be able to pull voters away from Joe Biden. And the best way for him to do that is to make Joe Biden look unacceptable, like a greater risk than a vote for Donald Trump. And thus far, as I said, he has not been particularly successful in doing that. Joe Biden is a much tougher candidate to caricature as Hillary or Bernie Sanders would have been. 

So where does that leave this race right now? Well, it leaves the president in a very precarious position where he does need a lot of things going his way in order to win this race. The first is that there's another intervening event that makes everything that we're talking about right now somewhat insignificant because this other thing is such a big deal. I don't know what that is. I'm not pretending that there's some big October surprise around the corner. But obviously, we live at a time where nothing that we're doing right now was anything that we had predicted back in 2019.

But even so, as I said at the very beginning, even events that would seem to be game-changing events do very little to change the contours of this race, in large part because this is a race about Donald Trump. And very few people are ambivalent about him, and very few people's minds are going to be changed about him. So can Joe Biden then lose this race? There are plenty of opportunities. Plenty of challengers or candidates who were front-runners have gone on to stumble in the final weeks of a campaign. We know we have a debate coming up--the very first debate between these two candidates at the end of September. There's a lot of pressure on Joe Biden, and the Trump campaign is arguing that he's going to struggle. You hear Donald Trump calling him "Sleepy Joe," essentially accusing him of being senile, not really knowing where he is at any given moment in time, what day it is, what time it is.

The problem with that strategy, though, is that it lowers the bar for Joe Biden. It says to voters, "Well, gosh, this Joe Biden character, if you listen to Donald Trump, he sounds like a mess." But then if Joe Biden comes out and is able to basically clear that very low bar, it doesn't leave Trump with many arguments then left, that Joe Biden's not ready or prepared for this job. 

The other big question mark, and this is what a lot of us are struggling to understand right now. We're not actually going to know a whole lot before the election. We'll have some ideas but: Who's voting and how they're voting? And we have, we know, more voters than ever who say that they plan on voting absentee. Voting by mail. For many of them, it's because they are worried. They're worried about going to the polls. We saw during the primaries a record number of people turn in ballots by mail. We also know it's because it's a political statement. The president has derided vote by mail, absentee voting, saying it's rife with fraud. The Attorney General has also weighed in, saying that there's possibility that your ballot is not so secret, that people can know how you voted, the people who are tabulating these ballots--so stirring a lot of disinformation about the voting and how it works.

And so what we're seeing as well is Republicans overwhelmingly saying that they don't want to vote by mail, they're going to vote in person. So why does it matter if you vote by mail or in person? A vote in October is the same as a vote in November, which is true, except there are risks to both of these. The first risk is that if you are a first-time vote-by-mail voter, you may not read the directions as thoroughly as you should have. And we saw a lot of ballots thrown out in the primaries, for example, because of people not signing in the correct place, but mostly because they failed to meet the deadline that, in most of these states, you have to have your ballot in to the election office by Election Day. If it slips your mind and you throw it in the post office box on Election Day and it doesn't get there in time, it's not going to count. So that's a risk that you take. And the other risk is by suggesting that somehow these polling places are not safe, maybe voters who just didn't get their act together to get an absentee ballot, voters that you want to get their votes because you know they're on your side, state, "I don't know if I'm willing to take that risk to go and vote. Democrats keep telling me I should have voted by mail, and I didn't. So I guess I'm just not going to show up."

The risk for Republicans, by only going to vote in person, is: What if we do have a spike in some of those states of COVID? What if there are people who say, "I am much more worried than I thought I would be about voting in person." And of course, voting in person, it means you only get one chance to get those folks out to the polls. And we all know, on Election Day, things happen that you didn't plan for. The dog gets really sick, and you have to go to the emergency vet. Something happens in your family, or just, quite frankly, you forget. Seems hard to believe, considering that politics is being piped into our brains 24/7, but there's a risk then in only counting on voters going to the polls.

But the other question that we don't quite know is: Who is voting? Because so many people now are going to get these ballots. Either they request for an absentee ballot--in some states, like Nevada, for the very first time, every voter is going to get a mail ballot delivered to them to their address. So, people who might not have been part of the process could come in and vote. What the president and his allies are hoping as well is that the polls are underestimating the intensity of not just the Trump voters, people who voted for Trump in 2016, but voters who were not engaged in the process back then--maybe they're registered to vote, but they just didn't show up in 2016, that they identify with Donald Trump culturally, socially. They see what's going on in the country. They feel more motivated to show their support for him. And they show up and vote.

Now, because they haven't voted in the past, they're not really included as likely targets or the kind of voter that could potentially turn out. And so they get underestimated. What I think ultimately is going to happen, though, is we're going to have a record turnout. And that means that both sides are likely to see all kinds of people show up, that maybe they hadn't planned on coming to vote in the first place. So that adds a little layer of uncertainty. But as I said at the very beginning, what we're seeing in the polls is a clear trend, a trend that's a little bit better for the president. He's not down as much as he was earlier this summer. But nonetheless, this six- or seven- or eight-point lead of Joe Biden's is big enough nationally to give him an Electoral College win.

Remember Hillary Clinton in 2016. The national polls said she was somewhere between one and three points ahead of Donald Trump. By the time all the votes were counted, she ended up winning by two points. So the national polls were right on, spot on. Two points, though, wasn't quite enough. Got her a popular vote, didn't get her the Electoral College, although even winning by just two points got her within 80,000 votes. There were about 78,000 votes cast across three states, Michigan, Wisconsin, Pennsylvania, had they flipped to Democrats, Hillary Clinton would be the one running for reelection this year, not Donald Trump.

So if the national polling average continues to shrink, and you see Biden up by only two, three, four points, well yeah, the potential then for Trump to win an Electoral College by a narrow margin, like he did in 2016, is certainly much higher. You could give him a shot, a really good shot, at doing that. But when the polls get up above five points, what we know is that there just aren't enough states in the Electoral College map, battleground map, that are that far to the right of the national average.

If you think about the national average, let's say here's the national average, states like Florida, Pennsylvania, Wisconsin, a little bit right of that. And then places like Georgia, Iowa are much further to the right of that. But they're not eight, nine, 10 points to the right of that or six or seven points in the case of some of these battleground states like Arizona or Pennsylvania or Wisconsin. So the president really needs to shrink it down much further than that.

Before we we go to questions, the other thing I wanted to just touch on is what Congress could look like in this next election, after this next election, and what that would mean. We know that when it comes to presidential election years, especially now, fewer and fewer voters are splitting their tickets. In 2016, for example, we did not see one Senate race where the candidate, the Senate candidate of one party carried a state that his or her presidential nominee did not. So in a state that Hillary Clinton carried, no Republican senator was successful, or candidate was successful, who was up for reelection. States that Donald Trump carried, the Democratic candidate was not successful in winning a Senate seat. So, how the presidential race goes is going to be very important. 

But really, we're going to be focusing in on five, well, it's probably more like six or seven states. Georgia gives us two races because they have a special election and a regular election. North Carolina, Arizona, Colorado, Maine--those are the key battlegrounds. Those are the states that Democrats are counting on winning. If they win those four, it is almost certain that they will win the majority. But they've also expanded the playing field into places like Iowa, Montana, and, as I said, Georgia, where the Democratic hopes are really resting on Biden doing well there or Donald Trump not doing well depending on your perspective.

But there is a very good likelihood, a little better than 50% chance, that Democrats do have control of the Senate and the House. That, combined with control at the White House, would mean for the first time since 2009, Democrats would have total control in Washington. But we also know it would look different than 2009. The majority, the number in the Senate would be very narrow, anywhere from a 50-50 Senate with a tie-breaking vote by the vice president. It would be then Vice President Kamala Harris. Or maybe if it's a really good night for Democrats, they have 51 seats. But as we know, that's not enough to break a filibuster. And as such, there's going to be a lot of discussion as we get into the heart of 2021 about whether Democrats are going to keep the filibuster, whether they will vote to abolish it for other purposes, or whether we'll even be having this discussion.

Personally, I think as we get into 2021, a couple of things are going to be really apparent. And it makes a lot of the discussion here kind of around the edges of: What are Democrats going to do if they take control? To sort of put that in--I wanted to sort of put that in perspective--if indeed, we are going into the January, February of 2021, let's think everything's going as planned, we have a vaccine, and it's ready to go. Well, now that's going to be the main focus for the White House is making sure that everything goes smoothly in its production, its distribution. The other thing we know is, we could see a lot of economic devastation that comes to bear in 2021. We know that state and local budgets are being hit really hard. We know that we're going to have a number of businesses that are going to be declaring bankruptcy. We know that people are going to be defaulting on their mortgages, and we're going to still have conversations about people being evicted from their rental properties. And, if we continue to see the sort of K-shaped recovery, a discussion about what's happening to those who are the most vulnerable in the wake of the pandemic. And so those to me are the issues that a new president is going to have to tackle and the new Congress is going to have to tackle before we can get into the sort of wish list of items that folks are going to want to see from their party. So, I'm going to leave it there so that we have enough time to get some questions from you all, and hopefully, answer anything that you all are thinking about.

Aron Szapiro: Well, thanks so much. This is Aron Szapiro, head of policy research here at Morningstar, and I'll moderate the questions, and lots of them have come in. Let me try to synthesize a few of these into one question. To what extent is Senate control going to be a key variable in a potential Biden administration's ability to govern? And I've gotten some kind of specific questions here. You know, if there is a Democratic-controlled Senate and a Biden presidency, do you think the corporate business tax will be raised? If it's divided, does it make it more difficult to advance the kind of fiscal policy response that Democrats have been proposing to deal with COVID? So let me kind of, I know it's a big question, but let me get your take on that.

Walter: They're very good questions. And I think we have two issues here. The first is one, as you said, Democratic Senate control--let's say that it is divided. You have a Democratic White House, a Democratic House, a Republican Senate. We kind of know then what the challenge is going to be for a White House being able to balance a very Democratic House and what it wants to get and then somehow make its way through the Senate.

Even if it is a Democratic Senate, as I said, it doesn't have 60 votes. So a lot of this stuff will either have to be done through reconciliation, which is procedural way that, for example, the tax cuts, the most recent round of tax cuts in 2017, made their way through the Congress, made it through the Senate. 

As I said, it's also possible that early on this debate about the filibuster starts and ends, and Democrats decide that it's worth abandoning the filibuster. I don't know that I agree with many who've said that that's a definite thing that's going to happen. I think there are a lot of senators on the Democratic side who represent the swing or more-moderate areas, many of whom have already come out and said that they don't want to see it eliminated completely. They might want to see some rule changes, which would mean that perhaps it's just sort of massaged but not completely gotten rid of. And finally, I think the White House is probably--a Biden White House is probably going to look for doing as much as they can, much like this administration has and the Obama administration did before that, as much as they can through executive order. And there is plenty that, as we've seen under this administration, and--what's the word I'm looking for--and the repercussions of that. They've been able to not suffer repercussions from that. They've been able to do a lot through that means of executive, some people call it overreach, but executive authority to do many of these things.

Szapiro: Another area we're getting a lot of questions: There are a number of commentators who are nervous about a close election and civil unrest that might result from that. And I guess I would add, you even see people like David Brooks of The Times writing columns predicting that that is the most likely outcome, but given the distribution of plausible events, ranging from a Trump victory to a Biden landslide or a Trump landslide, how likely is there an outcome where there's civil unrest and things hinge on fights about signatures on an absentee ballot and the like?

Walter: I know. I'm hoping, as with many things in life, the more you prepare for something to happen, the less likely it's to happen. So that we overprepare for this potential thing happening. We know that the states, for example, are doing a great deal of work getting themselves prepared as best as they can, for any potential wrinkle in this new way that we're voting system. We know that both sides are putting lawyers and all kinds of experts on the ready should something that, as you pointed out, is coming down to 200 ballots in one state occur, that they're prepared for that. I just think we live in these times where what has been so predictable is the unpredictability of it, that we are so exhausted by the current climate we're in that people are so on edge, that there is a fraying around the edges of our sort of societal net, that it doesn't feel that crazy.

And I'm hoping that I'm wrong. I'm hoping, as you pointed out, that we're kind of getting ahead of ourselves. But what I think is a real worry, whether this turns into literally riots in the streets or extended election handling in the courts for weeks and weeks or maybe months, what worries me more is what's going to be happening online. And let's say that election night ends, we don't have a declared winner until the weekend, there's a period of time there where a lot of conspiracy theories are going to be knocking around. Many of them will be put there on purpose by the Russians and others to sow this level of discord.

And what this means is even if all goes well--the new president is sworn in, there's no violence, nobody's holing themselves up in the White House, the military doesn't need to be called out, all of that stuff--the damage that's being done to the integrity of the system, of all the systems, the election system, the presidency, the ways in which we think about just casting a ballot, once those are damaged, they're very hard to get people to trust them again. And the foundation shifts further and further. And that's the sort of stuff that I worry about. If 40-plus percent of the American public believes not just that the person in the White House is there illegitimately--the person who won is not really the winner, it was rigged or whatever--that's not good. But if they also believe that the system in and of itself is so broken and is so not to be trusted, then we're headed to a very, very dangerous place.

Szapiro: Well, on a more uplifting note, I'll change gears slightly ...

Walter: Sorry, sorry, I don't want to be a downer, but ...

Szapiro: It's great to get the perspective. A number of people have asked: What would your crib sheet be for what you should look for on election night? As people are sitting in their living rooms, if they see Florida called early, either way, does that mean it's over? Does it only mean it's over for one candidate or the other? And I recognize that things may change over the next 47 days. But you know: What would your crib sheet be today for what to look for on election night?

Walter: Great question. And I think Florida is a really good place to start. It is very hard for Donald Trump to win the White House without winning Florida. I mean, it is literally the backbone of his Electoral College skeleton, I guess we'll say if we're going to keep within that analogy, and so to lose it, it really is all but impossible to see him win.

Now, the good news about Florida, despite its role in the 2000 snafu election, is that majority of people there vote early or vote by mail. This has been going on for quite some time. So, Florida has a good history, they know what to expect. They know how to count these votes, and they can count them before Election Day. See, this is the real challenge for many of these states. It's not simply that they're going to be overwhelmed with vote by mail. It's that in a place like Pennsylvania, you're not allowed to start counting those, processing those until Election Day. So they start to pile up. And no matter how well prepared you are to process those ballots, it just simply is not possible to do it in a 24-hour period.

Florida and North Carolina, also on the East Coast, are able to at least start processing them and then they can tabulate them very quickly once the polls are closed and can pretty much, I'm not going to guarantee it, but there's a pretty strong likelihood that we'll know the winner of those two states. As I said, if Biden wins both of those, Trump loses both of those, it's all but impossible. And while we may not know the final results in the Midwest, Pennsylvania or Michigan, because of those things I just pointed out, that's going to take a long longer time for them to tabulate their absentee ballots, we'll know what the story is.

If Donald Trump wins both of those--look, he's got a good chance to win. Not guaranteed, he can win both of those and still lose, it will come down then to this path in the Midwest. And then the one state that--unfortunately if you're here like me on the East Coast--is, I think, going to be the most important is Arizona. And I say "unfortunately on the East Coast" because obviously we're not going to know a lot by 11 or midnight on East Coast time. They're also a state that does primarily vote by mail and early vote. However, they're also a state that traditionally, they've taken a while to count all the ballots. So, if we could go to bed on election night with some of the big states called, as I said, the Florida and North Carolina, but without Pennsylvania, without Arizona, without a Michigan, Wisconsin, it's going to be impossible to call it or to know exactly what the outcome is going to be.

So look for those two states early on. And let's hope that because these secretaries of state, I've interviewed a lot of them and these election officials, they really are focused on getting this right and being prepared, as we discussed at the very beginning of this Q&A session, being prepared for the unpredictable and whether it's thousands of absentee ballots or thousands of people showing up to vote early or vote in person on Election Day, that all the worries about a potential election night collapse won't come true.

Szapiro: Another question I've gotten here a few times that I've also been mulling is: Most of the models that are out there, and the sort of weighted average of polls or straight average of polls, show a lead for Biden--not an insurmountable one--the betting markets have it at somewhere around 50-50. So I guess the question is, why do you think that is? Are people hedging? Are they actually using these betting markets for protection? Or just what's going on there?

Walter: It's a really, really good question. And I think it makes a whole lot of sense when you see what happened in 2016, that everybody looked at the polling averages, and they said, "Well, of course, Hillary Clinton's going to win." And then she didn't. And the response to that, instead of, "Well, what did happen? What did the polling get right? What did it get wrong?" it was, "Well, I guess, polling doesn't capture anything." There is this belief that the polling was so wrong and so awful and so bad, that actually--do the opposite of what the polls are telling you.

What we learned after 2016, and there was a lot of postmortem analysis of the polling--by the polling industry, by the Association of Pollsters, by individuals who do their own polling and aggregation. And I don't want to get too deep in the weeds, but I think, the following things that they found. The first is, as I said, the national polls were right. It was the state polls, in part because we just didn't have that many state polls. And even the campaigns themselves weren't doing, the Clinton campaign in particular, they weren't doing polling. They were doing analytics. And that missed the shift that happened over the weekend, those last four or five days.

Remember I said, during my talk, that there were so many undecided voters. Well, those voters broke, they broke late, and they broke overwhelmingly to Donald Trump. Nobody picked that up because nobody was polling. And I don't think you're going to see that same mistake again. There was also for the first time a real recognition that there is a new gap in this country. We've always had a gender gap. We know we have a gap in how people vote based on what race they are, whether you're white or African American. But for the first time, white voters' divergence on political candidates based on their level of education became really clear. And so now, pollsters are weighting on education, whether you have a four-year degree or whether you don't.

But I also think, and for people who work with data can appreciate this, that we as humans, are often not very good when it comes to probability. And if you looked only at those charts and those models that showed you the percent chance, the likelihood of Clinton or Trump winning, you saw overwhelmingly that Clinton was favored anywhere from 70% to 95%. Now, you know, intellectually, that a 70% or an 80% chance means that it is not guaranteed to happen. But unless you start seeing numbers in a 70%, 80% category, you say, "Well, of course, it's going to to happen, right?" If there's an 80% chance of rain, it always rains. It's very rare for it not to rain. Everybody makes the extra point in football. There are those rare moments where they don't, but come on, you get the extra point. But nobody focuses. They focus much less on the possibility that it might not happen. And that, I think, is where we're spending so much time today. I think if the mistake in 2016 is we spent so much time focused on the 70% or 80% chance of what was going to happen, today, we're spending a lot of time on the 20% to 30% chance of it not happening. Both of those aren't a really good way to think about the world.

The other thing I think that's happening is--look, as I said, Donald Trump still has his base. There is not a collapse, a total utter collapse of Trump support. He's losing support among certain groups. He's not doing as well among seniors, among independents, among women as he did in 2016. But there are a lot of people out there, especially those who are Biden supporters or who, more likely, are just Trump dislikers, who cannot, for the life of them, understand the people in their own lives who are still voting for Donald Trump. To them, these last four years have been a vindication of everything they did in 2016. "See, this is why I voted for Hillary. See, this is why I didn't vote for Trump. Don't you see that?" And then their friends or family, people in their lives say, "I think he's done a great job. I don't understand. Of course, I'm voting for him."

And that gets translated into, "Oh, he's going to win. I mean, if these people are still voting for him and he won, well then he's going to win again." I do think that's a lot of where these markets are, and finally, that Donald Trump does seem impervious to the rules of the game, the normal rules of politics. He says and does things that nobody else could get away with. And he doesn't seem to suffer political consequence. And until he does, or unless he does--not going to believe it.

Szapiro: I think we have time probably for one more question just coming up on the top of the hour here. And I thought this was an interesting one to close out on as we look forward over the next few weeks. Has a debate performance ever had a meaningful impact that was lasting on the polls?

Walter: You have to go all the way back probably to 1980. And the problem with going back that far in history is that the country was also at a very different place. We weren't as polarized as we are today. We didn't have social media. We didn't have 24/7 cable news. But it feels kind of like that today, only in that you had in Jimmy Carter at that time an incumbent who was really struggling. I mean, deeply underwater. But an opponent who Carter had labeled as not ready for primetime. He was too old. There was a lot of worry that well, this guy, a Hollywood actor, and he's kind of too radical. Could he really fill the job of president of the United States? Is he up for this challenge? And so there were still questions in a lot of voters' minds about whether this guy was ready for the big time, for the big job.

And there was that one debate. The two of them face off. And Ronald Reagan did an excellent job. And after that, what you found is that the bottom did fall out for Jimmy Carter, and we obviously know the outcome. But I think it's fair to say, yes, that there's some parallel, but at the same time, so much of what we know in 2020 about this race, or I guess maybe a better way to say this is: We live in a time where it's harder and harder to change people's minds because they are inundated every day with influences that only serve to solidify their current point of view. So whether it's the people that you follow on Facebook, what you're watching through news programs, maybe that you only talk to certain types of people about politics, more likely than not what's going to happen with this debate, and debates plural, is people see what they want to see from those debates, which is why it's always challenging for me when people say, "What did you think about the debate? What do you think happened?" I always say, "Well, it doesn't really matter what I think" because my own personal and objective sense of this is not going to sound right to a whole lot of people. How they perceive it is what's the most important.

Szapiro: I think that's a great place to end. Amy, thank you so much for sharing your perspective, for the overview.