David Beard:
Hello and welcome. I'm David Beard contributing editor for Daily Kos Elections.
David Nir:
And I'm David Nir, political director of Daily Kos. The Downballot is a weekly podcast dedicated to the many elections that take place below the presidency, from Senate to city council. You can subscribe to The Downballot wherever you listen to podcasts. And please leave us a five star rating and review.
David Beard:
Election season is in full swing. And that means we have a lot to talk about. So what are we going to cover today?
David Nir:
We have a bunch of things to discuss in including the GOP's continued abandonment of their Senate nominee in Arizona, a huge scandal that erupted in Ohio's 9th Congressional District involving stolen valor, a fascinating new study on the Latino electorate from a major research firm, and a new map that we produced in house at Daily Kos Elections that we are extremely excited to share with all of you. And then we are going to be talking with TargetSmart CEO, Tom Bonier, whose tweets you may have seen about fascinating registration statistics across the country. He has a lot to tell us, and we have a great episode ahead. So weird. It seems like the situation for Republicans in the Arizona Senate race just keeps getting worse and worse and worse; what's going on now?
David Beard:
It's really one of the stranger situations I've seen in following sort of ad buys in politics over many years at this point. But the Senate Leadership Fund, which is the independent expenditure PAC that is connected with Mitch McConnell and Senate leadership canceled the remainder of its nearly $10 million reservation in Arizona after previously canceling $8 million just before Labor Day and the NRSC canceling $3.5 million, all of which was supposed to go to help Blake Masters beat democratic incumbent, Mark Kelly. But now none of that money is being spent by national Republicans and they seem to be continually pointing towards Peter Thiel, who Masters once worked for, who spent $15 million to get him through the primary. So they keep asking Peter Thiel to come in, spend that money again, be that backer of Blake Masters to get him through this Arizona general, and Thiel just doesn't seem interested in it.
David Beard:
So we're left in this position where money keeps getting canceled and replacements not being brought in by the Republican party or by any sort of outside group either. So Thiel's PAC has spent about $1.7 million since early September, including reserving a 30-minute infomercial that's going to premiere October 1st. Which is quite a strange use of money, but according to multiple outlets, that money isn't coming from Thiel himself, but from other donors. So he just seems totally uninterested at this point in spending any more money on Master's behalf and with the national Republicans also no longer interested it seemingly in spending money on Master's behalf, it's very unclear how he's not going to get absolutely destroyed in the ad-buying arena with the amount of money that Kelly has and the outside money Democrats are going to be putting in. So it's going to be very hard for him to come back from this.
David Beard:
We've seen like TV advertising has a real effect, and usually that effect is canceled out because both sides spend roughly equal amounts of money in top tier competitive races. So any top Senate race or top governor's race, the money goes in and largely cancels out, but you have to spend it or it doesn't cancel out. And overriding TV ads from one side is really seriously going to hurt you in persuasion audiences. So it's totally inexplicable what Republicans think it's going to happen here, or if they're just sort of abandoning Arizona to Peter Thiel's whims.
David Nir:
It seems like they are making the biggest desperation play ever by threatening Thiel's protege saying we're going to abandon him unless you show up. But so far Thiel seems happy to let Masters do his own thing. Same thing with his other candidate, JD Vance in Ohio. It's really, really strange, and on top of that, it's not as though these big Republican groups, the SLF and the NRSC seem to be pouring more money into other races that they think are better bets. It really feels like they actually are trying to conserve resources here.
David Nir:
So yeah, just really inexplicable. And if you sort of step back a little bit further, the GOP path to taking the majority in the Senate just seems really narrow now because if John Fetterman's lead in the polls actually holds that means Republicans would need to flip two seats. And if they are actually giving up on Arizona, that means there are only really three plausible other states that they can make flips in. That's Georgia, Nevada, and New Hampshire. But we saw just how disastrous their New Hampshire nominee's Don Bolduc with his views on abortion. So it just seems that the GOP path is incredibly, incredibly tight at this point.
David Beard:
Yeah, It's really not what you would've expected six months ago, but here we are.
David Nir:
Well, here is something that we would've expected six months ago in Ohio's ninth congressional district; Republican JR Majewski was just busted this week in a new report from the Associated Press that said that contrary to his claims to having served in combat as an air force veteran in Afghanistan, Majewski actually never deployed to the country. In fact, he didn't get any further than the US ally of Qatar, where he helped load planes at an air base very far from the fighting. Majewski is a real piece of work. We've talked about him before he is a QAnon conspiracy theorist, he put together pro-Trump rap videos that are obviously terrible, and he even went to the Capitol and appeared to participate in the insurrection on January 6th. He is absolute scum. And the good thing is that longtime Democratic Representative Marcy Kaptur has been running really aggressive ads, targeting Majewski for all of the ways in which he is a threat to our Republic.
David Nir:
This is a district that Republicans gerrymandered quite extensively. It used to stretch from Cleveland to Toledo. Now it's just in the Toledo area and northwestern Ohio, and it got significantly redder. It went from being a district that Joe Biden won by 20 points to one that Donald Trump would have won by about four points. So just an enormous swing. Republicans obviously are salivating at the thought of taking on Kaptur, but Majewski has been about as terrible a candidate as they could possibly imagine. And just to give a sense of how depraved this is, he didn't just make up a story that he went and served in Afghanistan. He gave some interview last year, where he claimed that while he was in the desert in Afghanistan, on deployment, his grandmother died and that his commanding officers couldn't even find him in the middle of nowhere. And he missed his grandmother's funeral as a result of this.
David Nir:
Obviously a totally sociopathic liar and someone Republicans are stuck with, not because Democrats meddled in the primary. In fact, Republicans had two perfectly standard, at least by their standards, candidates in the primary who were both members of the legislature, but this QAnon, Trump-rapping nut won instead. And it's because he represents what Republicans actually want. Now, again, the district, like I said, it changed a lot, thanks to GOP gerrymandering. So Majewski could still win, but he and his flaws are absolutely keeping this race on the board for Kaptur. And I can guarantee we will see some really hard-hitting ads from her on this score soon.
David Beard:
One of the things we've really seen at the Senate and gubernatorial level is that these bad Republican candidates have a real negative effect on their polling and presumably on their results in November. We never know if that's going to actually change the result, but it's having an effect. And as you get further down the ballot, you wonder the degree to which these effects will break through. Because people just are less familiar, the lower and more localized you get with their congressperson or with their state legislature tend to just vote more for a party or how they're feeling about something broader. But with stories like this, you can really imagine this breaking through and really genuinely benefiting Kaptur.
David Beard:
So one of the things that I've been keeping an eye on since 2020 is the Latino vote, its movement from 2016 to 2020 and where that might lead going forward. And just this week, Equis Research put out a memo about better understanding the Latino electorate and where they stand now that we're about seven weeks out from the election. And they found that while there was about an eight point swing from 2016 to 2020 in Trump's favor, that movement hasn't really progressed or receded from 2020, but largely held still to where it was in 2020. And that comes from sort of two forces in opposition. One of which is Latinos' concerns around economic stewardship under the Democrats, and obviously the past couple of years and inflation has not helped that. There are two contrasting forces that are keeping Latinos from moving either more towards the Democrats or more from the Republicans from 2020.
David Beard:
And on the Democratic side, there's concerns about poor economic stewardship and that hasn't helped in the past two years with high inflation and things of that nature. But on the other side, there's real concerns in the community that Republicans just don't care about people like them, working-class and middle-class Latinos. And that's preventing them from really moving to the GOP in greater numbers. A couple of other issues have become top of mind in addition to the economy: gun control, particularly in the wake of Uvalde, and abortion, of course has become top of mind for practically everyone in the wake of the Dobbs decision. But the important thing that they emphasize is that you can't just substitute these issues for the economy. The economy remains a key factor and the thing that people are most concerned about in the Latino community, and that has to be addressed front and center.
David Beard:
They also observe that among 2020 Biden voters who disapprove of the president's job performance, young Latinos, Latino men, and self-identified conservatives are overrepresented among that set. So those are really the folks who are most persuadable. They don't approve of Biden's record, but they voted for him in 2020, and presumably are open to voting for Democrats, but have concerns right now. And another flag that they had that was especially worth watching was the Spanish-dominant community, which is Latinos who prefer to receive their news and media in Spanish. And that had some of the highest undecided rates among Latinos versus those who prefer to receive information in English. So that's something that obviously campaigns can work on specifically and push to... Democrats tend to spend a good amount of money on that, pushing to the Spanish-language television, Spanish-language radio. And that's an important way to reach voters who have a tendency to be more undecided than English language voters.
David Beard:
And a couple of states where this is the most important is, of course, Arizona and Nevada. We saw Republican gains in south Texas, in south Florida, less so in these really key swing states of Arizona and Nevada, and Democrats really need those margins to hold up, to hold those Senate seats and to either hold the governor's seat in Nevada and pick up the governor's seat in Arizona. And then despite being smaller communities, they can have a real impact in states where there's going to be really, really close races in states like North Carolina, Wisconsin, Pennsylvania, where we've seen close races in the past and the Latino community can make a real difference. So I found that to be a really interesting summary of where the Latino community is, where their votes are and what we should look for as we move towards November.
David Nir:
And that study was from Equis Research, E-Q-U-I-S. You can Google them and find that online. One last thing we want to talk with you about is a new map that we at Daily Kos Elections released just this week. We are super excited about it. We call it our congressional district hex map, and you can find it by going to our Twitter account. That's @DKElections. It's right there in the pinned tweet at the top.
David Nir:
Or you can go to dailykosdata.com, and you will find the hex map in our repository of all of our data sources. And I'll give you a moment if you'd like right now to go pull it up, because it'll certainly be easier to understand this map if you have it in front of you.
David Nir:
But in short, maybe the best way to sum this up is you might remember those garbage maps that Donald Trump used to love circulating that showed his 2016 victory on the county level. And Trump loved those maps, of course, because they were painted heavily in red.
David Nir:
In fact, he even retweeted one that said "Impeach this" shortly before his first impeachment. He supposedly handed out copies of the maps to visiting dignitaries or reporters to brag about his win. But they're totally, totally misleading for the simplest reason of all, which is that land does not vote. People vote.
David Nir:
And much of the United States, especially out west, is incredibly sparsely populated. So, these county-based maps, or these just traditional maps of any kind, totally over inflate Republican performance, and it makes Democrats look invisible. The Daily Kos Elections hex map is designed to remedy that problem.
David Nir:
On a traditional congressional district map, obviously the rural districts, just like the rural counties, look enormous, and the urban districts disappear into invisibility just because they're so small and packed close together. On our hex map instead, we render every single one of the nation's 435 congressional districts to almost exactly the same size. That way every single seat pops out.
David Nir:
So, whether you're looking at election results or the outcome of a vote on the floor of Congress, every vote is represented equally. And that only makes sense because each congressional district has a roughly equivalent population. And also, every member of Congress obviously only gets a single vote on the floor of that chamber.
David Nir:
Now there is one big caveat with drawing a map in this way. We preserve the states of shapes. That's been brilliantly done by our own Daniel Donner who created this map. But in many of these states, urban districts will wind up appearing to be in a different part of the state than they actually are in real life.
David Nir:
It's especially true in New York state, for example, where you have so many districts in New York City. But if you want to represent them all of equal size, the urban districts get shoved upward into Upstate New York.
David Nir:
Now this is simply a trade-off that you have to make if you want to show all districts and make them all visible. But we think that especially when you use this hex map alongside a traditional map, you get a much more accurate portrayal of election results at the congressional district level.
David Nir:
If you click on the links that I mentioned at the top of this segment, you will see, for instance, that we have taken our other new data set, which is the result of the 2020 presidential elections according to the new districts that are going to be used in the November midterms, we have mapped them on this hex map. If you did it on a traditional map, again, the blue districts would be almost invisible. On our hex map, you can actually see the districts that voted for Joe Biden for president. It's really, really illuminating.
David Nir:
And I also want to make clear, this version of the hex map uses the brand-new congressional districts. We have published prior hex maps using older districts before. But the reason why we're so excited about this, of course, is because these are the new districts.
David Nir:
So again, check out the pinned tweet at @BKElections, our Twitter account, or go to dailykosdata.com. And there are also templates that if you are so inclined, you can use yourself to make maps using our hex map to your heart's content.
David Beard:
Coming up next, we're speaking with Tom Bonier, the CEO of TargetSmart, a political data firm that helps Democrats. Stick with us.
David Nir:
Joining us today is Tom Bonier, the CEO of TargetSmart, which is one of the leading Democratic political data firms. Tom in particular has been posting some really eye-popping figures on voter registration statistics nationwide, and we are very excited to talk to him about those numbers. Tom, thank you for joining us.
Tom Bonier:
Thanks for having me.
David Nir:
So, we describe TargetSmart as a political data firm, but I'm not sure if folks necessarily have a particular idea of what that means. So, maybe you could start by breaking down what TargetSmart actually does for Democrats.
Tom Bonier:
Yeah, so as you said, at our core we're a data firm, which means collecting as much information as we can about the American electorate. So, we work with Democratic campaigns, progressive orgs, labor unions. And when they're out there doing their field organizing, when they're doing voter contact, they have some very basic needs, which means we need to have an accurate address for someone so we can go knock on the right door, which doesn't sound super sexy, but is actually important, especially when people move around a lot. Having the right phone number, having a cell phone number.
Tom Bonier:
So, we build starting with the publicly available voter registration data that states put out. And then we append as much information on top of that from as many sources as we can to get a more accurate reflection of who this voter is, and really most importantly, what motivates them. So, we'll build models. We want to understand their partisanship. We want to understand what issues might motivate them.
Tom Bonier:
It's so difficult these days to get anyone's attention for more than a few moments because the media landscape is so fragmented. So, we specialize on trying to find that needle in a haystack to better understand the voter so hopefully you can have a relevant conversation with them in the moment that you might have their attention for.
David Nir:
So, what are some of those other sorts of data sets that you mentioned that you use to enhance your understanding of individual voters that you get voter registration records from the states?
Tom Bonier:
So, it varies, but most common would be consumer data sources, the sort of data that's used for marketing perspectives. For our application, it can help fill in gaps that aren't available at the state. So, for example, and we'll talk about this more shortly, I'm sure, gender. Not every state reports gender with voter registration. And so, it's obviously important to have that data in terms of doing voter targeting.
Tom Bonier:
And so, we'll sometimes rely on consumer data sources where they will have that sort of information, or demographic data someone's likely educational attainment. We saw after 2016 how the polls were wrong. And one of the biggest things they were wrong about was they weren't taking into consideration educational attainment and how important that is, especially when you look at white voters and how predictive it could be of someone voting for Trump versus Clinton. So, data like that, income data, that sort of thing.
David Nir:
So, you mentioned some of the types of clients that you have a few moments ago. Can you be a little bit more specific about how exactly they're using this data that you provide?
Tom Bonier:
Yeah, so there's basically two sides that we're all probably familiar with, what people call the hard side and the soft side. The hard side being direct candidate work and working with Democratic Party, Democratic National Committee, state Democratic parties. And we do a lot of work where Democratic campaigns are using data that we're helping feed into. We're not the only source for that, but we're one of many sources providing consumer information, that sort of thing.
Tom Bonier:
And then there's the soft side. So, that would be labor unions, (c)(4)s, super PACs, progressive orgs. And to them we're providing a full voter file where they'll take this data, and how they interact with it will vary based on the organization and based on their capacity to intake large amounts of data, frankly.
Tom Bonier:
Some have teams in-house where they take in these raw files and they'll perform their own analytics on them. They have their own tools that they're using, and they'll use that to parse the data. So, in the end, where it comes out the other side is maybe one of their organizers doing a door knock and a canvas.
Tom Bonier:
And it's from the data we provide, having the right address, the targeting they do to figure out is this someone who's likely to be a target for our organization to the organizer taking it. And then there's actually a feedback loop there with a lot of these organizations where then they'll collect data at the door and then bring that back in so that they can get an even better depth of understanding of that voter and really what makes them tick.
David Nir:
Now you have lately been publishing a lot of your own research using this data that you gather in particular about the results of the astonishing referendum in Kansas this summer. We have talked about that vote a lot. Of course, abortion rights supporters turned back a measure that would have excluded abortion rights from the state constitution by a huge 18-point margin. This is in a state that Trump won by double digits. I think that that final margin really astounded most people. And you have found some really striking data emerging from these voter registration numbers, specifically about women in that election. Tell us what you found.
Tom Bonier:
That's right. I have to admit, even though I'm a numbers guy, so to speak, and spend most of my time digging into election data, that I was surprised by that outcome. I mean, maybe not. I think most of us thought going in to the extent that we were paying attention to that vote in Kansas that it could be close. There was one poll out there, one public poll that showed it within margin of error.
Tom Bonier:
And then like everyone else, I'm following the election results on election night in early August and see... When I noticed it was when Dave Wasserman tweeted his trademark, "I've seen enough" tweet. And it was less than 90 minutes after the polls closed there, which I thought, well, okay, that's weird. How is their data this early that's so obvious that he's out there calling it? And you look at it. That point, I think it was winning by about 20 points. In the end, the no vote won by, I think, just over 18 points.
Tom Bonier:
And so, I wanted to understand how that happened. I mean, there's an obvious component, right? We all knew after the Dobbs decision was handed down in late June that it would have an impact on this election. I think you'd be hard pressed to find someone who would argue otherwise. The question was to what extent? Will it actually reshuffle the deck in this election, and could it go from a red wave election to something else? I don't think anyone knew. And this was the first hint that, well, maybe something might be up. Let's figure out how this happened.
Tom Bonier:
And so, to your question, what I did is looked at the voter registration data to see, well, who was registering to vote in Kansas leading up to that election. And so, what I did is I set June 24th, the date of the Dobbs decision being handed down by the Supreme Court, as that inflection point, and looked at voter registration in Kansas before and after that date.
Tom Bonier:
And found that before June 24th, it was basically what you would expect. The new registrations were basically evenly split between men and women. And then after, women were accounting for 70% of new voter registrations in the state. And I promise you, I was convinced that was wrong. I was convinced I did... I'm not the most technical person even though I work with data. I'm lucky enough to work with people who are far smarter than me and far more adept with those things. So I'm thinking, I just messed something up. Let me do it again. I did it several times. Finally did it enough times on a few different approaches, it's right. And I've never seen anything like it, but what it was showing is something that actually made sense. But I think we've seen enough things over the last six years, especially, that haven't made sense where you would expect, well, this is going to create outrage. Kids in cages. You could go through the list, and we're all painfully aware of them, where you see these things that should have a big impact and they don't. And this was one case where it should have had a big impact and it did, in this one narrow case. So it was just seeing women engaged at this very high level.
Tom Bonier:
That was the first thing you could see because, to get a little bit into the weeds, eventually every state puts out a list of who's voted in the election, but it takes time. So at that point, the voter registration was all we had. So it was an indicator. It wasn't a promise that those women voted or anything else, but it was an indicator that women were far more engaged after Dobbs than they were before. And it specifically seemed to be centered around that vote on protecting choice in the state constitution.
David Nir:
That's an extremely dramatic shift from essentially 50/50 male, female voter registration to 70/30. How did that break down in terms of raw numbers though? Are we talking about significant numbers that could actually move a race?
Tom Bonier:
No, that's an interesting thing. And I'm glad you bring that up, because it's a question that, when I share this data, usually on Twitter, I'll get that question a lot. And the fact was the overwhelming majority of voters, in any election, are not going to be first time voters. They're not going to be new registrants, but why the new registrants are relevant... And in Kansas it was, I don't remember the exact number, but you were talking about, to be clear, tens of thousands of new registrants in an election where almost 900,000 people cast a ballot. The relevance there was that surge of intensity, enthusiasm, among segments of the electorate generally are first apparent in voter registration data. We saw that in 2018. We asked a similar question in 2018, the Parkland Massacre happens on February 14th, 2018. The March for Our Lives happens two months after. And you see this surge of youth engagement and activism.
Tom Bonier:
And then it's met with this level of cynicism from the establishment that says, well, but young people don't vote in midterms, which has generally been true. But similarly, it felt like, well, something had to be different this time. And so we started tracking voter registration and saw, well young people after Parkland registered at a much higher rate than before. And fast forward through 2018, I think we all remember what happened, but young people not only registered at a higher rate, they voted at much higher rate. They almost doubled their share of the electorate between 2014 and 2018. If that doesn't happen, the blue wave doesn't happen the way it did, and Democrats wouldn't have retaken control of Congress at that point.
Tom Bonier:
And so that's a very recent example of how surges in voter registration... Even in 2018, the youth surge wasn't necessarily driven by new registrants. There was a lot of young people who came out in 2016 in the presidential and who, if past patterns held, they wouldn't have come out in the midterm. They're just these presidential-year-only voters, but that didn't happen.
Tom Bonier:
So in Kansas, yeah, the number was small in terms of, at least from a relative sense, in terms of how many new women voters we're talking about. But the theory there is that that would suggest that women, in general, were much more engaged and that women in general would turn out at a much higher rate come election night.
David Beard:
Now more recently, you've gotten a chance to look at the voter file from the referendum result in Kansas. And you found impressive results, both among women and among younger voters, and voters of color. So tell us how that broke down.
Tom Bonier:
So I was more excited about this than I hate to admit, than most things lately, because again, when you're a data geek and you're an elections nerd, and you have vote history for a key election come out, we wait for it. And then it's kind of this mysterious thing where the state puts it out, and I happen to see that's in our system. So I like rush to start running counts to see, okay, we saw this surge in women registering to vote. Did it actually turn into more women voting? And so for context in Kansas, usually in past general elections, even in primary elections, though this turnout was much more like a general election, women account for somewhere between 51 and 53% of ballots cast. So women are slightly more. They turn out a higher rate and there are more women in general, slightly more women than men.
Tom Bonier:
In this election, women accounted for 56% of ballots cast. So again, we have data that goes back a certain amount of time. I'm not aware of an election in Kansas where women accounted for that large of a share of the electorate. I can't find one. Maybe it's happened, but I'm not aware of one. So the first box was checked that women were coming out at a much higher rate than they had before.
Tom Bonier:
But to your point, it wasn't just an overwhelming surge of women. It was skewed much younger. And to the extent that people say that young people don't vote in midterms, well, this was a midterm primary election in early August. An election that was handpicked by the anti-choice movement for the pro-choice position to fail. The idea that young people, women, Democrats, progressives wouldn't vote. The state was handpicked. Yet young people came out at... Here's the stat that blows my mind more voters under the age of 30 voted in this Kansas primary election than voted in the 2018 general election in Kansas. And like I said a moment ago, 2018, the youth vote was really good. So it was already a high benchmark, and younger voters in Kansas actually exceeded that.
Tom Bonier:
And to your last point, it was a diverse electorate, at least grading on the curve for Kansas, which is not an especially diverse state, but you have a growing Latino population. Latino voters accounted for a larger share of the electorate in this election than they did in any election other than the 2020 presidential. Huge number of Latino voters came out. And when you look at those Latino voters, they were super young. The median age of a Latino voter who came out to vote in this primary election was 35. The median age of a white voter who came out to vote in this election was 55. You don't see separations like that, of 20 years, when you're talking about median age. It was amazing the extent to which this coalition that was women, younger voters, Latinos came out to produce such an impressive result. I'm not sure I've entirely wrapped my head around how unusual that is, let alone the political establishment.
David Beard:
Now we have this great result in Kansas. We can look at the voter file in Kansas, but we don't have something like that for the other 49 states, and that's where these midterm elections are going to be decided. But there is something you've been able to track and that's voter registration data across many more states. So tell us what the voter registration data is saying in states other than Kansas.
Tom Bonier:
So that's the cool thing. What we did is after we saw the surge registration in Kansas, the next obvious question was, is this happening anywhere else? And I didn't expect to see the margins that we saw in Kansas because Kansas was so immediate. There was the immediacy of that constitutional amendment vote before any state has had anything like that. But we ran the numbers in states where they had produced a voter registration update recently enough that it gave you enough time, since the Dobbs decision.
Tom Bonier:
And what we found, and this was analysis that we started back in August, and have continued since then, that we found almost entirely, the consistency of this is shocking. I think at this point we've looked at 46 states. In all but four, of those states have seen women account for a larger share of new registrants after Dobbs than they did before. And in fact, in many of these states, the gender gap that you're seeing is double digits. Huge gender gaps. And these are key states.
Tom Bonier:
The pattern's actually interesting because I think maybe some people initially thought of Dobbs as something, well, that's going to motivate blue state voters. That's a fringe liberal thing. Those are the only people who are really going to get fired up. The real voters are just concerned about the economy and inflation. And what we saw is the opposite. This was something that was firing up voters in red states, that you were seeing women registering at much high rates. And in fact, when you look at the states of the base gender gaps, Alaska, Idaho, Louisiana are all in the top five. And then you also have states like Wisconsin, Michigan, Pennsylvania, Ohio, that all had double digit gender caps. In fact, the smallest gender gaps, which were still representing increases among women registering to vote, were in blue states, which in the end, it makes sense when you look at that.
Tom Bonier:
Like Oregon, Oregon's one of the few states that actually didn't see an increase in gender gap at all. Well, Oregon's a state, first of all, that almost everyone's already registered to vote because of automatic voter registration, and they do a very good job managing their elections there, but also choice is protected and enshrined in their constitution. So it's not something that's necessarily going to motivate voters like it would in the state where they have had to live with a reality where women can no longer access basic healthcare.
Tom Bonier:
So again, to a point I made earlier, this is one of these things where when you look at the data, you say, well, it actually makes sense in a way where we haven't seen things make sense in the past, in terms of a reaction to, really, something tragic happening to this case, I'd say is relevant with the Dobbs decision. So yeah, really around the country, seeing surges in women registering to vote.
David Nir:
Now this seems to buck the broader, or more typical, trend where the party in power tends to lose the registration battle. Was that what we had been seeing before the Dobbs decision came down?
Tom Bonier:
Yeah. I mean, in the end, the party in power tends to lose the ‘everything’ battle in these midterm years. If you look at the history for midterm elections, and to your point, the party that is not in the White House historically wins the midterms by very wide margins. The average House national popular vote victory is over five points. And generally that's then reflected in fundraising numbers, in primary turnout, in special elections.
Tom Bonier:
And what we've seen is when you look, even before Dobbs, Republicans were underperforming. They were still likely poised to go and take back the House and the Senate, but they were underperforming. And I think there were reasons for that. When you look at January 6th, when you look at the extremism of the party, when you look at the kind of candidates they were nominating or were winning primaries, when you look at Trump still being the standard bearer for the party, I think there were a lot of reasons that Republicans were underperforming. But still when you look at the polling, Republicans took a lead in the average of generic ballot polls, national polls, which is generally a pretty good indicator of where things stand in these races. They took a lead back in October of last year, and they held that lead. It wasn't a huge lead, but it was a lead up until not that long ago.
Tom Bonier:
So Dobbs really did change things. And I think without Dobbs, it would've taken something else for Democrats to have a chance in this election, but Republicans were underperforming somewhat, at least in terms of historic precedent.
David Nir:
We've been talking about women voters in particular and also younger voters, and voters of color. And in a way, we are sort of using these categories as proxies for Democratic voting strengths. But in around 30 states, voters actually are able to register by party. Are we seeing any comparable surge in registrations for people who are actually signing up and registering as Democrats?
Tom Bonier:
We are. There's significant surges. That's the interesting thing. Because again, when you share this data on women registering to vote, there are some people who will invariably respond, well, how do we know those aren't pro-life women? A fair question though, as unlikely as it might seem. So we look at party registration in those states that you mentioned, in the states that we don't have a sense of modeled partisanship, but we're also looking at the other factors you mentioned: age, race, ethnicity, that sort of thing. But when you look at party registration, Pennsylvania is one of those states that has party registration. And you look at the difference where before the Dobbs decision, Democrats were actually slightly out-registering Republicans by a few points prior to Dobbs. Since Dobbs, Democrats are out-registering Republicans by closer to 15 points. So there's actually been a huge surge in Democrats. So it's not just that it's women registering to vote at a higher rate, it's Democratic women, it's younger women.
Tom Bonier:
And then you're also seeing younger men beginning to catch up at this point. I think that's another great question that people have asked is when they see these numbers is, well, that's great. It's great that women are standing up and taking the lead and responding to this, but where are the men? Which is a very fair question. Where are the men? And in a number of states now we are seeing the men beginning to register at higher rates. Again, younger men, younger Democratic men, likely progressive men, registering at higher rates. And when you look at Kansas, actually seeing them turn out at a higher rate.
David Beard:
The big question, of course, that we all have, that we don't know is how this is going to translate to the midterm elections. We all have our hopes, but obviously we won't know until Election Night. But what do you think we can take pretty confidently from all of this data and these results that we've seen? And what do you think is more, we'll have to wait and see in terms of interpretations of all this put together?
Tom Bonier:
Oh gosh, you mentioned the word confidently and I feel like having lived through the last several election cycles, I don't feel confident about much. The person who feels confident about anything in this era probably isn't paying too close attention, but we do have cues that are beginning to paint a picture. That, to me, I do have some things that I'm confident about. I don't think that's a prognostication of an outcome, but rather a prognostication of a general context. Meaning we have these special elections that have happened in New York and Alaska. Especially New York-19 and Alaska, two big Democratic wins, and we can look at the turnout data from those, and we can see that women drove those results as well. So you're actually, again, when you talk about the Kansas results, people might add the caveat that well, but that was literally a vote on choice. How do we know those voters when choice isn't literally on the ballot will come out? Well, New York-19 and Alaska were good indicators that that's happening.
Tom Bonier:
The other data that we're looking at, and we can dig deeper on this if you want, is data on early vote. There are a handful of states where you can begin to request ballots. Georgia is one of them. And that's been in effect, I think for a little over a week now, voters in Georgia could request a mail ballot for the general election, believe it or not. And what we found there, is that women are requesting ballots at a higher rate than they did at this same point in the 2020 election, or at least a larger share of the ballot request in African American voters, especially African American women, are requesting a much higher rate. Meaning in 2020, at this point, African American voters accounted for 30% of the ballot requests in Georgia. Right now, they account for 37%. So huge differences. So there are those little hints that we can read that would suggest that, yes, this is something that's going to continue through election day.
Tom Bonier:
In terms of your question about how confident can we be, it's going to have an impact. I think the bigger question at this point is, if you take the dynamics that are really opposing each other on either side of this, on one hand, you have the fact that it's a midterm election and everything we talked about that should have set this up to be a so-called red wave election. And you have inflation, still a factor. You have the president's popularity rating still not being very high, but improving. You have those factors, and on the other hand, you have everything else. You have Dobbs leading the way, but this narrative of Republican extremism. The one thing I am confident in predicting is that the outcome, in terms of control of Congress, especially the House, is going to be very close.
Tom Bonier:
There will not be a significant wave election in one direction or another. You're not going to see either party picking up 20 seats in the house. This is going to be something where control of the house will likely be decided by just a few seats, the sort of thing where we probably won't know the outcome until a week or two or three after we all recall how long it takes to count these mail ballots in some of these states that tend to take their time. And so I think that's where we're headed at this point. We still have a decent amount of time so that the caveat stands as always, that things could change, but that's how things are looking at the moment.
David Nir:
So Tom, you mentioned early voting or the requesting of mail ballots, and I'm glad you did it. I really want to talk about this because I feel that in years past, it's the kind of data that a lot of folks can almost read into whatever it is they like and find tea leaves in that show that their side is going to benefit from those numbers. And it's even more confounded now, now that mail voting is so popular among Democrats and conversely, of course, Trump has made mail voting and early voting almost anathema among many Republicans. So when you're looking at this data on early voting and mail ballots, what is actually useful to look at, what do you look at when you are trying to make use of this data?
Tom Bonier:
Context. The most important thing is context, because every caveat you added is so important, right? If you just look at who's requesting a mail ballot at this point, you would say, wow, Democrats are doing amazing, because to your point, president Trump basically led the way and took something that used to be a fairly Republican mode of voting and mail balloting and turned it into a very Democratic way of voting by telling people that mail balloting was fraud. And then you have the impact of the pandemic and the changes in voting laws in 2020, many of which still are in place, some of which have been rolled back.
Tom Bonier:
And so in the end, we have to view all of these data points in context of historical precedent, to the extent that that historical precedent applies. So the comparison I made about Georgia, if I see that a group is outperforming their share of the electorate from 2020, and I know that Georgia was a very close election presidentially and obviously two runoff wins for Democrats in the two Senate seats, well, if we're outperforming those numbers, I'm going to feel generally good about that. Does that mean Democrats have it wrapped up? No, because it's algebra and there are two sides of that equation. We were able to fill in one of the variables in terms of Democratic early turnout, but there's still the question of democratic Election Day turnout. And as we've seen and experienced in the past, what's that Republican Election Day turnout going to be? And that's the one that we won't be able to fill in until the polls are closed and it's all done.
Tom Bonier:
And so the caveat that I always mentioned to people, we can look at these numbers, we can feel encouraged by them. We can feel hopeful if the numbers are such that they would suggest they're a reason to be hopeful, but in the end we should never feel overly confident that it's predicting some outcome because that last variable won't be filled in until election day, and if you wait until then it's too late. And so you always have to do the work and assume that the other side is going to come out in bigger numbers. So I'll be paying attention to them. I'll be sharing. I always share the early vote data. I take great interest in it, but you should always view it in the context, look, same with polling. We all consume polling, especially these days, maybe we over-consume it.
Tom Bonier:
But even the polls you need to view in context. The polls are just sort of a ‘what if’ prediction, where I mentioned that Kansas poll. The Kansas poll said that that was a one point race and then the initiative then ended up failing by 18 points. Why was it wrong? It really badly under-predicted youth turnout. It really badly under-predicted women turnout and it really badly under-predicted Latino turnout. But if you viewed that as a ‘what if,’ that poll wasn't necessarily inaccurate, it was accurate if that sort of turnout materialized. No one knows who's going to vote in this election exactly. And so these polls tell us what if. If women come out at the same rate they did in 2018, our polls will show us that. There aren't many polls that are actually predicting women coming out at a higher share than 2018, which is I think an important caveat as well.
David Beard:
Well, Tom, this has been a fascinating look at this data. Where can people find you online and learn more about what you do and about Target Smart?
Tom Bonier:
Yeah. Well, thanks for having me on to have the conversation. I've really enjoyed it. So on Twitter, I'm @tbonier, T-B-O-N-I-E-R. I try to share interesting stats usually based on the voter file, based on turnout, based on early vote. Our website targetsmart.com. We have great analysis at insights.targetsmart.com where we try to put out something that's a little bit longer form maybe once a week. And then pretty soon we'll be launching targetearly.targetsmart.com, that'll allow users to actually go and look at the early vote data and break it down by state and by party and age and gender and view it in the context of previous election. So lots of places to find us.
David Nir:
We have been talking with Tom Bonier, the CEO of Target Smart, one of the top democratic political data firms. Tom, this has been extremely illuminating. Thank you so much for joining us today.
Tom Bonier:
Thanks so much for having me.
David Beard:
That's all from us this week. Thanks to Tom Bonier for joining us. The Downballot comes out every Thursday, everywhere you listen to podcasts. You can reach out to us by emailing thedownballot@dailykos.com. If you haven't already, please subscribe to The Downballot and leave us a five star rating and review. Thanks to our producer, Cara Zelaya and editor, Tim Einenkel. We'll be back next week with a new episode.