8: What South Carolina and the Early Vote Tell Us About the Midterms

I had a band in college called Boom Boom, where. No! I'm playing. Yes! I think he has depths that we're only just now getting to. We're only opening the doors of the inner sanctum. That's what this podcast is about. Getting to the inner sanctum. Welcome to Still Counting. I'm Galen Drake. I'm Claire Malone. I'm Nate Silver. On today's show, it's wall-to-wall midterms coverage, including the latest in Nate's. Shocking. I know, right? Including the latest in Nate's flipper election forecast. People have been pointing out that it's sort of getting close to a horse, which I think might be true. Yeah, it is getting close to a horse. We'll look at flipper and the battle for control of the Senate, including a new New York Times-Sieta poll. We're also going to discuss whether or not pollsters are hurting, look at the early votes, and more. And we're going to be sprinkling listener questions in throughout the episode. This is a service journalism episode. But before we get to flipper updates, we have some housekeeping. Still Counting will be dropping a special pre-election bonus episode next Monday, as well as our regular episode on Wednesday, which obviously will take in whatever has happened on Tuesday, which, if you are living under a rock, is Election Day. Guys, what are your plans? What are you going to do on Election Day? Anxiety, Max. Look, there's not much to do on Election Day. Election Day. Do you vote in person on Election Day? Do you do it beforehand? So, okay, full disclosure. My polling place is still at my previous address, which is next to my. Voter fraud. So I'm going to my trainer today, and I will vote. Is it the same congressional district? Otherwise, you're admitting to voter fraud on the Still Counting podcast. For years, I voted in Ohio when I did not live there. This is fraud, guys. You can't just be saying this. This is like when Bill Simmons admitted that he had a proxy better for who. I lived there part of. I lived there for part of the year. Oh, boy. I don't know about this. Okay, Nate, I think that Claire is teeing us up to talk about what we're actually doing on Election Night, which is that we've got a live stream going over on Substack. We're teaming up with The Argument and Matt Iglesias. You and I are going to be there, and Claire is stopping by. Yes, I'll also be on the Crooked live stream for a little bit. But, yes, Nate and Galen and all the people at The Argument are going to be doing some good stuff. Yeah, there'll be a live stream. And or a live blog. Some of that will be free. Some of that will be privilege for paying subscribers of our respective newsletters. And I'll also say, over at GD, we just have, like, a ton of stuff going. We've got a pollster off tomorrow between Republican Patrick Ruffini, who's been more bullish on Republicans, and Laksha Jain, who's been more bullish on Democrats. We've got, like, an hour-by-hour guide coming up, some final looks at the polls, and a late-night podcast on Election Night. Basically, like, election Twitter come to life is what the next six days will be. And we also want to remind you guys that CrookedCon is just over a week away. Crooked published the official schedule at CrookedCon.com. It has over 25 sessions on November 7th at the Washington, D.C., Convention Center. For example, our live show at 9 a.m. on Stage 2. This is going to be a first for us, a live show at 9 a.m. I'm a morning person. I was up at 6.45 today. Nate, you have become a morning person. You were not a morning person. We're all morning people now. What time did you wake up today, Galen? A little before 6. I woke up at 4.45. Oh, my God. Okay. Let's talk about the midterms. Get your tickets at CrookedCon.com. Okay. We are going to check in now with the latest from Flipper, where Democrats' chances of controlling Congress are holding steady as of Tuesday. The model gives Democrats about a 75% chance of taking the Senate, which is a notch down from last week, 77%. And their chance of winning the House is hovering around 95%. Okay. This is our second-to-last chance to talk about this before Election Day. So, Nate. Penultimate. Penultimate. Total layup for penultimate, and you didn't go there. I'm sorry I didn't line by with you. I like using those round numbers. They've been stuck or consistent at 75. And 95, respectively, if you round to the nearest kind of whole number. I will say, a bunch of new polls that were interesting this morning that we'll talk about, particularly in South Carolina and Minnesota, have not entered Flipper yet. And I'm curious what that will do. You know, my sense is that, like, despite some weirdness and some hurting that we'll talk about, this latest round of polling, we finally are seeing a heavier volume of polling, has been pretty good for Democrats. I wouldn't say any. I wouldn't say any sign of, like, a last-minute Republican rebound nationally. You can debate some cases like Texas. I would also not say anything too crazy. Okay. But, yeah. But you want to talk about some states? Or what do you want to do? Well, yeah. Well, first, let's start off with this. A Washington Sun reporter tweeted on Monday about the Flipper forecast, and I'm quoting here, Well, Democrats are kind of, I was going to use a term that was, I was going to use a term that was inappropriate for our podcast, right? Oh, wow. Like, I don't fucking give a fuck. Does it start with a P? My job is to actually make a forecast, right? Democrats' job is to panic and or spin. And probably both panic and or spinning would lead you to take a more moderate view of the race. I mean, look, I'm going to say, I'm trying to figure out how not to get myself in too much trouble, right? People have been talking about this obvious case that, well, you have all these Democratic ties or leads in red. And they come up with a seat or two short, right? Very possible. That's what the 25% chance represents, more or less, right? There's also a really obvious case that people are too fucking scared to make that it would be a landslide. Because, look. Cue Fleetwood Mac. And the Dixie Tracks. Gas costs $4.30 a gallon. Trump has a 37% approval rating. Trump lost by eight and a half points in the U.S. House midterm in 2018. Trump is 80 years old. Trump, I'm saying Trump a lot because he's the most important factor in the election. Trump is at war with Iran. Yep. Trump is out there campaigning and talking about men's butts when Republicans don't want to talk about him, right? The number of campaign rallies. Hey, he talked about Nicki Minaj's butt, too, this weekend. Let's just be comprehensive here. So, I guess, you know, Trump is an ass man. Bisexual king. Bisexual president. And I believe that we will talk about early voting, which is kind of like I'm reluctant to do that, right? But, like, certainly among some groups, Democrats seem to be turning out low-prensity voters. But, look. People are too afraid to, like, take this upside case seriously. And, yeah. If it happens, there'll be plenty of evidence that, like, this is a really, really bad environment for Republicans and Trump. Can I just say here, like, I think we are getting at two full freakout over polls and everyone's tuning in and everyone has their own sort of suspicions and doubts and whatever. And at a moment like this, I think it is helpful to, like, put all of the horse race polling to the side. Like, you can still look at other data, but just put the horse race polling to the side. Donald Trump's net approval rating is negative 23. That puts him on par with George W. Bush during his second midterms or worse or around. Or where Nixon was or worse. That's historically bad. When Gallup asks Americans, which party do you think would do best addressing the issue that you think is most important plaguing the country? It is Democrats by a healthy margin. There's just even if you're not pitting, you know, Josh Turek against Ashley Hinson or whatever, there's a lot of data. And we've all we've talked about gas prices ad nauseum on this podcast. There's a lot of data that backs up the story that the horse race then suggests. We're seeing. Yes, very much so. OK, before we even get to any listener questions, I just want to I want to talk about some polling that came out today. But specifically, let's dive into a race that Ted Cruz recently said was neck and neck. And the Times poll, which came out this morning, seems to agree. And that's South Carolina. Trump has obviously visited there recently in the Times poll. Annie Andrews and Lindsey Graham's sister, Darlene Graham, were basically tied. Annie Andrews recently got some key endorsements from a former Republican and former Democratic governor of the state. You guys, how did this race become kind of the last minute surge? Do you mean Dr. Annie Andrews? I do mean Dr. Annie Andrews, a pediatrician. Thank you. Thank you, Nate. Why is it important that she's a doctor? Is this does this play in at all? Is it just is this candidate quality? Is this I've been I've been thinking about titles. Yeah. Yeah. I've been thinking about this race. I've been thinking about this race all morning. So in some of the other states, such as Ohio, people like, well, Ohio is a really red state. Now, if you actually look at this national environment, we've been talking about and shifted over 11 points from 2024, then Ohio should be a toss up. And so if Sherrod Brown's a better candidate, then that's not that abnormal. Right. You really would need a big Democratic overperformance in South Carolina. There are a couple of ways to do that. One is based on candidate quality. I don't think. Andrews is any type of electoral superstar. She has kind of fairly typical, progressive, moderate mix of positions, right? She tries to brand herself as a pediatrician, which she is. She has lots of photos of her with, like, sheriff. Trying real hard. Trying real hard. Well, okay. But, like, I think it's more about, like, Darlene Graham's weaknesses. I agree with that. I mean, Darlene Graham has had some sort of embarrassing debate moments. You know, the other thing I have actually been thinking about, I'm writing a book about Maha. I think a lot about Maha. South Carolina, the Spartanburg area, was the center of a measles outbreak. And I think one of Andrews' first ads, back when Lindsey Graham was still alive and running. Right. She brought up measles. And I think she's actually, I believe in the write-up of the Times poll, they noted that she was doing, like, a bit better in the Spartanburg-Greenville area, which is sort of the area where it was all happening. I don't know if that's coincidence. I don't know. But I was like, oh, that's an interesting, it's interesting at this moment to have a Democratic pediatrician doing well in a state that's had that. Just to note the little kind of, like, valence there. That's also an area where Darlene Graham's primary opponent outran her as well. So I think Darlene Graham may also not be a particularly good fit as a candidate there. I have also been thinking about South Carolina a lot this morning and digging through the crosstabs of that New York Times poll. It's worth saying here. Trump won South Carolina by 18 points in 2024. I think Nate was getting at this, but it's not a state that under any normal circumstance is close. Remember all the excitement around Jamie Harrison and then he lost by 10 points. But looking at this poll, here's what caught my eye. Trump's approval rating in South Carolina right now is net negative 13. That's really fucking bad. Going into the 2018 midterms, I looked this up. You know, there was a bunch of different polling, but one that I pulled from Winthrop. Trump was 5. 5 points underwater. So he's significantly more underwater now in South Carolina than he was back in 2018. The other thing I noticed is that Republicans have a healthy lead in the generic ballot polling in South Carolina. So if you just ask South Carolinians, would you rather have a Republican or Democrat win for Congress? By 8 points, they choose the Republican. But then you pit Annie Andrews and Darlene Graham against each other, and it's 48-48, exactly tied. I mean. And where is a lot of that movement coming from? It's coming from. I mean, we are seeing like pretty healthy turnout amongst black voters across the South so far in the early vote, which we'll get into. But you're seeing a pretty big shift amongst white college educated voters. These are the swingier, you know, it's interesting. White college educated voters in the South are much more Republican than white college educated voters in the North because they tend to be more religious. And that makes them less elastic, less swingy. But you are seeing that. Though on the generic ballot, they prefer Republicans by 15 points. They're only supporting Darlene Graham by three points when you look at that head to head. That's a lot of numbers. But what I'm trying to say here is this is a unique situation. This is unexpected. And if this turns out to be either a neck and neck race in the end or Annie Andrews even winning, it's a big fucking deal. Well, it's a unique situation or we're fitting a convenient story to polls that are too hot for for Democrats. But but look, you want. Hard data. Darlene Graham ran when her primary over Ralph Norman by only five points, despite Trump's endorsement. So that's a story of dissatisfaction. And yeah, you have had across the South extremely high and many of these states are required to track voter registration by race, by self-identified race. And like from Louisiana to Georgia and North Carolina, these seem to be unambiguously the strongest. Strongest numbers for Democrats in the early vote, which we'll talk about more in a minute, I suppose. I kind of ran some. I ran a little math and I asked what if what if black voters turned out at a presidential rate in South Carolina and non-black voters turn out at regular midterm rates? Right. And it's mostly black and white state. Not a lot of Hispanic voters there. If you do that, then that produces about a nine point swing based on turnout alone. It's not going to be that big a differential like that would be kind of unprecedented. Right. You need three things. Right. You need a candidate quality advantage. You need high black turnout and you need depressed white turnout. Right. I can see all three happening. It's a parlay, though. My barometer for where people's heads are is always my college group chat. Shout out. One of whom submitted this South Carolina question. They also have thoughts. About your hats, Nate, and are intrigued by your choice every week. But I want to I want to skip forward since we're. No, I mean, in a good way, but just like curious about, you know, why the mood strikes you in certain silver was feeling jealous of all the attention Flipper was getting. Right. Silver is the successor to 5E Fox and he feels like he should be more of a star. And I think I know what I'm getting getting you for Christmas hats. This episode is brought to you by the International Rescue Committee. You've seen the headlines, wars, floods and crises are forcing millions of families to flee their homes with nothing but what they can carry on their backs. Tonight, many of them are living in makeshift shelters in the cold without enough food, without clean water, without the medicine their children need to survive. You can help. The International Rescue Committee is on the ground in more than 40 countries, providing families with what they need to survive food, clean water, medicine, safety. And right now, for a limited time, every gift you make will be matched. So your donation makes an even bigger impact for the people who need it most. Families who have lost everything. This is a moment to act. Please give now at rescue.org slash rebuild. That's rescue.org slash rebuild. Still Counting is brought to you by Made In. If you're considering the pros and cons of different cookware brands, you should know that Made In has more of the pros. Pros like Tom Colicchio and Brooke Williamson and many other professional chefs who all trust their cookware. Pros like Tom Colicchio and Brooke Williamson and Brooke Williamson and Brooke Williamson and Brooke Williamson and Brooke Williamson and Brooke Williamson. Pros like Tom Colicchio and Brooke Williamson and Brooke Williamson and Brooke Williamson and Brooke Williamson. to Made In cookware. Fact is, Made In has a longstanding relationship with professional Pros like Tom Colicchio and Brooke Williamson and Brooke Williamson and Brooke Williamson. chefs. The company evolved from a 100-year-old kitchen supply business and works with multi-generational craftsmen to make each piece. They make exactly what demanding chefs are looking for, including a wide selection of curated products from carbon steel to stainless clad, plus plateware, glassware, and more. The three of us recently had the opportunity to select a piece of made-in cookware each. Does it say something different about each of us, what we selected and or our significant others but our household went with a 12 inch uh stainless clad frying pan i can imagine getting one of those nice like uh japanese steaks a little little salt do you know the trick nate you got to turn off the heat really high before you put anything in it and put a little i'll get a little get a little splash of water make sure it whatever kobe beef whatever then you can cook without it sticking on your stainless steel absolutely but perhaps Pros like Tom Colicchio and Brooke Williamson. Pros like Tom Colicchio and Brooke Williamson. And yeah, you get kind of close, but Nate, I'd be curious what you think about the rest of the electorate there. So, John, this is one of those Creole names, John Cuvion. I probably in French find that too much, but he's a data observer, a longtime observer of Louisiana politics, who is very versed in the early voting data there, right? And there, the early voting turnout among black voters is so strong that Cuvion is saying, I'm pretty sure Republicans will win in Louisiana, where there's been very little polling. But I'm not totally sure, and I think it's closer than typical, right? And so the argument is, you know, look, Flipper is very skeptical about Louisiana and Mississippi. We're actually lower than prediction markets there. The argument would be, though, that that would provide proof of concept for a slightly more purple state in the form. Okay, here's where I want to bring in our first listener question from Andrew, and he asks, in your view, which races are not getting polled enough this cycle? Does anything stand out for you guys, you know, particularly in this environment where it feels like, oh, all of a sudden, we've got a lot of stuff on the map? Wait, I want to see if Nate and I picked the same state. Minnesota, I mean, the New York Times designed to remedy this, right? High-quality polls of South Carolina and Minnesota. I mean, there has been other polling of South Carolina. Actually, the Times poll is not an outlier, but, like, that Minnesota poll is interesting. Which is Peggy Flanagan plus 14. 14, which is a partisan benchmark. If you have a plus-nine environment in a plus-five state, it should be 14. So maybe all our contemplation about Minnesota a few weeks ago was also speculation based on incorrect polling. Yeah, what states would you pick, David? Okay, why? Kansas. There have only been two public polls. They're from the first half of October. Both of them are from Insider Advantage. And one shows Adam Hamilton up one point, and the other one shows Roger Marshall up one point. And so I don't think we have a really strong sense. Do you have a gut based on other states? All the people I've been talking to seem really bullish for Democrats in Kansas, and Republicans seem— I mean, I was talking to Jessica Taylor at Cook Political Report earlier this week on GD, and she was saying, you know, she sees inside polling, which, again, we don't always put that much stock in, and she talks to—she was talking to Republicans who are so pessimistic about Kansas that they could see a world in which they lose Kansas even if they're winning, you know, Michigan or Maine. I want to—can I say—and the other one I'd say is Nebraska. There are no public polls from October. No, in Nebraska, we're calibrating based on party internal polls, which both suggest a close race, right? It's really been remarkable how, like, even the Republican internal polls are like, oh, yeah, we're up ahead three points in this really red state. Like, hmm. That's not great, right? Yeah. But, like, I do want to—I don't want to abandon South Carolina because, again, it's unlikely to be the tipping point state. Probably if Democrats win it, they're having a really good night, right? But it does provide, like, what I'd call, like, Susan Collins insurance, right? Like, let's say Democrats do really well among Black voters. Well, what races does that help in? North Carolina and Georgia. In fact, part of the reason those races aren't even on the radar is, I think, because of the Black vote, right? Texas, a fair bit. Ohio, a little bit. And then maybe South Carolina. What if they're—what if this white—these white states are fake in the polls, right? But the Black turnout surge is real. And what if Susan Collins wins? You could actually—if you pick up Ohio, Texas, North Carolina, South Carolina. Weird. Weird. But, like, that's four states. Yeah. Which also brings— Which also brings us back to what you said about Minnesota. The New York Times published a poll showing, again, Peggy Fine got up 14 points in Minnesota. The reason that's so interesting to me is that there is no party registration in Minnesota. And so pollsters can't use a lot of the tools that they have been using to try to get closer to the expected baseline of the electorate in a state like that. It's closer to what, for example, Ann Seltzer was doing in Iowa back in 2024 because there are just fewer things. You have fewer— Less ability to wait in that race. That suggests that one of two things is happening. That either all of this fancy schmancy stuff—whatever, I'm being a little flip—that pollsters have taken on during the Trump era to try to reach Trump voters is leading them in the wrong direction this time and actually is under-sampling some of the shifts that we're seeing. Yeah. Okay. Potentially possible. Or that poll is experiencing exactly what Ann Seltzer experienced. And we should sort of put that to the side. Nate, you were talking a little bit about, you know, what if we see a huge Black voter surge? Listener Ryan asks, what framework should we use to determine if results are a blue wave? Like, you know, turnout is—you know, increased turnout of certain demographics is one. How else are you thinking about that theoretical outcome? Well, it's—unless polls are way off, it's a blue wave. The question is, what's the magnitude of that? I mean, the whole $64,000 question has been, like, this rock in the heart. It's a place where it's likely to be a very blue environment. But how does that match up against these R plus 10 or R plus 11, or in this case, South Carolina, R plus 18 states, right? But, like, look, I mean, look, if Democrats pick up the Senate and win Texas, no one's going to have a contrarian take that it wasn't a blue wave, right? If they pick up the Senate and win Kansas, if they win the Senate, it's a blue wave. Let me ask an even more sort of particulate question about the actual— Night of election night and then the day or two after. And I—you know, if this had been 10 years ago, it would have been like, oh, this is just for election nerds. But I think we all know that the way results are reported, the way the television narrative goes, as people metabolize the results, often can affect the way races are written up or the way pundits are tweeting about them on Twitter. So, like, can you say, you know, let's theorize, OK, it's like 9 o'clock next Tuesday. But what states should be—should people be watching to have an indication of, OK, wow, this might be a blue wave? Or are we really sort of saying, you know what, like, you're not going to know what's happening, what's real until, like, maybe Thursday, maybe later? Like, do you guys have a sort of the smart—the smart person's guide to watching this election without being irresponsible and ill-informed? I want Nate to zoom in on specific races here if he's watching them. But I just want to shout out that at— 7 o'clock, polls close in Virginia, Georgia, and Florida. Florida also counts pretty quickly. And at 7.30, polls close in North Carolina and Ohio. If things are going the way we expect them to go, I think we should be able to say who won the House by 8 o'clock and have some pretty—I mean, because of the—and look, Latino voters in South Florida and Florida in general have proven to vote differently from Latino voters across different parts of the country, namely Texas, Arizona, and the Northeast. I don't know that Florida is—like, if Republicans are collapsing with Latino voters in Florida, then I would trust that they're also collapsing with Latino voters in Texas. But if they're holding their own with Latino voters in Florida, I would say we don't know enough yet. Like, maybe they're still losing ground with Latino voters in the Rio Grande Valley, et cetera. Latinos in South Florida and Latinos in South Texas are two of the most unique groups, and I think those are not things that I would take to extrapolate. Even one to the other, necessarily. Look, there's a Kentucky district. The Lexington district is competitive, even though it's fairly Trumpy, right? Like, that would be a bad sign if the GOP loses it. North Carolina, I think Roy Cooper would expect to have that called fairly soon. If it's not, that's a bad night for Democrats, right? If Florida isn't called fairly soon for Ashley Moody, maybe a bad night for Republicans. But, like, people—I don't know. You know what I do on election nights? I just click F5 and reply. Refresh the needle. And polymark it. That's all I do. And then pretend to be smart. That's all I do. Great. Pay for his livestream, everyone. That's what he's doing. Okay, I do more than that. Gallo's humor to whomever he's losing. Let me get another couple of reader questions in. This one is from Sean, and he asks—or she. I've known it's the correct way, S-E-A-N. There's a correct way to spell Sean, and it's not. You know that I feel like those are totally different names? I have this intuition. I went up to a guy. I was at a little poker charity event, right? And he's like, my name is Eric. And I'm like, Eric with a K? I just fucking knew. It's my savant-like skill. Sean, spelled correctly, says, Is there any reason to suspect that there might be a coattails effect in states where gubernatorial and Senate elections are taking place? For example, in Iowa, Rob Sand is polling better than Josh Turek, or the reverse, in Georgia, with Jon Ossoff and Keisha Lance Bottoms. Thoughts? Can I give you a technical answer to that? Please do. I'm curious about this. There's not a coattails effect to the extent that isn't already priced into the polls. Okay, so the technical way to do this is say, take the Senate polling average, take the gubernatorial polling average. Me and Eli sort of built into an article on this just last week, right? Just the fact that Rob Sand is plus six or whatever he is, and the governor's race mean that Turek will outperform his plus one. And the answer is basically no. That there's not additional information provided, but you know. But if Sand's a popular candidate, then maybe that's already filtered into and makes Turek easier to brand himself as kind of nonpartisan, right? And if Sand has good turnout operations, maybe it helps a little bit. But like, but, you know, we're forecasting relatively narrow governor versus Senate splits with some exceptions. We probably don't have time to talk about Vermont. Vermont's a really interesting race, for example. You know, I think we should put a little respect on the name of split ticket voting, right? Like a lot of people imagine, oh, you know, John Ossoff is going to win Georgia by 9, 10 points. Of course, he's going to, you know, pull Keisha Lance Bottoms over the finish line if you want to frame it that way. And yes, I think there are coattails effects. I think that and sometimes, you know, two different candidates can help each other out. Maybe Keisha Lance Bottoms helps black turnout and maybe John Ossoff helps with maybe sort of like swingier, affluent white suburban voters in, you know, around Atlanta, whatever. Well, we can dig into. All of that after the fact, but we're going to see, especially in a midterm environment where we don't pay like 95 percent of our attention to the presidential race differences between candidates. There's going to be a huge split ticket in Maine. There's going to be a significant one in Michigan. There's probably going to be one in Texas. We are going to see one in Iowa and Georgia, and we're actually going to see a big one in Florida. And so, you know, there is that sort of 10 to 15 percent of the electorate that does. Tend to show up to vote more regularly, but changes their mind, they will, especially in an environment and midterm environment like this, sometimes be voting for people of different parties for governor and Senate. Nate, you just published a whole thing on this where you looked into whether split ticket voting was declining. Were you going to say something about that? Because I'm curious about that. No, it is declining, but the degree of decline is kind of consistent with with the polls. I mean, George is the one example where because it's kind of a turnout driven state and because, like, you can imagine a Republican turnout being quite despondent in the Senate race. But then again, the Republican governor candidate is self-funded and has shitloads and shitloads of money, right? Yeah. Bottoms would be the one where if she wins and that race is very close in the polls where you might give an extra gold star to John Ossoff. OK, Nate, I know you're going to have stuff to say on this. On Tuesday, Marist put out a poll showing Troy Jackson up by two points in Maine among likely voters. However, among registered voters. The poll found Jackson has a nine point advantage. What gives? What's going on here? Yeah, Silver Silver Wolf has a question about that. Hey, guys, what the fuck is up with that Marist poll? That's Silver Wolf. Is Silver Wolf like an ethnic white from the outer boroughs? Yeah, is it New York, New Jersey? I mean, you you you said on Twitter some some rough things. And then you also said, I guess I should be you know, I should give credit to Marist for for for, you know, pubbing this like sticking with their kind of weird results, right, or their consistency in this weirdness. What's what's going on in this poll? So obviously, the poll is incoherent and kind of indefensible in one way or another. And you can aggregate that. I don't care. I'd love to. I'd love to hear the Marist people explain what the fuck's going on with their poll. So I'm going to focus on the gubernatorial race. First, because this is the far weirder result, right? In our polling average, the Democrat Hannah Pingree, am I saying that right? Or is it some weird French Acadian thing? Hannah Pingree has been leading by around 10 points in almost every poll in the registered voter version of the Marist poll. Hannah Pingree was ahead by 11 points in the likely voter version of that poll. She was tied. So they're showing an 11 point swing. They are also showing. Undecided in that race, jumping from 2% among registered voters to 6% among likely voters. That makes no fucking sense, right? No fucking sense. You can have like a enormous turnout differential and say, well, what if Pingree gets only 60% of her vote and the Republican gets 70% of his, right? Even that doesn't close the gap, right? I think they're terrified to publish their actual result and are hurting. I think it's simple. And they had to reverse engineer some cockamamie way to do this that totally fucked up their gubernatorial poll, which now is totally out of line with any other gubernatorial poll in the race and that nobody believes is, you know, I, you know, it, it doesn't make sense. Um, now could it be that kind of, this is really not a registered voter versus likely voter poll, but like two entirely different. Different ways of processing the same inputs maybe, but like the fact that you increase the number of, um, undecideds in a likely voter poll, usually what happens is I take a likely voter poll. You have fewer undecideds because people who are undecided at this point probably are not going to vote. Right. I think there could easily be, well, there are two explanations, right? Either their algorithm is buggy and needs to be repaired or they're scared of their own shadow, right? There's not a flattering explanation. These polls make no fucking sense. Stuff like this, like hurting has gotten worse over the decade of Trump. Has it always been bad? And now more people are tuned into it. I do want to point one thing out, which is that Ann Seltzer, who Claire, you did the profile interview with, right? She was sued by the profile. She was sued by the president of the United States for publishing an outlier poll. There's no crime in that. To be wrong. Right. You know, I don't know what kind of chilling effect that would have potentially, but I don't know, possibly not zero, right? And there are some cases where this like, so in Maine, there was also, excuse me, in South Carolina, there was a Citadel poll that showed any Andrews winning unregistered voters and losing among likely voters. That kind of makes sense given the history of South Carolina, right? And South Carolina is a low turnout state. Turnout matters. Maine is a high turnout state. The democratic early voting numbers look fine to good there. They often do, but a slightly wider spread than in the past. And like, um, they weren't comfortable. They weren't comfortable publishing Jackson plus nine or whatever the raw number was, right? Maybe they're not comfortable because they don't believe their turnout model. And it's kind of a case of two rights making a wrong, but they did very weird things. Like the percentage of college educated voters was lower in their likely voter sample than the registered voter sample, right? That does not happen in midterms. I do think that hurting can be a problem and people aren't straight up publishing fake polls that show, you know, 48, 48. Like a lot of the polls that we saw in the final stretch of 2024, because folks are waiting and putting likely voter models on the results. Like you can tinker with things a little bit that get you closer to where you feel like things might be. And sometimes that can also be subconscious, not that you're doing it actively. It's just part of the process. And as we've talked about this type of polling, especially now is both a science and an art. I do want to play like devil's advocate. I don't have a strong position on this Marist poll, and it sounds like Nate has dug into it a lot more than I have. Maine is a little bit of a unique state in that it's the oldest state in the country. And Susan Collins is also an old school Republican. And so you might expect things to happen a little bit differently in Maine than they would elsewhere. So oldest state in the country means that the people who are likely to show up are an older voters vote at higher rates. Then young voters. And so the people who are likelier to show up are these older voters who, you know, maybe Susan Collins fits as an old school Republican. And when you, you know, get, I'm going to interrupt you because their poll actually has youth turnout increasing in the likely voter version. This poll is, is indefensible, right? If they aren't sure whether Maine is a, and by the way, Jackson plus two is a completely reasonable as a, as a forecast. It's completely reasonable as you know, our model is close to Jackson plus two. It's actually a state. I wrote this weekend where I'm subjectively a bit less confident in Democrats chances than flipper. Right. But like that, are you basically saying that this is maybe like an entrenched problem at that? Like they're kind of forced to put out their polls or you're saying they should withhold the polls. I mean, what, like if you publish two versions of the same poll in the same state that have a plus 11 and a tie in the same race, you should not be publishing that poll, right? You should look at what the fuck is going on with the way you're collecting over your voter sample or the way you're likely voted. That's. That's not publishable, right? That would be, if I had an intern send me that poll, I'd be like, okay, bro, we got to improve here like that. It's, it's not. Um, okay. I, you hired me. Okay. Let me say, let me say one more thing about the Marist poll, right? You should not be using likely voters to fix your problems with your sample, right? It may well be. It may well be that they think their registered voter sample is fucked up. So fix that instead of saying there's going to be an 11 point swing, which is like quite close to like statistically impossible, right? Like it's just bizarre. Still counting is brought to you by Sundays. Sundays for dogs is different. It's human grade, real food. You'd feel good eating yourself. The Sunday's recipes contain 80% plus. all natural meat, roughly twice the meat content of some leading frozen brands. brands, plus nutrient-rich ingredients like kale, ginger, and blueberries gently air dry rather than exposed to high heat. No fillers, no synthetic additives, no chemicals, just simple, complete nutrition. It doesn't look or smell like dog food. It looks like high-quality jerky. And it's vet founded and vet formulated, created by Dr. Tori Waxman to meet her high standards as both a veterinarian and as a dog parent. And the best part? It's effortless. No fridge, no freezer, no prep, no mess, just scoop and serve. I hope listeners are sticking around to hear this ad just to hear how good you are at reading things. And by good, I mean. You are so good at reading things. You're so good at reading things, Nate. You know how to read. You get the quality of a home-cooked meal with the ease of kibble. Over 100,000 dogs are now eating sundaes, and many dog parents notice the same things. Shine your coats, better digestion, more energy, and finally getting excited at mealtime. For even the pickiest eaters. We don't have dogs, but we all know dogs. And we love dogs. Right, guys? We also all know people who get excited at mealtime. That's right. Make the switch to sundaes. Go right now to sundaysfordogs.com slash count and get 50% off your first order. Or you can use the code COUNT at checkout. That's 50% off your first order at sundaysfordogs.com slash count. sundaysfordogs.com slash count. Or use code COUNT at checkout. Oh, we got it. We got it. We got to represent the dogs, not just Flipper. Crown Coins Casino. Winning never sounded so good. What would you do if you won big? At Crown Coins Casino, real people win real prizes every single day. Hi, this is Michael from Georgia. I won 40k and finally renovated my home. Visit crowncoins.com. Sign up in seconds, enter code crown and get a free lucky wheel bonus. You may win up to 10,000 sweeps coins. Crowncoins.com. That's C-R-O-W-N coins.com. No purchase needed. Code valid for new users only. Void when prohibited by law. 18 plus terms and conditions apply. There are so many headlines about this. So many headlines about the early vote. The New York Times and NBC are tracking early vote data in the Senate battleground states, which is so far outpacing 2022. And we have many, many listener questions on this topic. So let's start with Billy. Billy asks, By the time you answer this question, we will have received a good amount of early vote data, both in person and vote by mail. What is your view of which way the data leans if forced to choose? Prens, Billy says. Please don't give the usual early vote data is hard to interpret and polls are still more instructive answer. And I appreciate Billy knowing which way most people answer that question. Guys, don't bullshit Billy. He wants the straight dope. Okay, so here's a straight story. Early voting data is hard to interpret. Okay, that was the most predictable setup. You know, I think that this year is different. It feels different in a couple ways. One, because we're still in a time when voting patterns are changing, right? Pew actually asked voters, how do you plan on voting this cycle? And fewer Democrats said they planned on voting by mail. And this could be because people have concerns about the integrity of the US Postal Service after Trump has been sort of, making his claims about the illegitimacy of mail voting or trying to put your give the USPS more responsibilities of trying to sort through which ballots should be counted, whatever. So that may be a signal to Democrats of do less voting by mail, which could mean that they do more early voting in person or on election day. And so in an environment where things are still shifting, and we don't have well established patterns, like we may have had for like the decade prior to the pandemic, it's important to keep in mind that things are still shifting. That being said, I know I did still sneak in a copy out there. My apologies, Billy. That being said, there's more folks really trying to glean stuff from the early vote in a rigorous way this time around. The New York Times is doing it. Zach Danini at Vote Hub is doing it. And what they're trying to do is not just figure out who is returning ballots, because oftentimes the analysis will just end at, well, like Democrats are outpacing their early vote numbers from the last midterms or the last presidential election. They're outpacing their early vote numbers from the last presidential. What they're trying to drill down on is whose low propensity voters are turning out for both Democrats and Republicans, because that could be suggestive of a turnout advantage overall. And what's the summary of that? Who are the low propensities? This is going to go back to what we were already talking about. We are seeing very high black turnout in Georgia and North Carolina compared to this point in previous elections. We've already talked about that a little bit. We're also seeing very high turnout of Hispanic voters in Texas. Now, again, we don't know how those people voted. We just know that they voted. And South Texas has been very swingy, very idiosyncratic, hard to predict. Does really high turnout in South Texas amongst Latino voters portend good things for Democrats? I would think, given the way that they have swung over the past two years, yes. But I would not bet the farm on it, because again, even if we know, you know, some of the details around like the racial makeup of the area where the votes came from or the registration, we simply do not know how they voted. Early voting data also seems to show, as polls do, and we'll get to that, a big turnout gap among young voters, right? Trump picked up a lot of votes from young men in 2024. As I said before, young men aren't the most reliable people to turn out in midterms historically. Or in general. And you see that young Democrats are turning out. Young men are not reliable. Or in general. Period. Period. Thank God I'm not a young man anymore. That's another gap, right? Where young Democrats seem to be turning out, young Republicans who are maybe new converts to Trump are not. Look, I think overall, the early voting data seems to be quite consistent and validating of the polls, which shows, you know, maybe Georgia and North Carolina not being particularly close. Something very. Very positive for Democrats going on with Black voters. Some of these irregular Trump voters either not turning out, or we can't tell who you vote for just whether you voted or not, right? You know, Nevada is another state where our friend John Ralston, friend of the newsletter, has been tracking early voting data in Nevada for years. And in that state, it's quite predictive. We're down a lot across voting in Nevada. Most people vote before Election Day. Those numbers seem quite strong for Democrats in our Senate. There is a gubernatorial race where perceptions have shifted from lean Republican to a flip there, potentially. Yeah, like Patrick Graffini, kind of, who has been one of the more bullish forecasters, and a very smart guy on Republicans this year, is like, well, I collected early voting data and to kind of, I'm paraphrasing here, right, to kind of challenge narratives. And it's actually not great for Republicans. It's not bad, but not great, right? So like, I don't know. There's a question of whether it's. From a forecasting standpoint, this should probably give you more confidence in Flipper and polls and models, right? The reason why I'm not out over my skis is because Flipper is already a little bit more highly confident relative to prediction markets and models, not a huge difference. But it seems like the least ambiguous. And Arizona, the numbers seem not great for Democrats, right? But it seems less ambiguous than in the past. But as Claire mentioned, important. Democrats are terrified of their votes not being counted if they're sent in by mail. And so, you know, and they might be terrified of like, what if Trump tries to shut down late ballot counting, right? So they want to vote early in person. And those different tranches, if there is a big surprise, despite what are at this point strong Democratic numbers in early voting, it will be differential motivations to vote in different ways. And that is an important, important, not just coming or yes, caveat. Yeah, I just want to add, one more thing to that, which is first, that the early vote is not included in Flipper in case people were wondering. But the other one is, some of this, again, differences in the way that people vote. And I think Kansas looking at the New York Times data illustrates why we have to be careful, right? Registered Democrats account for 46% of early voters versus 40% of Republicans. But roughly half of the early ballots come from Johnson County, which is pretty Democratic leaning. That normally only accounts for a quarter, a quarter of the statewide vote. So the electorate that we're seeing right now is geographically unbalanced. Do we see Republicans come out on election day and bring that to a more balanced place as we saw in, for example, 2024? Or is this, are they capturing a turnout differential? We just can't, like at this moment in time, nobody can tell you the answer to that question. Here's my simplified way of thinking about it, right? In midterms, there are three questions. Democrats turnout, Republicans turnout, and what's happening with swing voters. We've probably done this before, right? And in midterms, I think that's the correct balance. In presidential years, where turnout is higher and more balanced, then I might say we would need to really put more weight on persuasion, right? But like, I think we have ample evidence that Democrats are going to turn out in this election, right? So to salvage the election, Republicans need to turn out their And so they kind of need both those things to go right. and that's kind of where you get to flippers 75 percent right if republicans turn out their voters to match these early democratic votes which they very well might do and if they hold down these democratic candidate advantages and maybe susan collins pulls it out or mike rogers in michigan whatever else right then they can win it's not in the least far-fetched to say rebels could win but like what if both those things go wrong that's when you get like the south carolina flipping that way that's still i still we just updated flipper by the way as you're taping this andrews is now up to 31 percent in south carolina that's what's not crazy not crazy okay we've covered the bases that we absolutely wanted to cover and we got i think close to 100 questions sent to our email address still counting at crooked.com this past week so we're going to try to run through some of them maybe a little bit faster uh as we close out here and claire the first one is for you and it's a it's a broader question about the media it is from josh josh says nate has been doing poll-based analysis data journalism whatever you want to call it for something like 20 years the way political data is discussed in the media and general discourse has changed a lot in that time has his journalistic project been a success oh my god i'm curious for you to evaluate oh wow well i do think it's true that um you now hear even in the you know in the decade that i've been doing it or in in the in the in this realm i'm not doing data journalism um but i do think that political reporters are better at you know describing the idea that okay this is one poll that i'm talking about to get a better picture of things you should look at the aggregate of polls i really think that that message has gotten out there the thing that's obviously changed in the past decade is well in some ways it's gotten more complicated right because just the changes in our our habits around the you know people screening their calls more polling has gotten harder we've also got a more distrustful media environment i mean not just media environment people are more distrustful of american institutions across the board from like government to doctors you know so um so people are more distrustful and i think that the questions that people ask around polling and politics are more existential and nuanced and like i sometimes have them so i guess it means nate's project has succeeded and that people um are more educated on the basics but i think it also means that the more sophisticated media consumer or unsophisticated media consumer has more questions about the validity or whether or not polls are um you know manipulating election results right then you get you we just get into really meaty territory and so it it you know i really we were talking about this a little bit earlier the ann selzer poll which is obviously she she she does it with the she did it with the des moines register i think it's very clearly protected under the first amendment um but i don't think that's an outlier instance where trump is suing a pollster but i think it's fair for people to have questions about you know what polling does like i think the whole electability conversation that we end up talking about is so you know skewed by polling yeah so polling based polling based that's a good jumping off point for two follow-up questions one from homa and one from rob homa says why are we still talking about polls isn't it true that whoever answers to any kind of poll is a weird person a normal member of the general public doesn't answer polls or surveys or even any type of phone calls especially from strangers um i think that's a good jumping off point for two follow-up questions one from homa and one from rob says how do i respond to skeptical friends and family about polling methodology when i get i've never once gotten a call these pollsters are only engaging with supporters etc etc so not only is there skepticism amongst the general public there's skepticism amongst the people who probably consume the most polling based analysis and content still counting listeners i think this year has featured the most sophisticated discussion of polling and polling errors that we've had in a long time and i think that's a good jumping off point for two follow-up questions one from um which is the work of a lot of smart people who have been doing this for a long time and and some smart new blood right you know the early voting discussion is better because of people like zach danini at vote hub right who actually came in ahead of time and said we're going to think about this ahead of time and not just ad hoc all the way um you know everybody kind of understands that a correlated polling error is possible they understand that democrats have tended to underperform their polls in in red states the seltzer poll was kind of in some ways the death of poll as poll and the kind of birth as poll as model or as forecast right like that's kind of what i think the marist poll represents not in the right way to do this but like i think plus two represents maris's forecast of the race mediated by their fear of being criticized and or sued by president trump perhaps right and i think their poll was actually plus nine right but they maybe have not modernized their polling techniques and their little seltzery with the way they're they're one of the more traditional polls and have missed high on democrats in in recent years right so from like a from a flipper standpoint we actually kind of we almost want we want people hurting right but we kind of do want people like it's a consensus based product right it's a consensus product and it's like almost like it's almost like a prediction market almost right yeah so it's fine for like prediction markets but like um do we really fucking i just saw the nathan fielder movie this weekend do we really know what the fuck's going on right um not until we get actual hard data in the form of election results really the answer is you should be skeptical of the polls right like the polls are imperfect there is an average error however unfortunately all their you know yeah all their error they are the best tool we have yeah i would say just to add on one little thing i you know i think that the criticism of the kind of journalism that we came up in the journalism that this podcast dabbles in not dabbles in traffic's in is that we are telling you what the data says right and that often leads to people saying like well you're just shilling for centrist candidates right because it always comes into we always come into the electability conversation what does it what does candidate quality mean and i think for many people uh many journalists right you often think about like okay what does the horse race say where's the candidate going but many partisans many people who are in the activist and movement side of politics say you're miring us in um you know a certain kind of politics and i think that it's important like i just want to i'm not weighing in on either side but i'm acknowledging that in the ecosystem of politics or horse race coverage does often end up sort of i think leading people down the garden path of um probably where most americans are right which is in the center and in the ecosystem of politics there are obviously movement and activist actors who have loud voices who say you are stunting the conversation around american politics i just like i think we should acknowledge that in the conversation around polling and the kind of conversations that we end up having to me that's the joy of data it forces you outside of yourself i live in new york city amongst a bunch of freaks so how am i ever going to learn what the rest of the country thinks it's through data like this which is why i love it still counting is brought to you by zip recruiter we're all trying to save time these days it's like days before the election if there's any time of year that i'm trying to save time it is right now i mean come on nate's not even sleeping we all have our own time saving hacks for everyday life but if you're a business owner how can you save time hiring that's zip recruiter zip recruiter has a new feature that quickly lets you see the most interesting and qualified candidates first so you save time by meeting the right people faster and now you can try it for free at zip recruiter.com slash counting zip recruiters powerful matching technology finds qualified candidates quickly candidates can tell you in their own words why they're interested in your job and there's no wonder that zip recruiter is the number one rated hiring site based on g2 save time and meet great candidates sooner with zip recruiter four out of five employers who post on zip recruiter get a quality candidate and they're going to save time and they're going to within the first day try it for free at zip recruiter.com slash counting that's zip recruiter.com slash counting meet your match on zip recruiter winning never sounded so good what would you do if you want big at crown coins casino real people win real prizes every single day hi this is chelsea from dallas i just hit the jackpot won over 1 million 500k visit crown coins.com sign up in seconds enter code crown coins casino and get a free lucky wheel bonus you may win up to 10 000 sweeps coins crown coins.com that's c-r-o-w-n coins.com no purchase needed code valid for new users only void were prohibited by law 18 plus terms and conditions apply i have another question for you claire from kevin in the past few federal elections we heard a lot about the gender gap this year i haven't seen much about the gender gap but then i found out it's been replaced by the masculinity gap I will say, I don't know what that is. Maybe that says something about me. So I'm asking about the masculinity gap and what role it's playing in this year's elections. Is it a thing? Is it changing votes? I imagine they mean amongst candidates. You know, it's actually funny when we were having whatever, 15 minutes ago, we were talking about young men and how they're not reliable. I mean, I do. I do think that during the presidential election, sexism is alive and well. Right. I mean, I think, you know, we in the 2024 election, people were surprised by the shift of Latino men, right? Younger men. There was a lot of talk about men. There really was. And that certain demographic groups, it was sort of the, you know, the age old thing of like, turns out sexism is stronger than racism. Right. Like that kind of age old battle. And so I do think we ended up in the last two years talking a lot about men. You know, we've been talking for 20 years about how men are in crisis. I think there's a lot of really interesting reporting and data about the way. Young men say that they feel that they are raised to maybe not succeed in an increasingly sort of service based or, you know, sit at your desk based economy. There's, you know, lots of conversations around men in America. And it's interesting because for many years we were talking about the George Bush soccer mom and, you know, wither the mother. Right. You know, I actually thought of that again in South Carolina with two female candidates. One of them is a pediatrician. And, you know, like there's. The gender conversation in politics when Trump first started was very men versus women. Actually, Claire, you want to know the the gender gap in that race since we got the New York Times. Yes. University. OK, so it's two women candidates, of course. Annie Andrews, the Democrat, leads amongst women by 11 points. So 54 percent supporting Annie Andrews, 43 percent supporting Darlene Graham. And then Darlene Graham leads by 15. 15 points amongst men. So 55 percent support Darlene Graham, 40 percent support Annie Andrews. To me, that says partisanship swamps gender, at least as far as, you know, the candidates are concerned. I'm not sure that, you know, when the listener was saying the masculinity gap, I don't know if they're talking about the masculine. Like, yeah, the masculinity that we're always talking about, that the Republican Party has sort of taken on this almost, you know, like comical, like the jawline of, you know, military. Members and what it says about testosterone, you know, sort of implying that James Tallarico is a feat. Like maybe that's what it's talking about, where it's like it's just outright saying, like, is it is it sort of the idea of, you know, maybe Trump playing to his base? The idea of, OK, he won a lot of unexpected male voters in 2024. So let's just like let's get even more male, like put those monster truck wheels on and go. I don't know. I think they're talking about, like, Abdul El-Sayed's masculinity. Like, he's trying to be like Fuckboy Island and like John Ossoff trying to be like, I don't know, whatever he's doing. What about Tim Walls, man? He was a football coach, a real man's man. I have been noting the Abdul El-Sayed, like, I lift, like, literally ads, right? That does feel like a play to men who voted for Trump in Michigan, right? Like, those are those are the Obama. Obama, Trump voters, right? Like, you know, I guess we maybe now they're the I lift voters. Like, what kind of creatine powder do you get at Costco? I don't know. He's right, though, that that that that feels in the age of sort of like the Huberman sort of, you know, gains Maxer that that that political messaging has to shift with that. In South Carolina, I just want to point out that in my research this morning, I don't know if there's a way to, like, show this on the screen later. I came across, like, the first link for Annie Andrews, and it's like an Act Blue page that has Darlene Graham looking very dowdy, right? She's in black and white and looking down and has this kind of ugly sweater on, right, with this oversized flag pin, whereas Annie Andrews is in scrubs with a stethoscope looking up or hair done really well, right? And, like, you know, maybe women candidates are allowed to be meaner to other women, and I don't know how that dynamic plays out. Internalized misogyny. Internalized misogyny is a real thing, Nate. But I do I also think it has to do with, like, there's also an age gap among those people. And maybe it's also just showing, like, out with the old, in with the new. I haven't seen, obviously, the imagery you're talking about. But I think people pay much more attention to what a woman looks like or her presentation, whether or not she's running for office, unless you're Trump paying attention to John Husted's ass, in which case wear those form-fitting trousers. Plus. I just see the clip where he used to compliment Ken Paxton's ass. He's like, I don't know. He said, I don't know about that. Badass. I just want to say one thing when it comes to the generic ballot here, because I did. I South Carolina is such an outlier. I want to say that part of what's happened to the gender gap is that men have pulled more even while women are overwhelmingly supporting Democrats. So on the generic ballot in the most recent New York Times, the Seattle University poll was basically 50 50 split. Forty seven percent of men supported the Democratic candidate. Forty six percent of men supported the Republican candidate amongst women. Fifty six percent supported the Democratic candidate. Thirty seven percent supported the Republican candidate. And so the gender gap, at least right now, is not so much that men and women prefer something wildly different, just that men are split and women overwhelmingly prefer the Democratic candidate. Yeah. And I mean, during the Trump era, like setting aside even Roe, if you can write the overturning of Roe. But even Trump policies or Trump candidates choosing to make. Issues like IVF, which is sort of like bipartisan popular, you're alienating women voters even more. You know, so there there have also been, I think, ways that. For whatever reason that we don't have to get into here, strategy wise, that the the the GOP under Trump has just not actually like courted women voters in a way. OK, here's a more wonky, modely question. Kevin asks with so many. Deep red states showing Democrats ahead by a couple points in polls, but with neither candidate above 50 percent. Wouldn't it make sense to assume the late deciders in these states would break disproportionately toward the Republicans, allowing them to win all the red toss up states by about a half a point to two points, given the underlying partisan lean of their states? Nathaniel, so a logical argument, however, I have generally. Found that doing things other than split undecided 50 50 does not improve forecasting performance, right? Look, if you think that the polls are too high for Democrats in red states, the way that you do that is probably by blending in some fundamentals instead of monkeying around with the undecided vote. I mean, again, this is totally wrong. It is true that when there are more undecideds, that there's more uncertainty in a race. Flipper accounts for that. By the way, the thing I did find interesting about the Times poll in South Carolina this morning was it was. it was 48-48 right it wasn't like 42-42 so i think that i think it's actually like a mostly smart heuristic flipper kind of gets there in a slightly slightly different way the thing it's a little more subtle though if you're undecided late in the race you're by definition not typical of um of the overall population you know what i mean think about it this way the people who don't change their minds are people who or sort of have a sense of which way their fault are people who pay a lot of attention to the news people who have oftentimes a higher degree of education those are the most ideological people in the country they vote most frequently they don't really change their minds the people who tend to change their minds are people who don't pay a lot of attention to the news don't buy one party lock stock and barrel they might already have been cross-pressured even if they were voting for republicans regularly because that's how their community votes that's how their parents voted what have you and sort of once there we saw this amongst the latino vote i mean that's a little bit different because there's even less of a sense especially amongst first generation latinos that they're inheriting partisanship from their parents but once you get a sense that your community is is frustrated or ready to vote in a different way than they have been voting that creates a permission structure so like maybe it could be people in the livestock trade in kansas maybe it could be uh you know you know you get you catch my drift there but yes these are atypical people like if you've known how you were going to vote in these midterms at the conclusion of the last presidential you like we're not talking about you right now also if you're undecided on october 28th there's a good chance that you won't vote or you might vote third party or you might undervote that race maybe you have an opinion the governor's race you have an opinion on the governor's race you have an opinion on the governor's race you have no opinion in the senate race you undervote the senate race deliberately right um so that's why i'm like so maybe in some countries in many countries they don't list undecideds they just list the they remove the undecideds and then just add everyone up to 100 right having the undecideds is valuable i've looked at this in u.s elections and international elections and u.s primaries right it is right that when there's more undecideds there's more uncertainty so it's valuable information and pollsters should include it but by the time you get to late october i'm not sure they're going to vote period we got to get to late october i'm not sure they're going to vote we got a question from somebody named agamemnon and i can't not put this question to you all it it's framed as a more wonky question but uh and sort of flip based take but this is also a broader discussion do the classic or deluxe model of flipper take into account possible effects from voter suppression in states like north carolina georgia or texas does this include texas congressional districts where voters might have to drive long distances to go to a polling station this is this is a rather precise question but we historically have gotten a lot of these types of questions which is like well what about barriers to voting how is that represented in the forecast um there's a maybe a specific answer there about flipper but a broader question about the impact overall i guess that must be a neat question since the short answer is no um for various reasons one is that in the short term voting restrictions can be motivating to the population that's targeted with disenfranchisement or increased degree of difficulty right you know one reason why you might see this lopsided black turnout is because black voters feel disenfranchised with a calais decision in the weakening of the voting rights act by the supreme court right and they might feel like their votes might not be counted if they wait to vote late or vote by mail and so therefore they're going to be very certain to vote early in person which is i guess the most secure way to get your your vote counted right you know another reason is uh we i guess it helps in some sense to have like a benchmark of like what did the polls say if there are shenanigans i don't want to get too cute about that right but like it's i don't like blending things right it's like here's what the polls say assuming that votes are counted and tallied properly but there are reasons a little more subtle too which is like you have to be very careful even if you were machiavellian who is agamemnon was agamemnon like a greek god or was a mac anyway i'm getting confused here trump and republicans often now have the more marginal voters and you have to be careful if you're being even if you're being cutthroat if you're trump that you don't wind up making barriers to your own voters to voting right there was one year in 2020 when we included shifts in ease of access as like it filtered into like the prior of the partisan lean score basically and i think it held up as being somewhat useful in historical data however since 2020 we're really over this entire like trump era we've seen a shift from democrats being the irregular marginal voters to republicans being those irregular marginal voters instead and so so it's not clear directionally even if like greater voting restrictions would help with trump yeah i mean it's clearly like the huge thing that it's it's hard for us to i mean we want to talk about it but it's not necessarily database is like how are like what potential um you know voting interference could happen and whether or not that's like voter intimidation of latinos right by placing you know immigration agents around polling places or whatever you want to put it as and this is something that like you know the political parties have been doing for decades mostly the republican party i mean sort of you know barry goldwater and the nixon era was kind of you know famous for um for polling place intimidation i mean like i think that's something that it's basically like you have to watch on election day and we're already seeing i just think we should note obviously the doj um has filed a suit in the state of alaska um sort of setting up the casting of aspersions on a close result between mary peltola and dan sullivan because of the dan sullivan's plural on the ballot so there's obviously some kind of like a kind of like a stuff that will unfold in this election no matter what as claire you're suggesting there's an array of different ways that if trump intends to try to mess with the you know active voting the process of counting the certification of results he could you know we take this more seriously because he has and we'll be tracking that you know in the final run-up to election day when we talk about the results after the fact and so on and uh so we will come back to those questions i mentioned we also talked about the different array of things that could happen um on a recent gd politics podcast i want to get to some rapid fire questions that are a little more fun than voter intimidation before before we close here uh the first one is a quick one for you nate andrew says does flipper model take into account prediction market odds no and i i think it's a terrible idea to do that we want to a we want to beat the prediction markets right no you idiot no it's it's becomes a self-consuming snake eating its tail right you want independent flipper might be wrong but flipper is an independent assessment every day and that's what you want okay here's a question from tom in what election cycle do you think people were most tuned in or least tuned in personally i found 2016 to be all encompassing when it happened yeah wow you can't even compare 2016 was just like it was the beginning of the cultural movement it's like it was it was uh like the monoculture's last stand that was that was it it was our baptism as you know 538 the 538 politics podcast 2020 had the highest turnout yeah because people were bored during fucking covid but it's different to see like midterm turnout has been higher in recent midterms and you know frankly we see that numbers for the show and the various content we produce right like midterms are kind of two-thirds of presidential year interest now instead of like one-third to put a rough gauge on it yeah okay finally sherlock hound says since everyone knows that 5e fox is nate's fursona what animals would claire and galen be claire's favorite question uh first of all i just like i can't not bash 5e fox enough hated him did not like 5e fox i know i can i just i like what is going on there where does the hate come from i just think it's tacky i just don't like clip art um i think i guess like you know the greatest animal ever greatest movie star ever was miss piggy so maybe a pig quiet piggy quiet piggy to quote yeah and i'm a woman right to quote the president of the united states quiet piggy give me a t-shirt for my birthday the republican commentators wanted to pretend that the masses can't understand camp even though miss piggy is an icon of oh my gosh our childhoods come on people like in the last six months i have watched some original muppet shows and man and the new muppet show which is good and like that that stuff holds up miss piggy miss piggy's humor reeks really holds up i'll just say i'd be an otter for the gays um and we're gonna leave it we're gonna leave it at that that is it for today's episode of galen did you provide your answer yeah i said an otter otter oh my god which is not a twink not a bear an otter um okay like i said thank you guys i'm excited we're gonna get to pod again before next wednesday uh folks yes spread the word people make people in your life listen to this podcast tune in on monday come for wednesday after election if you live in dc come and see us in dc but um it makes a really big difference if you actually tell your friends and family about the show and please do it yes and rate us review us yes uh but only not only only nice only nice let's if we're if any polls are getting rigged this election cycle let it just be that one go rate us but only if you're going to give us a five-star rating right if you don't have anything nice to say don't rate us if you don't have anything nice to say don't rate us if you don't say anything at all even if it's a three-star among registered voters you can make it a five-star among your likely voter review and that is the perfect button for today's episode that's it for today's episode of still counting listen wherever you get your podcasts and subscribe on youtube.com at still counting podcast we'll be back next monday for our premium term bonus episode and then again on wednesday evening for our regular episode looking at the results still counting is a crooked media production executive produced by galen drew claire malone and nate silver our team includes sol rubin emily for claudia shang charlotte landis reed churlin johanna case ben hefcote matt degrote katie long and adrian hill our music is by breakmaster cylinder our production staff is proudly unionized with the writers guilds of america east winning never sounded so good what would you do if you won big at crown coins casino real people win real prizes every single day hi this is michael from dallas and i won 40k and finally renovated my home visit crown coins.com sign up in seconds enter code crown and get a free lucky wheel bonus you may win up to 10,000 sweeps coins crown coins.com that's c-r-o-w-n coins.com no purchase needed code valid for new users only void were prohibited by law 18 plus terms and conditions apply looking for a new way to take control of your health if diet exercise and willpower alone haven't gotten you where you want to be visit orderlymeds.com to learn more about glp-1 treatment options with a convenient virtual process you can connect with licensed medical providers from home and see if treatment may be appropriate for you. Visit orderlymeds.com today. That's orderlymeds.com. Compounded medications are not FDA approved. Eligibility determined by a licensed provider. Individual results may vary. See website for details.