Chapters
Shownotes Transcript
I like the idea of having enough kids that they can raise each other. Oh, wow. Wow.
I'm upstate at the moment and my grandmother's like telling me stories and she is one of 13 children and she was telling me how the kids started naming the children. After birth, it was like, it's your problem now to the rest of the kids. And it was like, name it, feed it, figure it all out. I'm telling you, earlier generations, man, they were tough. Yeah. I don't know. I have one kid and I am barely hanging on. So Galen's great grandma, props to you. Truly a saint.
Hello and welcome to the FiveThirtyEight Politics Podcast. I'm Galen Druk. It is officially election season. Labor Day traditionally marks the time when general election campaigning truly ramps up. Summer vacation is over, TV ads flood the airways, and pollsters switch their models from registered voters to likely voters. Over the past couple months, we've seen a historically rare advantage in polling for Democrats. Over
Over the coming two months, it's a big question whether that advantage will remain. In past midterm cycles, the party out of power has traditionally gained ground heading into November.
Today, we're going to talk about what history suggests we can expect from these final two months of the midterms. And we're also going to play a little game of midterm trivia to jog our memories about how past midterms have gone. For example, what percentage of Democratic House ads mentioned health care in 2018? And what year was the largest midterm landslide ever?
We also have another good or bad use of data to discuss. We're going to look at how news outlets reported the latest U.S. life expectancy numbers, which have now fallen two years in a row. Here with me to discuss it all is politics editor Sarah Frostenson. Hello, Sarah. Good morning. Hey. Good morning. Also here with us is senior writer Amelia Thompson DeVoe. Hey, Amelia. How's it going? Good. How are you, Galen? I'm
I'm doing well, you know, easing back from a pleasant Labor Day weekend and trying to get my brain into the headspace that I just described, which is the final two months of an election season. But, you know, I think good, all things considered. Also with us is elections analyst Jeffrey Skelly. Hey, Jeff, how's it going? Hey, Galen. It's going, man. It's going. It's going. Oh, my God.
Love to hear it. Okay, so I want to make sure we have ample time for our trivia game. So we're going to dive right in. For today's good or bad use of data, we're going to talk about how long Americans live.
Last week, the CDC published the latest life expectancy numbers for 2021, and it showed that average life expectancy of Americans had fallen again. Life expectancy fell to 76.1 years from 77 years in 2020. That had already fallen from almost 79 years, so nearly three years in the span of two.
This got a lot of attention, and these numbers are pretty nuanced, especially when you dig into the crosstabs. So I wanted to break down how we should think about this using some of the ways that it was framed in the news.
So the first example I want to look at is from Stat News, which has statistics in the name. You would think they would do a pretty good job of reporting out examples of data. So in the lead, they write, Americans born in 2021 can expect to live for just 76.1 years. The lowest life expectancy has been since 1996, according to a new government analysis published Wednesday.
This is the biggest two-year decline, 2.7 years in total, in almost 100 years. The COVID pandemic is the primary cause of the decline. However, increases in the number of people dying from overdoses and accidents is also a significant factor. This confused me a little bit, and hopefully you can help me understand.
How should we think about life expectancy numbers in practice? Because if this is mostly about COVID, is this really an impact on the life expectancy of people being born today?
I think it's like, it's a little hard to wrap your head around. And the first thing I'd say is that the most important thing to remember is that life expectancy numbers are not individual level predictions and they're not fortune telling or sort of like destiny. You know, your destiny is not written in the stars in the 2021 life expectancy numbers. I think a better way to
A better way of thinking about it is that it's more of a snapshot of the collective health of the country, everyone who's alive right now on average. So when you dig into the tables of life expectancy reports, you don't just see the life expectancy for someone who was born in 2021. You also see the average years left for people at different ages. So
it's more nuanced than just saying, you know, someone who is born in 2021 on average will live 76 years because obviously like a lot of things will happen between now and then and life expectancy numbers do change. But I think the bigger takeaway of what we're seeing from these numbers is that there has been a big shock to the collective health of the country in general,
the US population on average is not living as long or not projected to live as long as they did even a few years ago. And I think most crucially, you know, we saw this trend in the first year of the pandemic and it hasn't reversed. It's actually continuing to go in the same direction, which is it's continuing to decrease.
And so I think that direction of the trend, as much as anything, is one of the most important takeaways from this report. But I'm curious to hear what you guys think. Yeah. I mean, that's what really surprised me is that not only was life expectancy lower this year, it was lower than it was in 2020. And, you know, the big difference there, as we all know, is vaccinations. You would have thought that maybe things would be getting better, you know,
This is a different example, but one thing we have seen in regards to violent crime is, yes, that also spiked during the pandemic, but now cities are kind of reporting there could be declines now. And I would have expected something similar here with life expectancy. I do think one thing that's complicated by this, and the Stat News article did address this earlier,
is that it's not just COVID. You know, there are also a number of deaths that are due to overdoses. And this is something that the U.S. has been battling for a long time now. I think we can all remember the Angus Deaton case paper from, you know, the early 2000s here, kind of talking about life expectancy reducing, how opioids were playing a factor in that. The situation's only gotten worse in different parts of the country, affecting wider swaths of
groups of people. And I think, you know, COVID, it's something else that is kind of often an underlying condition that exacerbates other diseases. And so perhaps that's a factor in this as well. But the fact that it's continuing to deteriorate, that surprised me.
Yes, sir. Just to add to what you said, according to the CDC data, 16% of the decrease in life expectancy came from unintentional injuries, which is sort of a broad category that encompasses a lot of overdose deaths. So whether that's related to COVID and sort of the knock-on effects to society of the pandemic and shutdowns and things like that, or just
extending a broader trend in terms of increased overdose deaths, you know, that is a significant chunk of the decrease in life expectancy as well. Jeff, I'm curious how you saw this data in the context of
Thinking of like, if a person is born today, their life expectancy is 76 years. And I should also add, you know, at this point, according to epidemiological data, it looks like over 80% of Americans have had COVID. A lot of Americans who haven't had COVID are vaccinated. And so I'm wondering if, should we expect this to eventually snap back closer to where it was?
in the way Sarah described, you know, maybe it wasn't going to happen in 2021 because it took a long time for a lot of people to get vaccinated. There was hesitancy and amongst the people who are hesitant, probably a lot of them have gotten COVID at this point. Right. I mean, actually that Stat News article also described life expectancy as a measure in a way that I liked, which is it's not a prediction for a single individual, but it's something of a check engine light for society's health as a whole.
So you saw it go down in 2020 with obviously the pandemic swirling, and then it continued to go down in 2021. And some of the other data that we looked at has shown that
Whereas in some other developed countries, you saw a bit of a snap back in a positive direction, you know, life expectancy jumping back up a little bit in 2021. It continued to decline in the U.S. And, you know, maybe that has something to do with response to actually getting vaccinated, to masking, to other COVID cautionary measures, because obviously that became a big conflict in this country about how to approach COVID.
Sort of what came to my mind immediately was the Spanish influenza back in sort of 1918, 1920, that period of time. And you actually did see, obviously, World War I was going on, and that was a factor in a general decline in life expectancy. But if you even just look at
the life expectancy of women who, for the most part, were not serving in the military. There was a noticeable decline in life expectancy during that period because the Spanish flu was killing millions of people. It was a huge decline. Yeah, yeah, massive. It was like 12 years, right? Yeah. But it snapped back. I mean, I looked at that same data, Jeff, and it just looks like it's like a V. Like it's like 1917, boom, down 12 years, boom.
boom, back up. And then I think 19, correct me if I'm wrong, but I think 1919 was actually, the life expectancy was actually higher than 1917. Yeah.
Yeah, it may depend on which particular group you're looking at or whatever. But the point was that it did snap back. And so that suggests to me, you know, it's sort of it could go in two directions. It's in the United States at least. It's either has COVID become so endemic and that going forward, are we just going to have minor variants and it's going to be just like a, you know, a somewhat more deadly flu variant?
and things do snap back to some extent. It's not that COVID is the entire story, but clearly it's a big part of it. Or, you know, is COVID or other potentials
viruses, epidemics, which is something that gets talked about, like we may be a greater threat for that moving forward. Is that going to be a thing that continues to affect life expectancy such that we don't see that same degree of snapback or even, you know, a noticeable snapback? And I mean, who knows? We're not auguries. We can't read the signs on this exactly. But nonetheless, like the history would suggest that it will snap back to some extent. I
I find the comparison fascinating.
to the early 20th century challenging though, to be a little bit of a devil's advocate cynic. If you look at the, what was happening with life expectancy around the time of the Spanish flu, it was increasing quite rapidly. And I mean, like makes sense. There are huge advances in health technology, huge advances in basic sanitation, like things like chlorinated water are being introduced around this time. Um,
antibiotics are being discovered. So I think the snapback makes some sense just in the sense that like this was a time when a lot was changing with life expectancy. As Jeff mentioned, there was also a world war, which sort of complicating things as well.
And things were really kind of on the upswing. And that's not what we've been seeing with life expectancy in the U.S. in the 21st century. I mean, it's been sort of pretty stable. Like the fact that we're now seeing changes in terms of years rather than months is noteworthy. But there's been some up and down in the top line life expectancy number over the past decade.
The fact that there's been some down has been really noteworthy because I think there's like an expectation in modern society that life expectancy should continue to go up. We should continue to improve. You know, like we're not going to like live to the age of Moses, I don't think. I think we're going to live forever according to Jared Kushner, right? All right. Well, it's different sources. Yeah.
Indeed, indeed. People say different things, you know. Yeah, but so we're already in a context where, you know, overdose deaths are playing a big role in certain populations. There were some studies a few years ago showing that just like deaths at midlife
had increased a lot across the board. And the one other thing I want to throw out there when we're thinking about the pandemic is we're not just talking in terms of health outcomes about deaths that are directly due to COVID, although obviously that's a huge amount of what's going on here and a huge amount of the tragedy of the pandemic. But also people deferred health care during the pandemic to a really significant extent.
I mean, it was, you know, felt dangerous to go to the doctor for a long time. It was hard to get elective procedures. And a lot of, you know, a significant amount of what influences life expectancy in this country is chronic conditions. And so if you have a chronic condition or you have an undiagnosed chronic condition and you weren't getting health care for a year, two years, then
That could have really significant health impacts on the U.S. population that I think we don't even fully understand. I mean, we're already not a super healthy country.
population compared to other wealthy countries. And I think the idea that Americans just kind of might be even less healthy because of the pandemic for reasons that are actually secondary to COVID and are more about health care access is something that we should think about and I think makes me a little bit more pessimistic about the snapback we've been talking about.
Yeah, no, I thought that was an interesting point in the stat article that around 2010, according to the CDC, that's when life expectancy in the U.S. started to stagnate. So it is different, as you were noting, Amelia, from what we saw with the influenza in the early 1900s, where like life expectancy was increasing, like skyrocketing.
Maybe America is on the decline and you see that in terms of mortality. I mean, we know from studies like where you live is increasingly tied to all kinds of health outcomes and educational outcomes and wealth. And I think some of the great inequality we see in this country is reflected then in our mortality rates as well.
Yeah, it was interesting. Since 2010, this latest CDC data showed the life expectancy gap between men and women had grown by a year. And if you break down looking at the crosstabs, this most recent data, you see a couple trends. That's one of them, the widening life expectancy gap between men and women. Another is that Native Americans and Alaskan natives saw the largest decreases in life expectancy in 2021.
Next was white Americans. And then after that were black and Hispanic Americans who had seen larger life expectancy decreases in 2020 than white Americans. And so you can break it down a bit and find some interesting trends, which brings me to
So according to the New York Times write-up of this, quote,
The reduction has been particularly steep amongst Native Americans and Alaska Natives, the National Center for Health Studies reported. Average life expectancy in those groups was shortened by four years in 2020 alone. They go on to quote Noreen Goldman at Princeton, "...the continued plunge was all the more upsetting because it occurred after a successful vaccination campaign," she said, adding, "...the Native American population did quite well in the vaccination efforts,"
And that made us feel that 2021 would not be as devastating as 2020. That was wrong, and it's pretty hard to swallow. So reading through this article, I was thinking, is the suggestion here that the vaccination campaign didn't work amongst Alaska Natives and Native Americans? And so I went to the CDC data to look at where the reduction in life expectancy fell, and it was actually...
21% of the decline in 2021 was due to COVID versus 50% for the overall population. Another 21% of the decline was due to unintentional injuries, i.e. overdoses. Another 19% chronic liver disease, 5% suicide. That's not what percent of deaths, that's what percent of the decrease in life expectancy is attributable to those things.
So it seems like, you know, this population in particular is experiencing something quite different from just there's a pandemic in the United States. You know, of course, overdose deaths and suicide and things like that may be exacerbated by measures we took as a result of the pandemic.
But I was curious sort of what you made of that distinction in the data when you look through the crosstabs and actually saw that different things are impacting, you know, different Americans. For example, like more of the decrease amongst in life expectancy amongst men is attributable to suicide and overdoses than women and things like that.
I think it's what you're citing here, Galen, that makes me kind of land on the bad use of data for this. It's not that I don't think we are touching on a real problem, a real decline in mortality in the U.S. I just wish it had been framed more as kind of a mystery because it's not clear cut and it's not affecting different types of Americans in the same way.
And I think that's what needs to be probed more is that, yes, the pandemic has had an effect in terms of life expectancy. I'm not refuting that. But there are also underlying factors. The fact that, you know, life expectancy increases were stagnating as of 2010. You know, that's 12 years ago at this point. We know from other research that America has a problem with suicide. That's the top way that people die by gun violence.
We know that America has a growing problem with drug overdoses. We know that that impacts different communities differently. And yes, the pandemic, as you were saying, with isolation and being alone for so long, probably exacerbated that in some ways. But it's also separate from the pandemic. And that's where I wish some of the coverage had started to kind of engage with this more as a question of like, what is really happening here?
What can we say about what we're seeing from the pandemic versus other contributors? And I think it was like the other contributors, namely overdoses, was kind of buried in the coverage. Like you had to read deep to get that point. And I think that should have been higher because I don't think that problem is going anywhere anytime soon. Yeah, like there's kind of two stories here and you have to tell them both of those stories with care because a lot of the COVID decline
is liable to snap back. I think there's some debate there about how quickly and in what exact ways, but is liable to snap back, especially when we look at other countries. Some of those other things may well not. Right. Yeah. I mean, I think given the trends we've seen over the past decade, you know, why would they? I guess my last question here before we move on is almost a cynical one, or I don't know if I like hesitate to ask it because it seems so morbid, but
How does this all play into politics? I mean, this is a somewhat new development for a country where life expectancy has been increasing decade after decade. Of course, it has slowed, you know, before now. But is there a political component to this, given that we're a politics podcast?
The question that I have is about health behavior and whether that is increasingly something that is correlated with partisanship or influenced by partisanship.
It feels like almost a crazy question to ask, because why would something like going to the doctor be affected by whether you're a Republican or a Democrat? But we've talked a lot on the podcast about how increasingly all kinds of aspects of people's identities and behaviors are shaped by their partisan identity. And we already saw some differences in how people approached health behavior even before the pandemic started.
I was looking back at an article that I wrote in the aftermath of the 2020 election with FiveThirtyEight contributor Meredith Conroy. And we were talking about why sort of Trump's machismo and bravado around COVID-19 might have resonated with so many men. And part of the reason was that there were already differences between men and women in terms of things like willingness to go to the doctor, wearing seatbelts.
And one of the things that I do think we see in this data, you know, not to undercut the very good point Sarah was making that it's not all about COVID, but I think there is some evidence in the 2021 data that, you know, health behaviors like,
wearing masks and getting vaccinated did lead to different outcomes for different populations. I mean, given the staggering health inequalities in this country and the way that the COVID-19 pandemic affected different racial and ethnic groups,
It's striking that life expectancy declined more among white Americans than black and Hispanic Americans. And in a New York Times article, one of the experts, they quote, attributed that to health behavior. He said that it reflects greater efforts by black and Hispanic Americans.
Americans to get vaccinated, wear masks, and take other measures to protect themselves, and the greater tendency in white populations to push back on those behaviors. Obviously, that doesn't explain everything. There are a lot of other factors. But I do wonder if, especially given some of the polarization we've seen around trust in experts, trust in science, trust in institutions, if even something like
medicine and the way that people interact with the healthcare system will be increasingly influenced by partisanship.
Yeah, you know, Amelia, that brings to mind the fact that if you looked at the components of life expectancy and the decline among different racial and ethnic groups, the percentage part that COVID played among not Hispanic whites was 54% of the decline in life expectancy could be attributed to COVID-19. Whereas for
black Americans, it was 35%. For Hispanic Americans, it was about 25%. For American Indian and Alaska Native populations, it was about 21%. And for Asian Americans, it was about 17%. So it's like the non-Hispanic white figure stands out pretty markedly there, you know, that over, that slightly over half of the decline in life expectancy could be attributed to COVID-19. And we know from other data where they've looked at, you know, vaccination rates, looked at
at basically death rates from COVID-19 based on the partisanship of an area and places that Trump did well in tended to have higher death rates from COVID than places where Biden did particularly well in. So that's, you know, I think there is something to that, that you can see from these numbers.
Look at you guys using data. My answer on this is I think Republicans actually can spin this pretty effectively. And what I mean by that is I think you could look and say, look, the death rate was lower under Trump. Things didn't get better under Biden. I think you can argue that's a disingenuous read of the
data, but we are going into a midterm election. Life expectancy did get worse in 2021. I would imagine Republicans kind of latching on to that as a failure of the Biden administration. If this is the engine check or what have you of a nation's health,
I think there is a certain politics that come with the idea of decline. And we've seen that more broadly, and we've seen it not specific to like life expectancy itself, but this adds to the idea that something is not right. Maybe it increases a sense of nostalgia for a period when Americans were more optimistic, overdose deaths were not so prominent, when life expectancy was increasing by significant clips.
And so I think in terms of how
we talk about the country, how politicians talk about the country, I do think that this adds to this, you know, decline politics of which there's lots of history and study because we're not the first country to experience sort of like troubles like this. And I don't know exactly how I think people can see for themselves how to some extent the two parties relate to that. But to me, it's a little bit of a question going forward what that looks like.
I mean, I think the thing about these numbers is that people can read into them what they want. And I don't think, to your point, Sarah, the news coverage did a great job of really untangling how complicated this is and how some of it's about COVID and some of it's about other factors and some of it's about health behavior and some of it's about inequality. And really, it's hard to draw any single lesson from this.
I think people saw the headline of life expectancy continues to decline and everyone's in a bad mood. Everyone thinks the country's in decline. Everyone's unhappy with the way the government handled the pandemic. And it just becomes a thing where people kind of take what they already thought about the government and the administration and everything else wrapped up in the pandemic. And it would be very easy to just project that onto the number and not think any more about it.
All right. So, Sarah, you came down already with general bad use of data. Do we... Is that, Amelia, an agreement that the news media did not do a good job with this data? Yeah, I don't think so. I mean, I read a lot of articles preparing for this, and I was actually really surprised by how thin the coverage was. And I think part of it is, you know, like, this is something that people are going to be studying for a long time, and it is hard to report. I can...
Ken can confirm it is hard to write a story where you say, we don't know what's going on. But I wish there had been a little bit more exploration and a little bit more, you know, honesty about what we do and don't know. Curiosity. Yes. We love curiosity in the news. Jeffrey, are you going to make it a triple bad use of data? Yeah.
I guess I lean that way just because to the point that some of this coverage wasn't terribly nuanced. I think the COVID story is really easy to tell and that the other parts of this are maybe harder to tell and how those different factors interact with one another is even more difficult to sort of suss out. And so trying to lay that out and
1,500 words or what have you in an article is not easy. So bad use, but it's also a challenging topic to be sure. Bad use, but it comes with empathy for the journalists involved, says Jay. Precisely. I appreciate it. No, it's an important story. No, really. It's really important. All right. Let's move on to something a little less important.
Although, you know, at this point, who's to say? Which is the final two months of the midterms. Damn. Oh my gosh, Galen.
How do you feel great on vacation? Like really good? Easy. You go to Aruba. You'll spend your time relaxing on cool white sand beaches and floating in healing blue water. You'll immerse yourself in natural wonder and find your center on an island where things move at your speed. You won't just feel great. You'll feel relaxed, renewed, and ready for life. That's the Aruba effect. Plan your trip at aruba.com.
You're a podcast listener, and this is a podcast ad. Reach great listeners like yourself with podcast advertising from Lipson Ads. Choose from hundreds of top podcasts offering host endorsements, or run a reproduced ad like this one across thousands of shows to reach your target audience with Lipson Ads. Go to LipsonAds.com now. That's L-I-B-S-Y-N-Ads.com.
During the two months between summer's cultural end, and here we are not going to get into a debate about when summer ends because I think we know that there is disagreement on this podcast. Anyway, between the two months between summer's cultural end and election day, most of America's political energy is focused on turning out and swaying voters. Do you
will typically not see all that much in the way of legislation coming out of Washington. The focus is elsewhere. Today, as I mentioned, we are going to play a little trivia game pegged to this particular final stretch, but let's begin with setting the table for what folks can expect during this time. So Jeff, traditionally, how much do polls change in the final two months of a midterm campaign?
So I took a look at 538's historical generic ballot polling average and 62, 63 days until the election. It swings about two points and it's usually it changes in the direction of the party that's not in the White House. That's not always true. So you're talking about just sort of like the absolute change on average is about two points and usually in a midterm it's in the direction of
of the party that's not in the White House. So, you know, that's something that we could sort of expect. There've been a couple exceptions to that in midterms. In 2002, things actually moved in the Republicans' direction, even though George W. Bush was in the White House. Perhaps that had something to do with the fact that Bush was very popular. There was sort of a longstanding effect of after 9-11 and then sort of the lead up to the Iraq war where Bush was very popular. And so maybe the GOP benefited to some extent.
as he got closer to Election Day. And then also in 2018, the polls basically didn't change very much from where they were two months out to where they were in November. And maybe that was down to Democrats already had a really big lead, and they'd sort of maximize that lead to some extent, considering how partisan and polarized things are. Maybe it's hard to get much beyond a nine-point lead in the generic ballot these days.
But also there was in October, for example, the debate over Brett Kavanaugh's nomination. Maybe that amped up Republican enthusiasm some and sort of kept it from declining further for Republicans. Anyway, those are just hypotheses, but those are the couple exceptions, I guess. When there is going to be a blowout, and I guess in 2018 you could kind of say there was a blue wave, there was a bit of a blowout.
But in some of the historical blowouts, like how much change have we seen? You know, you're at like in 2010 or say 1994 or whatever, like how much volatility do we see in those blowout elections? Yeah.
So, yeah, if you look at, say, 2006, 2010, 2014, in each of those, the polls moved about three points in the direction of the party that wasn't in the White House in terms of margin between about this point in November. So you usually have seen some relatively substantial shift there in the favor of the out party.
I would note, though, and we've talked about this a fair amount, what's interesting is at least right now it is Democrats who are gaining in the generic ballot and not Republicans. You know, Editor-in-Chief Nate Silver, of course, has written a ton about how this could be an asterisk election. It's something we are taking very seriously. History would expect, as Jeffrey has outlined, that Republicans start to gain. But so far, it has been Democrats who have closed the gap and now lead in the generic ballot.
Just to lay a little bit more of the groundwork here, I mean, why do we see this movement towards the out party in the final two months? I mean, I think it's down to a few things. Maybe it's an artifact, partially, of pollsters shifting from registered voter polls to likely voter polls because...
They want to get a better read on who's actually going to show up in November. And they figure since you're getting close to the election, uh, more people are paying attention. So you're more likely to get an accurate read of who might actually show up. Um, so you start, you start actually asking voters like how likely are you to vote and then trying to use that information and maybe like past vote history if you're using, uh,
voter list so you know how often these people have voted to sort of get a read on just how likely they are to show up. And we know that historically that shift has tended to benefit Republicans. Republicans are more likely to be older. Older voters are more likely to vote than younger voters. So there's also white voters tend to be more likely to vote.
And Republicans tend to be whiter on average. So these are things that have often helped Republicans. I think there's also just the other part of that, which is that less engaged voters are becoming more engaged and less engaged voters are more likely to be independent, maybe swing a little bit. And
If they're not too happy with the status quo, as is often the case in a president's first midterm and second midterm for that matter, they will vote for the other party. So those are sort of factors that I look at as part of why you're seeing that shift that happens over those last couple of months.
Has there been another election like this one where there has been such a strong cross pressure between the sort of conventional wisdom, normal political forces that we would expect and something like the Supreme Court's ruling in June?
Because that's just what makes this so unpredictable to me. Like the Supreme Court just threw one of the biggest bombs it could have into this election. And I'm trying to think of another one where there were forces that were colliding with each other like this. And I mean, you throw like inflation into this also, but am I missing something? I mean, I think it's 1998.
I was going to say, I think you could argue that 1998 or 2002, because those are the two like, oh, wow, the president's party actually gained seats in the House. There's also an October surprise that I thought I might bring up later, but in 1962, the
Yes. The Cuban Missile Crisis. The Cuban Missile Crisis. Yes, is a very good example of an October surprise because JFK came out of it even more popular because the Soviets had backed down and they had sort of been – the US was viewed as having sort of won this confrontation, if you will. Yeah.
I think we all won because no nuclear missiles were shot off. But anyway, Democrats ended up only losing like four seats, I think it was, in the House. And they actually gained four seats in the Senate. So in a way, that was another example of sort of like the midterm going pretty well for the president's party. So October surprises are not only in presidential elections. They can also affect midterms.
Final question here. In a lot of the competitive Senate races that have been getting plenty of attention, particularly in states where you wouldn't necessarily expect a Democrat to be leading in the polls, I'm thinking Wisconsin, Ohio, to some extent Georgia, Pennsylvania –
You know, with the exception of maybe Pennsylvania in some polls, neither of the candidates, Republican or Democrat, are at 50% or higher, right? There's still a significant portion of undecided voters in the Senate elections. And that means that there's more uncertainty ultimately in the result. I think we generally think if a candidate is above 50% support, it's harder for the other candidate to come back because there's enough decided voters in their direction.
So when do undecided voters generally make up their minds in this kind of environment? Like when does the polling in that sense get a little less uncertain? I mean, I think the easy and perhaps only right answer to this is as you get closer to election day. Yeah. I mean, well, as you get closer to election day, you know, every race is going to differ to some extent by like how well known the candidates are.
how much maybe there's lingering frustration over the primary, but eventually like the party's voters will mostly get in line for their candidate. You know, for example, in Pennsylvania had an extremely, extremely close Republican primary battle. And so maybe there are some Republicans who haven't quite gotten on board with Mehmet Oz, for instance. And he also has like really high name recognition. Yes. And so, you know, that might be a case where it's like just a matter of
Some voters getting close to election day and be like, fine, I'll vote for my party's candidate kind of situation. So I think it's like undecided voters are tricky because sometimes those undecided voters never even actually become voters. You know, you say they're undecided, maybe they don't even show up. Sometimes, you know, again, it comes down to sort of like how well known the candidates are, what other lingering effects there are from like the primary, like how unfavorable attitudes are toward each candidate. Like it's just...
In general, we know that certainty tends to rise as you get closer to election day because you get fewer undecideds. But the number of undecideds is going to vary from race to race. And it's possible some of these might have large enough percentages that raises the uncertainty a fair bit.
I think the thing to remember always about undecided voters is that I think people tend to think of them as like swing voters who are decide, you know, like might vote for the Republican might vote for the Democrat. There's a lot of evidence showing that that is not a good description of undecided voters. Um,
in some ways they may actually be more partisan than, um, people who will just identify, you know, say like I'm voting for one candidate or the other. Um, and their decision might be quite predictable. Um, so I think that's,
you know, that's another thing though, that's sort of complicated about this year is that there are just so many different things going on. You know, there's the economy, there's what people think about Biden, there's abortion, you know, the sort of the pressures that might lead people to stay home, which is another really good point here. I think like people think of, again, of undecided voters deciding between who to vote for when often it's actually like, do I vote or do I just not, do I just not care? And I think that makes,
may make things a little bit more unpredictable too, just because it's not, you know, we don't know what the economy is going to look like in two months. And that could influence the way people are thinking. And I, I mean, I don't have evidence for this. So now I'm just kind of
entering the land of speculation and maybe I should stop talking. But I think like that uncertainty may also be sort of filtering down to voters, especially with the economy and just a sense of like, OK, things are looking better now, but is that going to last? I think that is a would be a reasonable attitude to have.
Yeah, I think that's a great point, Amelia. The whole concept of like undecided voters, you know, this isn't an SNL skit where they're just like so completely like ignorant that they, yeah, or that they're like, you know, swinging in the wind here. A lot of times it's whether or not you're, you know, you tend to vote for one party most of the time, but it's just a question of whether you're going to be engaged enough or animated enough to actually show up and vote. We obviously know midterm elections tend to have lower turnout than presidential elections.
How much so will vary from presidential election to midterm election. I would say that most signs, including sort of the current political environment we're in, the Trump era, if you will, would suggest that we can expect a pretty high midterm election.
turnout this time around will be as high as 2018. Who knows? But a really important part of that is just how much more likely are Republicans to show up than Democrats on average. In 2018, you saw Democrats were more likely to show up on average because, hey, they didn't like the guy in the White House. It was a very pro-Democratic environment. This time around, history would suggest that Republicans will be more likely to show up. The concept of differential turnout here with a Democrat in the White House is
However, I mean, as we've seen with the Jarek ballot still running even, you know, it's it sort of remains to be seen just how much that that will show itself.
Yeah, no, Gallup was looking at this question. And unfortunately, they did not ask it at the same point in 2018. But they thought that based on where it was currently, which is that 48% saying that, you know, they're really expecting to vote, very engaged, would have been on par with where it was in 2018. Because by the end of the 2018 cycle, 54% said that. So that's like a six point difference.
And frustratingly, too, this poll was done before the Supreme Court decision. So Republicans at the time had like a 10 point enthusiasm gap. But we have seen that and other polls shrink. I don't think we've seen Democrats like on average surpass Republicans, but it's been like neck and neck, which reflects the generic ballot. All right. Well, that is a good table setting exercise. But let's play a little trivia.
You're a podcast listener, and this is a podcast ad. Reach great listeners like yourself with podcast advertising from Lipson Ads. Choose from hundreds of top podcasts offering host endorsements, or run a reproduced ad like this one across thousands of shows to reach your target audience with Lipson Ads. Go to LipsonAds.com now. That's L-I-B-S-Y-N-Ads.com.
We're going to try to jog our memories a bit with a game of trivia focused on historical midterms. So I am going to ask the question and give the three of you a moment to come up with an answer and then ask you all to say the answer at the same time. Because in the past, I think order has benefited some folks. There's a lot of triangulation. We just want it to be a sort of
raw, like, first number or answer that comes to mind. You give it to us, and then we will sort it all out. So, are you all ready for this? How long do we get to think about it? Or is it just really, like, whatever? You know, I'll give you a little bit of time, and then I'll say, like, three, two, one, go. Okay. Okay.
So here it is. First question. What is the average number of U.S. House seats lost by the president's party in a midterm election since 1946? It's 1946. It's only going to get harder from here, probably. So I'm not going to give you too long. Here we go. On the count of three. Three, two, one, go. Twenty-six. Twenty-five. Twenty-seven. Twenty-seven.
Ooh. Whoa. Wait, I feel like there was a little delay there. There was some crowding in the answers. Sarah, you definitely answered first. No.
No, mine was 27. Okay, so. I was startled because I didn't think Sarah was going to say something so close to what I said. Look, I read Jeff Skelly's pieces for the site. What can I say? I think it's 26. I also feel bad because I feel like I should have known this right off the top of my head, and I was like, I think it's somewhere around 25, and I don't remember exactly. Is it 25? I don't remember. Yeah, it's 25, guys. Damn, Jeff Skelly. Damn it. All right.
I knew it was close. Very fitting. Very fitting. That was like impressive. Damn. Okay, so one point for Jeff. I think from here on out, these won't have been answered in previous pieces that Jeff has written, but... Written by Jeff? Um...
We'll see. I would like that as a requirement, perhaps for future questions. There's going to be an asterisk on that one, isn't there? Although, to be fair, if it's written by Jeff, it's probably been edited by Sarah and read by us. So, you know. Next question. What percentage of Americans said they had given quite a lot of thought to the upcoming congressional midterm elections in the summer of 2002? Jeff, if you actually know the exact number here, I wouldn't.
I will be wildly impressed. I have to imagine that you don't. I do not. And if I get it exactly right, it is a complete and utter fluke. Quite a lot of thought. It's quite a lot. Yes. All right. Three, two, one, go. 35%. Wait, did we get two 32%s?
Yeah, they both said 30. No, I said 35%. I said 32. Sarah is... I said 32. Whoa. So this makes me think it was in a story that you two worked on together. Not in the midterm of 2002, I don't think. I don't think. So we're going to have to... Did we nail it? No. All of you are wrong. Damn. Okay, well, that's reassured. Okay, okay.
but in a weird, you're wrong in a weird way. But I also think I'm just, I'm not accusing anyone of anything, but I'm just going to say for the next ones, you're going to have to write your answer down and then you're going to have to hold it up and say, I mean, listeners won't be able to see. I'm in favor of this. Listeners won't be able to see. So you'll have to still say it out loud. Um,
In the meantime. Oh, but you're just going to make us have a record of it? Yes, yes. We're not going to like hold it up? You can hold it up, but we'll also have to read it out loud.
So the answer to quite a lot of thought given to the congressional midterm elections in the summer of 2002 was 23% of Americans. Damn. The reverse of Sarah and Jeff were using, which was 32%. As a point of comparison, this summer, 48% of Americans surveyed by Gallup said they had given quite a lot of thought to the upcoming midterms.
Even after September 11th, it's more than double the number of Americans who are paying attention. Sarah and Jeff, you were the closest ultimately, so you each get a point. Jeffrey, two. Sarah, one. Amelia, zero. Oh, yeah.
So now you all have your paper and pens. We're going to move on. That's right. Will you do the Jeopardy song as we write? Number three. Thank you. Thank you, Trebek. According to the Wesleyan Media Project, what percentage of 2018 pro-democratic U.S. House race ads mentioned the issue of health care? And they started recording this from right after Labor Day.
Okay, that's enough of that.
Here we go. Three, two, one, go. 82%. 82%. 46. 46. 73%. Okay. And I was channeling my inner Sean Connery and just put your mom because celebrity jeopardy. Wait, hold your numbers up again because I'm going to have to do some quick math here. The correct answer is 59%. Oh.
Oh, damn right. Which means that Sarah won. Can you do the math for me, Jeffrey? Oh, yeah. Wait, wait, wait. Yeah, no, Sarah did. What was it? No, wait. What was Sarah's? 73. And it was 52. It was 59%.
Yeah, Jeff gets it. Oh my God, did I really not get that? What were the other ads even about? 13 points and 14 points. You guys are really running really close to each other. What were the other ads about? I don't know, Trump probably? Yeah, I guess that was a thing. All right. But it was like Trump and healthcare. Also, probably a lot of biographical ads. Like, hi, I'm so-and-so and I was in...
The military when this plane went down. Remember like the Doors ad? Although that was before Labor Day. I got to get one. I think I'm adjusting my what success means in this game. So the score is three for Jeffrey, one for Sarah, zero for Amelia. We are moving on.
It's interesting what a spread you guys had there the first time I asked you to write it down on paper instead of just saying it out loud. I would say that was by far the hardest question you had asked. Fair enough. I'll choose not to read anything into that. Okay.
Number four. Didn't come directly out of a story written by Jeff. Number four. Name one of the top two concerns for Republican voters in the 1986 midterm election, according to ABC News exit polling. 86. I love this question. Thank you, Emily Vanesky, our intern, for tracking down all of this fabulous trivia for us. You guys ready?
Three, two, one, go. The economy. Taxes. Russia. Like Soviet Union, the wall. Tear down your wall. Okay, so Jeff is the economy. Amelia is taxes. And Sarah is Russia. We will be awarding no points for this question. The answers, perhaps...
weirdly similar to today, although not weirdly, I don't know, politics maybe doesn't change that much, were combating illegal drug trafficking and party loyalty. Party loyalty? Those were the top two issues? Interesting. That's curious. What is party, how is party loyalty even an issue? I don't know, but I will. I object to the premise of this question. That's really fascinating. Actually, what's interesting, of course, is that right after
that midterm, Iran-Contra broke. It was like days after that midterm. So it's just funny to think about the timing of that.
But drugs, that's a good one in 86. So here's here. This is from a Washington Post write up of the exit polling. Democrats nationally in most of the key states indicated greatest concern with reducing unemployment and protecting Social Security. For Republicans, the most important issues appear to be combating illegal drug traffic and party loyalty, according to the ABC News poll. So there you go.
It's also about which items you include in that poll for people to say, unless it was open-ended. But my suspicion is it was, they had a master list that people picked from. All right. So it is still 3-1-2-0. Next question. According to a Pew survey from November 2006, what percentage of Americans said they got most of their news about the midterm elections from the internet? 2006. Okay.
Which is like a fun year to ask this question because it's kind of cuspy. Yeah. I'm not sure I know the answer this year though, which is... That's a good point. It's not, I like don't have a peg for it, you know? So I could be very wrong on this one too. Yeah, this one's gonna be tough. Is there like a heart style version of this where I can win by not getting any right? Yeah. But that heart style version is just winning in your heart. Okay.
Oh, wow. Wow. No participation trophies on the 538 Politics podcast. I'm a millennial. And I hate to lose. Okay. Three, two, one, go. 28%. 27%. I said 21. Okay. You are pretty close. It's 15% of Americans got... Damn it. So really... Damn. Damn.
You were all in the right direction, but Jeff, you got the point again. Yeah. Damn. Sorry, folks. Splitting hairs over there. Close does count, and thanks beyond for shoes and hand grenades. In 538 Trivia. Yes. Okay. In 1994, a landslide midterm year for Republicans, what percentage of registered voters said in a Gallup poll right after Labor Day that they preferred the Republican Party on the generic congressional ballot? Okay. Okay.
Three, two, one, reveal. 60%. Damn it. 46%. I said 44. And with that, Amelia is on the board. It was 49% of registered voters said that they'd vote for Republicans on Gallup's early September 1994 generic congressional ballot poll. 45% preferred Democrats.
Okay. I was thinking that it was like close, but then things got worse for Democrats as you got to November or something like that. If you listened to the introduction of this segment, you would come away with exactly that conclusion, Jeff, that things would look close and then things got better for the Republicans. I see. I see.
Sarah Frost and sin. I clearly tuned you guys out. Said no thanks. Sorry, I'll tone it down. Potential bonus point for anyone who can...
name the number of seats that Republicans won in 1994? Oh, this is for a point? This isn't just for fun? You know, we'll decide. Only if you get it, only if someone gets it exactly right will points be awarded. Okay, okay. Because it is a bonus question. Okay, here we go. Three, two, one, reveal. 58. 58.
44. Was it the total number that they won or the number that was changed? Oh, that was in the, okay. Because I would answer 63 for the shift. Okay, so I was talking about the shift. I was talking about the shift. Wait, hold on. Yes! So you were saying 63, Jeffrey. Sarah, you said what? 58. Oh my God, I'm thinking of 2010. Sarah, you got it, but you don't get any points because it's not exact. It was 54 seats.
What? This had to be exact? Yeah, yeah, yeah. Them's the rules, Sarah. Sorry. Oh, come on. No, bro. Like, there was no exact specification there. He said it. He said it.
He said exact. Okay, well, that's bullshit. I had the 2010 number in my head. So that was really dumb. I just can't. I just can't track. Wow, that was dumb. And after winning those 54 seats, the GOP became the majority party in the House for the first time since 1952. More historical facts for us. Okay, so final question. We're gonna wrap it up on this.
There's no coming back from this. There is no coming back from this. It is worth five points, so maybe you can... No, I'm kidding. Whoa, whoa. Any other technicalities? We can't do a Micah Cohen style, like Sarah and I add up our points and also double them type of thing? Potentially. Let's see how this final question goes. Okay. All right. So this is worth five points. This is worth however much Galen wants. It could be worth negative points. It's kind of a surprise.
It'll be a surprise. Interesting. Which midterm year had the largest landslide with the most seats flipped by one party? In all of American history? In all of American history. Oh, shit. Ooh. Ooh, I like this. Yeah, I bet you do. It's going to be something in the 19th century. Come on, Sarah. This is where Jeff's knowledge of...
19th century political history is going to give him the win. I guess that's true. You're saying all of American history? All of American history. All of American history. Since George Washington. Okay. I know it's one of, I think it's one of two and I don't remember which one it is. Well, you're going to have to just pick one and go for it.
To be frank, once we get to the 19th, like, I have some sympathy for y'all because once we get to the 19th century, I wouldn't be able to tell you what years were midterm years versus presidential years. So quite frankly, if you even just get a midterm year. Anyway, okay. Three, are you guys ready? Three, two, one. Reveal.
1898. I was listening to the podcast earlier. 1932. Okay, we have 1894, 1898, and 1932. It is 1894. Damn it. Whoa! Let's go. Wait, Jeff actually knew that? Let's go! I knew it.
I'm more proud of that one than anything else that has occurred. Yeah, that's good. That's good. What happened in 1894? Huge, well, panic of 1893, man. It was bad times, bad times. Huge Republican gains. What was the other? When Grover Cleveland was in his second term, Democrats lost 116 seats in the House and five in the Senate, and there were only 44 states in the Union at that time.
So as a percentage of all seats. People were mad. They were mad. What other midterm were you thinking, Jeff? It was like 1890 was really the other one because Republicans had lost a ton of seats before.
You had some wild midterm swings in those years. Benjamin Harrison won in 1888, but it was not very popular. Which is part of how Cleveland got back in in 1892. You forgot to mention two non-consecutive terms for Grover Cleveland there, which is a historical data point we will be referencing a whole lot, I bet, in the 2024 cycle, depending on...
how a certain someone's comeback attempt goes. Yes, yes. All right. So Jeffrey, unfortunately that question was worth negative four points. So the score is zero to one to one. Sarah and Amelia, you have tied. You will be getting the award cut in half and sent to you in the mail.
I joke. You are all winners, like I said, in your hearts. And thank you for playing along because the real winner here is the listener who has just learned more midterm political trivia thanks to all of you. So thank you, Sarah, Amiria, and Jeff for providing that service to our listeners.
Yeah. Well done, Jeff. Well done. Yeah. I, Jeff was the real winner. I'm a sore loser, but that question, he, he deserved credit for that. And we're going to leave it there. My name is Galen Druk. Sophia Leibovitz is in the control room. Chadwick Matlin is our editorial director and Emily Vanesky is our intern. You can get in touch by emailing us at podcasts at five 38.com. You can also of course, tweet at us with any questions or comments.
If you're a fan of the show, leave us a rating or review in the Apple Podcast Store or tell someone about us. Thanks for listening and we will see you soon.
