David Beard:
Hello and welcome. I'm David Beard, contributing editor for Daily Kos Elections.
David Nir:
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. Next week, we are going to be doing a mailbag episode. We are asking you, our listeners, to send us your questions about the 2022 midterms and we will answer them. You can reach out to us by emailing thedownballot@dailykos.com, or you can find us on Twitter at DKElections.
David Beard:
Less than three weeks to go. What are we covering today?
David Nir:
Well, we have to talk about the third-quarter fundraising numbers that saw Democratic candidates for Senate and House far outweigh their Republican counterparts. There's also a follow up on the shockingly close race for the Oklahoma governorship. There are some more data that we want to share with you. We're going to tell you all about that and you can find it at dailykosdata.com. And finally, there is an insane situation unfolding in the United Kingdom, but it's not likely to lead to a new election anytime soon. We're going to tell you why.
After our weekly hits, we are going to be talking with the director of one of the best-known Democratic polling firms; that is Public Policy Polling's Tom Jensen. He has so many interesting things to share with us about what he is seeing in the polls for the midterms, and we have a great episode in store for you.
David Beard:
This week, we were able to collect the third-quarter fundraising number for federal candidates, and so we've got a lot of interesting data there. So why don't you walk us through it there, David?
David Nir:
Yeah, so for the period of time from July 1 to September 30, candidates had to file fundraising reports showing everything that they raised and spent in that time, talking specifically about candidates for the House and Senate. And the overall takeaway is that Democrats seriously kicked ass. So here are the top lines, and we'll start with the Senate.
Among all Democratic candidates on the ballot in November for every Senate race that is going before voters, those Democrats raised $193 million to $104 million for Republicans. And this excludes any self-funding. And as a result, that means Democrats had an almost two-to-one advantage. They raised about 64% of all funds. Republicans raised just 35%. That last 1% is Utah independent Evan McMullin's fundraising haul.
Now, that's actually down from the spread in 2020 during the same time period, the third quarter when Democrats raised absolutely absurd numbers and had a 71 to 29 advantage. But this is still a huge, huge split in favor of Democrats and that's why Republicans, so many of them are relying on super PACs to bail them out.
On the House side, Democrats also enjoyed an edge. It wasn't quite as large. It was $176 million to $145 million. Again, this is for every Democrat and every Republican running in every House district in the nation. And as a result, that means Democrats raised 55% of all funds from donors, again, excluding self-funding, and Republicans just 45%. That is actually an increase from 2020 during the third quarter when Democrats had a smaller 52-48 advantage. So what does this mean?
Well, as I think we all know, money is by no means the sole determinant of outcomes in politics. We saw in 2020 so many Democrats, especially running for high profile Senate races, raised massive sums and still wind up losing. But the important thing here is that it means Democrats are in the fight almost everywhere. In fact, in both the House battleground and on the Senate side in the key races, Democrats overwhelmingly have advantages in the most competitive races.
Now, I mentioned super PACs a moment ago, and we've talked about this before, Republican super PACs have widely outraised their equivalents on the Democratic side and they are absolutely flooding the zone. But it's very important to remember that campaigns, candidate campaigns under federal law are entitled to much cheaper ad rates than third party groups. So Republican super PACs, like the Congressional Leadership Fund for the House or the Senate Leadership Fund, obviously for the Senate, have to spend far more to run the same number of ads.
Here's an interesting detail. In recent week, in the month of October, in the Arizona Senate race, Republicans actually outspent Democrats. So we're talking about the campaigns of Mark Kelly and his allies, and on the Republican side, Blake Masters and his allies. So the GOP outspent Democrats by a 51 to 49 margin. However, because most of the Democratic spending came from Mark Kelly, who is just a machine of a fundraiser, Democrats actually aired something like three times as many ads as Republicans managed to air for essentially spending the same amount of money. Now, again, aired the most ads is certainly not a guarantee of winning by any stretch of the imagination, but this means that Democrats have the money and the resources to get their message out in the final weeks of the campaign and Republicans are simply playing catch up and again, relying on super PACs to come in and save the day. And you would always, always rather be the side that has better funded candidates than better funded super PACs.
David Beard:
And I know we're all worried obviously about the outcome of the 2022 election, but all we can control is what we can do. That's give money, that's help out on campaigns, volunteer, et cetera. So I think it's great that these Democratic candidates have raised this money, that they're getting their message out there, and then we'll do everything we can and ultimately the chips will fall where they will.
David Nir:
And I should add that if you're interested in getting directly involved in key races in ways that don't involve giving money, please go to dailykos.com/gotv. That stands for Get Out the Vote. dailykos.com/gotv, and you can find something either in your area or virtually that will be right up your alley and it gives you a great opportunity to work directly with campaigns and the organizations that are supporting them.
David Beard:
Now, one race that we talked about last week that I just have to do another quick update on is the Oklahoma Governor's race. Now, as we talked about incumbent Oklahoma Republican governor Kevin Stitt is facing a tough challenge from superintendent of public instruction, Joy Hofmeister. Two big updates. First, the Republican Governors Association has started a million-dollar ad buy going after Hofmeister, connecting her to Biden and accusing them of taxing the oil and gas industry, pretty much what you would expect from them. But the news obviously is that the RGA is spending this money in Oklahoma, which they would not do if they didn't think that there was a serious problem here. One of the best indicators of races and whether or not they are competitive is whether or not these party groups from D.C. are getting involved because they get a lot of information, they have a broad view of the whole country, and if they think that they need to spend a million dollars in Oklahoma to save Kevin Stitt, then clearly there's a serious problem there.
The other big update is that two more polls have come out both by GOP affiliated pollsters showing Hofmeister ahead. One of them only had her ahead by one point, but the other poll had her ahead by seven points. And one thing that we've seen in deep blue and in deep red states is that the dominant party tends to overperform their polls at the end of the day. We've seen that in a very red state like Oklahoma, but then also a very blue state like California. So a small lead might not be enough for Hofmeister, even if it's real that she's up by a point or two. That might be overcome by Oklahoma's very dark red lean. But if she's up by something like seven, like that one poll showed, there's a real possibility that that could hold up on election day. And so if more polls come out showing her with sort of a lead in the mid to high single digits, you really start to believe it might happen.
Now still, we've got a ways to go here and RGA spending a lot of money here is not going to help her case, but it's definitely something to keep tracking.
David Nir:
So it's been a very data-heavy week for us at Daily Kos Elections. Of course, we were just talking about the third quarter fundraising numbers, and you can find our spreadsheets that catalog all of those numbers for both the House and the Senate at dailykosdata.com. And there, you will also find two other resources that we just published. One is our comprehensive candidate guide for the 2022 election. We list every single candidate in every House race, every Senate race, every governor's race, along with demographic information on each candidate. So race, religion, gender, whether they're LGBTQ, and more. We also include some demographic and political information for each district and each state.
This is a fantastic resource. You can look up absolutely any race. You can look up your own district, you can look up other districts. Again, you'll find that at dailykosdata.com. And the other resource that we released this week is called the House Vulnerability Index. It is designed to tell us which seats from each party are most likely to change hands in the November midterms. You can find out much, much more about it at dailykosdata.com. There is a detailed explainer from contributing editor David Jarman, as well as a link to the spreadsheet.
David Beard:
Now, lastly, I want to talk about something a little strange, which is not an election, and I want to talk about the United Kingdom and why there's probably not going to be an election anytime soon despite the absolute crazy series of events that seem to keep happening that you may have seen on Twitter or are in the news. But first, I have to give a little background. Boris Johnson was deposed by the Conservative Party earlier this year, which resulted in an election of a new prime minister by Conservative Party members; that was Liz Truss.
She was officially named prime minister on September 6th, and then two days later the Queen died, of course, and sort of that obviously consumed the first couple weeks of her premiership, dealing with that and the mourning period and everything like that. So the first big real event of her premiership was the mini-budget that rolled out on September 23rd, and it had massive tax cuts including a lot of tax cuts specifically for the wealthy. It didn't have any spending cuts; it didn't have any explanation for how it would be paid for. There are the sort of fiscal documents that were going to be rolled out, but they got delayed and are going to be rolled out later in the year. As a result, the financial markets freaked out. They went after UK bonds and other aspects of the U.K. financial stability that the government relies on to keep borrowing low.
As a result, the Tory backbenchers pressured Truss and her government into pulling back a lot of this economic plan that she had relied her entire economic structure on. This was her big project to enact all of these tax cuts for the wealthy and for really everyone. And as a result, create all this economic growth sort of in this Reaganite style. But the U.K. is not in a position to afford any of that, and the markets forced her to stop—a U-turn on almost all of it. And as a result, borrowing costs are a lot higher, the mortgage rates for a lot of people in the U.K. that don't use as many fixed-rate mortgages as we do here in America, they're going to go up in the next few years. And there's a lot of concern about the financial condition of the United Kingdom. And Truss has ended up getting some of the worst poll numbers I've ever seen with disapproval going into the 70s, approvals sometimes under 10%. The Labor Party is up 20, 30, 35 points on the Conservatives in polling. And the entire Conservative Party is just in a mass panic about the whole thing.
Just this week, there was supposed to be a vote to allow fracking, which had previously been banned, that Truss wanted to allow in local circumstances. But a lot of Tories are opposed to fracking. And the Truss government said that this was going to be a confidence vote. They said that you had to vote for this or you were going to be kicked out of the party.
But then the day of, they sent junior minister to walk that back and say, "Actually, it wasn't a confidence vote." But then after the vote, they said that the junior minister had actually informed people that it was not a confidence vote in error, and it was a confidence vote.
But then the prime minister, because she ran after the chief whip who ran out of Parliament during the vote, didn't vote in a confidence vote without an excuse. Technically, if it was a confidence vote, Truss should have been kicked out of her own party, which is just an absolutely insane series of events. No one really knows what's going on with the chief whip. There's rumors that they resigned on the floor and then just walked out and Truss ran right after them. The whole party is in this disaster area.
And one of the things I think a lot of people are asking is that is this going to lead to early elections? Because as we've seen in other countries with the parliamentary system like Italy or Israel, that often when a government collapses into infighting and everybody's at each other's throats, an election soon follows.
That's probably not going to happen in the U.K. because the Conservatives have a majority on their own. It's not a situation where there are multiple parties or one party might withdraw from government because they think they have an advantage, the Conservatives are the majority party in the Parliament in the United Kingdom; they don't have to answer to anybody else. And so as a result, as long as they all stay united on the idea that they don't want an election, there's not going to be an election any time soon. And because the Conservative party poll numbers are so incredibly bad, one of the few things they are united on is the fact that they don't want to have an election anytime soon.
It's one of the poor ways that parliamentary politics can sometimes work out in that a majority government that was elected and has become extremely unpopular can just ride it out right up until an election is forced to be called, which is often up to five years after the last election, and can just wait and wait and wait it out, and then you have this sort of lame duck period. Whereas if things are really popular, you'll sometimes see a majority government call early elections to try to take advantage.
But long story short, the Conservative party will probably remain extremely unpopular in the U.K. for the next two years but stay in power because the alternate is to bring forward a landslide loss earlier than the absolute latest possible date. That's probably what's in store for the U.K. Crazy things can happen, you never know, but that's what I would expect to see going forward.
David Nir:
Well, we are going to swing back to this side of the pond and talk about United States elections with our guest coming up on this show. After the break, we're talking with public policy polling's Tom Jensen about what he's seeing in the 2022 midterms. Please stay with us.
I am so pleased to introduce our guest this week. He is not only one of the best known pollsters in the democratic world, he's also someone I have known in politics for an extremely long time and is just an absolutely delightful human being. Tom Jensen, Director of Public Policy Polling, thank you so much for joining us today.
Tom Jensen:
It's really great to be with you.
David Beard:
You may have known Tom for a long time, but I bet you that I've known him longer because Tom and I both went to UNC. We were both there at the same time. And Tom was actually one of the first campaigns I worked on when he ran for student body president way back in... I don't even remember what year it is. We don't need to get into that. But we've been-
David Nir:
How'd that one go?
David Beard:
Not the best campaign Tom's ever worked on, I would guess.
Tom Jensen:
Well, it was actually very illuminating for being a pollster because I ran a very good insider campaign. I got endorsed by the Young Democrats and the College Republicans, even though there was a Republican candidate. And I had gotten endorsed by the Black student movement even though there was a Black candidate. Had most of the major groups on campus locked up.
And then I got 18% of the vote in a four-candidate race. And what that showed was that your insiders who were super paying a lot of attention are not the same as your normal voters. And I actually kept on going back to this in 2011 when the media was totally obsessed with Jon Huntsman. I was like, Jon Huntsman is the Tom Jensen candidate in the Republican primary for president. The elites may love him, but he's not going to come across well with the masses. And so that was the lesson I learned from my own experience as a terrible candidate.
David Nir:
I was thinking like Jeb Bush, did you have the super PAC blow $100 million on you only to lose? But that's totally, totally hilarious. Tom, from UNC then, tell us about how you got into the polling field.
Tom Jensen:
PPP, when I was 24, was still a pretty small organization. This job opened up, and it was really something that was mostly just focused on polling on local races across North Carolina. I did not have any sort of formal training as a pollster going into that, but I did have a pretty good base of political knowledge from having been very involved in North Carolina Democratic politics as a student and my first year out of college, and also a lot of comfort with statistics, which is obviously pretty important for polling. I'm very grateful to my boss, Dean Debnam, for giving me an opportunity to run a company that, at that point, I was not qualified to run and then give me the freedom to learn how to do it on the job. I started in November of 2007 and we got right into it with the presidential primary shortly thereafter.
David Beard:
And then over these past 15 years, PPP has grown a lot, so I'll give you a second to brag on yourself and how the organization and the company has grown over the years.
Tom Jensen:
I'll tell you our origin story briefly. Believe it or not, in 1999, both Raleigh and Durham, North Carolina, elected Republican mayors. And that's the last time either of those cities elected a Republican mayor. But certainly, it's a good example of how much the urban politics dynamics have changed over the last couple decades that it was ever possible to elect a Republican there in those cities.
But our founder was very involved in the 1999 Raleigh mayoral race. We lost 51 to 49. And our candidate was able to afford one poll for that race, and she did it sometime over the summer. And she really couldn't get any new information in the weeks leading up to the election about where things were trending and what issues were rising to the top and that kind of thing, so PPP was created to allow campaigns and organizations on smaller budgets to be able to afford to do polling. And we've basically grown from providing that service in North Carolina to now basically providing that service for Democrats all over the country.
Just today, we have polls going in Idaho, West Virginia, Texas, New York, Washington state. That's just a selection of what we have going today. We've been able to take that approach globally to helping people with lower-cost polling all over.
David Nir:
Now, Tom, you really leveraged Twitter, I think, to blow up PPP's reputation. And for many years, you would solicit all kinds of topics or states or candidates or questions, and then you would also poll on these really funny but topical and smart questions that really got folks buzzing. When was it that you figured out this almost one weird trick on how to become such a prominent pollster? Because it really seemed to work so brilliantly for you, and it was also super, super fun.
Tom Jensen:
Yeah, remember when Twitter used to be fun? Life was so much better then. It's not a coincidence that PPP stopped being fun the second that Trump took office, because that was really also when online just got so toxic that there stopped being much appeal in putting our work out there.
But I think the big story with the more offbeat stuff we did is that Barack Obama made the Republican party go crazy from 2008 to 2016. And basically, what we were doing during that period in a lot of our public polling was asking questions that might have seemed outlandish, but I think that really showed the direction that the Republican party was headed in and the extremism that was taking over.
One thing I really vividly remember is in 2011 doing a poll that found that Republican primary voters in Mississippi still thought interracial marriage should be illegal. That's a settled question. Why would you even poll on interracial marriage? But what we found was that, with the direction that the Republican party was going with its extremism in response to being so unhappy that President Obama was in office, a lot of those crazy views were taking hold.
And I actually give Daily Kos a lot of credit for being the first polling organization to poll on birtherism when that was something that was just chuckled at. "Oh, these people say that Obama wasn't born in the United States, so that's so ridiculous." You all polled on it and found that a lot of people really did think that. And I think that's ultimately the issue that propelled Donald Trump into the presidency. If you go back to the very start of the Trump movement. We were asking questions that were ridiculous, but they were really telling a story about the direction that the Republican party was headed in that I think culminated in Trump's election in 2016, and obviously has just continued on over the last six years with the party getting more and more extreme.
David Nir:
Tom, I remember you once sent me a blooper reel of you trying to record the poll questionnaire asking is Barack Obama the anti-Christ? And you just kept cracking up. You could not keep a straight face. And it took you 10 tries to be able to record a good take on that one.
Tom Jensen:
Yeah. And I bet I could record that now without laughing at all because the stuff that just seemed so ridiculous 10 years ago or 12 years ago has gotten way too close to reality in a lot of cases.
David Nir:
Well, since this conversation has taken a turn for the serious, we will stick with that mood. And Tom, I really want to dig into some of the mechanics of modern polling. Obviously, folks have felt for quite some years now that polls are broken or not as accurate as they used to be. But let's just back way up first. And I want to start with some basics. If you went into the field and you got 1,000 responses to a particular poll, what would those results look like just as is without trying to weight a poll? And we'll get into what weighting exactly is and what it means in a second, but how would that compare to what you actually think that America or the electorate would look like?
Tom Jensen:
You definitely would end up with too many women in a totally unweighted poll. You'd end up with too many white people and not nearly enough people of color. You'd end up with too many old people and not enough young people. You'd end up with too many highly educated people and not enough less well educated people. And usually, you would end up with too many rural people and not enough urban and suburban people.
And let me tell you how that plays out in two specific states that I think are pretty illuminating about some of the things that have happened with polling in recent years. In a state like Iowa, the fact that you end up with too many white people doesn't matter because over 90% of the population is white. But the fact that you end up with too many educated white people does matter because what that means is that your raw data in Iowa almost always ends up being way too Democratic. It was really true in 2020. You have to try to weight to correct for that. But polls have generally overestimated Democrats in Iowa over the last decade or so because of that dynamic where you get too many well-educated white people.
So now let's look at the flip of that. In the state of Nevada, you have one of the most diverse electorates in the country. I think you could pretty firmly say the most diverse electorate in a swing state. Well, there you end up with way too many conservative white people who live out in the 2nd congressional district and in the non-Vegas part of the state. And you don't end up with nearly enough Latinos, Asians, Black voters, that sort of thing.
Nevada's been a state over the last decade or so where polls have tended to really underestimate Democrats. You even quite close to the election in 2018 had Jackie Rosen trailing in the bulk of the polls. So that was sort of the flip of the situation where there, the issue with the racial imbalance makes it so that polls have traditionally underestimated Democrats. So it's different everywhere, the impact that weighted versus unweighted data has, but that just shows you in a couple states how it plays out.
David Nir:
I think those examples are extremely illuminating. And to me, the next obvious question then is how do you decide how to weight? How do you decide as a pollster, what do you think the electorate would actually look like if your poll yields too many responses from women say, how do you determine what percentage of the electorate do you think is actually going to be women?
Tom Jensen:
Yeah. So first of all, I'll tell you what we're weighting in most of our polls, and the big thing that's sort of a historical change there that I think a lot of pollsters have gotten more comfortable with, or maybe even just come to see as totally necessary, is we ask people on every poll who they voted for in the last presidential election. And that's something that we've always asked on polls, but we used to just sort of use that more for advisory purposes. And we now pretty definitely weight for that variable on every poll to make sure that we have a balance of Trump and Biden voters within a particular district that's at least within the ballpark of where it was the last time around. That ultimately, I think, is the biggest thing you can do to ensure that you have a somewhat balanced sample.
Of course, where that could go wrong is let's say in a state like Arizona that the people who come out to vote this year, even though Biden won by a fraction of a point, let's say that the people who come out to vote this year voted for Trump by five. Then obviously the polls are going to underestimate Republicans if that's something that ends up happening.
I don't think that's going to happen to that sort of extent this year, especially with the way that the electorate has resorted along educational lines. I think since most well-educated voters who are more likely to turn out for non-presidential elections or Democrats now that you're not going to see the sort of cataclysmic turnout dropoff that you saw for Democrats in 2010 and 2014, that's something to be aware of. So 2020 vote's a big thing that we weight for. We weight for gender, we weight for race, we weight for age, we weight for education. We weight for whether we're getting people on a landline or a cell phone.
And then in states with party registration, we sometimes weight for party as well. And all of that stuff is informed by a combination of things, Census data and the particular state historical voter turnout data in the particular state, that sort of thing, or how things are shifting from one election to another in the electorate in the state, for instance.
In a state like West Virginia where people are rapidly leaving the Democratic party, we're not going to weight to the same party breakdown that turned out in 2018 because that party breakdown doesn't exist anymore as people have sort of moved away from the Democratic party. So that's sort of a big picture look at it.
David Nir:
And in that raw data that you collect, does that inform how you might weight anything? For instance, if all of a sudden you are getting back polls with many more right-wing voters, does that tell you, "Oh, maybe there's actually a surge in enthusiasm among conservatives or Republicans and therefore, maybe we should weight this poll a little bit more heavily to the right?"
Tom Jensen:
Yeah, so historically the answer to that would definitely be yes. But if you did that in 2020, you'd really get burned because the raw data was way too Democratic in 2020. And I remember sitting around at this time two years ago, hoping that meant something about turnout. But what it really meant in that particular case was that Democrats were still at home concerned with COVID, and Republicans had gone back to living their normal lives, and we were getting a really imbalanced raw data because of that. That definitely was not an indication of Democrats being more excited because it actually turned out at the end of the day, I think globally, that Republicans turned out a little bit more than Democrats and there were enough Republicans voting Democratic that we still won the election in the big picture of it. So we're a little cautious about that.
What I would expect this year based on a lot of data, not just what we're seeing in our polls, but also what's happened in special elections across the country — which you all do such a good job of tracking for all the rest of us; thank you Daily Kos Elections — I think the electorate's probably going to be one to two points of where it was in 2020.
So let's say that obviously, nationally, the people who voted in 2020 voted for Biden by four. I'd expect that who turns out this year with Republicans having a little bit more motivation with the Democrats in the White House, which is historically kind of how it's always gone. I think you'll see a Biden-plus-two, Biden-plus-three electorate, something like that, but it's nothing like in 2010 where, I don't know if there's absolutely official numbers about it, but I think he basically went from an Obama-plus-seven electorate in 2008 to something that was probably about evenly divided between Obama and McCain voters in 2010. Democratic turnout just fell off the ledge. I don't think that's going to happen to that extent again this year.
David Nir:
One last question on this topic. A few weeks ago we had the Economist’s G. Elliott Morris on our program, and he is a polling and data analyst, and he said that in his opinion, and I think the opinion of a lot of others who follow this issue, that one chief difficulty pollsters have had in recent years is reaching a sufficient number of what I believe he called ‘very Trumpy’ voters.
And I'm curious to know if you agree whether that's a problem, and if so, can the answer just be as simple as weighting those very Trumpy voters a little bit more heavily to compensate for the fact that they don't seem to want to answer polls?
Tom Jensen:
Well, that is a very interesting question. So basically when you look at our polling over the last five years from 2017 to 2019, and in 2021, we had probably about the most accurate polling years we've ever had in the history of our company.
I remember on election night in 2018, we had polled something like 400 races in the month of October and just going through the results over the end of the week and it's like, got within one there, got within two there, and it was rarely more than three or four off the pace. That was one election year I was like, "Man, I wish our clients would've released all our polls because that would've made us look really good." I mean, you can attest to that, Nir, because I was telling you most of what we were seeing sort of behind the scenes.
David Nir:
Oh yeah, no, you hit it on the screws, Tom.
Tom Jensen:
So four out of the last five years, our polling was great. 2020, our polling, like everyone else's, was a disaster. So I think, basically, the answer to your question about whether weighting can sort of solve this problem of not enough Trump people answering the phone, I think it definitely can when Trump himself is not on the ballot because we've been very good in all the Trump years except when he was on the ballot.
But if he is the candidate again in 2024, there's definitely going to be some hard choices to figure out how to sort of account for that. But one thing I will say about the polling this year compared to 2020 is our raw data makes a lot more sense in 2022 than it did in 2020 when Democrats were so disproportionately answering the polls.
So I don't think there is as much or possibly even any problem with missing out on the Trumpy people this time around. I think it'll be a lot different for a midterm than it is when Trump is actually on the ballot.
David Nir:
So you mentioned 2022, so let's turn there. You've been doing obviously a ton of polls for clients all across the country, as you mentioned. Obviously, you can't just tell us publicly what all those polls are saying as much as we wish that you would. But give us sort of a broad overview of what you've been seeing and how it compares to what's sort of out there publicly with the 538 tracker and all of those things. Does that pretty much match up? Are you seeing things better or worse?
Tom Jensen:
So that's a really good question because the answer to it is that what we're finding in our polls for the most part isn't that different from the public polls, but the public polls often don't give a whole lot of context beyond the toplines. And the context beyond the toplines speaks to a reality where Republicans may very well outperform the polls right now by three, four points in a lot of cases. And it actually doesn't necessarily mean that the polls right now are wrong. So I'll explain what I mean by that.
Basically, what we've been seeing in our polling all year is that if you approve of the job Joe Biden's doing, you are already in the Democratic column and you've been in the Democratic column all along. And all year when we've sort of dug in on the undecideds in our polling, which is something that we do at pretty good depth for all of our clients, they've been very Republican-leaning throughout the year.
On the average poll that we do, we usually will find about 20% of undecided voters approve of the job Biden's doing, 60 or 70% of voters disapprove the job that Biden's doing.
So over the summer when we would do a poll and say, find a Democrat up 48 to 41 or something like that, but see that two thirds of the undecideds were Republican ... I mean, were people who were unhappy with Biden, we'd be telling our clients like, you're up 48, 41 right now, but we sort of see the number that you can grow to as 51 looking at these people who are on the fence and we see the number they can grow to as being more like 49. So even though you're up by eight in this poll, we think that this is more likely to end up being something like a two or three point win, that sort of thing. Because we've just been seen throughout the year that almost everybody who's on the fence is somebody who's unhappy with how things are going right now, and people who are unhappy with how things are going right now generally end up voting Republican.
And another piece of this is that in the statewide races where Republicans have weak high-profile candidates, we saw a lot over the summer that Republican leaning voters were sort of hesitating to support those people. Obviously there's been a lot of attention in recent weeks about how much Pennsylvania's tightening; how much Nevada has moved more into Republican direction; how Georgia got a lot closer, at least before Herschel Walker's latest set of troubles. That was all stuff that we could see in our polling was going to happen in July or August, because basically I think the main thing that happened over the last three months in those races is those undecided voters who didn't like Biden finally just decided, "Okay, I'm going to vote for Republicans because I don't like Biden." They moved into the Republican column.
So that's sort of where there's a possibility that races right now where Democrats are up by two could turn into races that Republicans could win by two, and the polls weren't even wrong, but that's just the direction that people headed in because there's a Democratic president and they're unhappy.
David Nir:
And we do have one example we can talk about specifically, which is the Illinois governor's race. You recently did a poll for the Chicago Sun-Times and it was released publicly. So tell us how that poll came out and how your results reflect what you've just been talking about.
Tom Jensen:
Yeah, so we found that J.B. Pritzker has a 15-point lead in the Illinois governor's race, but one big part of that was that the Libertarian candidate was polling at 7%, 8%. And this is something that we've seen in a lot of these races with weak Republican nominees, is that part of what Republicans who don't like their candidates are doing right now is saying, "Okay, well I'll vote for the Libertarian."
Well, we know that at the end of the day, 7%, 8% of people don't vote for the libertarian. That's much more likely to end up at 2% or 3%. Something I've been telling clients a lot, especially in regards to these races like Illinois, where Darren Bailey is obviously such a poor candidate on the Republican side, but I think one huge lesson we've learned over this last decade in politics is that you can trust Republicans to do the wrong thing in the end when it comes to Republican voters. So that 7 or 8% that we might see right now is saying they're going to vote Libertarian for governor of Illinois.
Tom Jensen:
And we saw something similar for governor in New Mexico on a public poll that we released lately. And some of the key Senate races, you're seeing unusual levels of Libertarian support right now. I think you can pretty well assume that the Libertarians are going to end up at about half of where they are right now. And let's say that a Libertarian goes from 8% to 4%, that's going to move three to one in the Republican's direction almost every time. So there's two points off your poll margin for Pritzker right there.
And then the other part of that specific poll was that when we looked at the people who were undecided, it was the same thing I'm talking about with Biden, except in this case it's with Pritzker. We found that if everybody who was undecided just voted for Pritzker if they approved of him, or voted for the Republican if they disapproved of Pritzker, that would cut four points off of Pritzker's margin because the remaining undecideds were so heavily unhappy with Pritzker. So between the Libertarian not really getting 7 or 8%, and the fact that all the voters on the fence were unhappy with Pritzker, your 15 point lead is all of a sudden a 9-point lead.
Now I know that if Pritzker really wins by nine, which I think is more likely than him winning by 15, that I can look forward to a lot of people tweeting at us about how we're Democratic biased idiots because we overestimated Pritzker's win by six points. But I'm telling you right now that there's a very good chance that we overestimated Pritzker's win by six points just based on sort the innards of the poll. And that's context that gets lost when people just don't look at anything beyond the toplines. Because again, this is something that we're seeing on almost every poll, that most of the voters on the fence are people who are likely to end up just biting the bullet, even if they're not thrilled with their Republican candidates, and voting Republican in the end.
David Beard:
So if we take the Biden approval/disapproval as sort of a universal baseline, though obviously, like you said, in an incumbent governor's race, it's a little different. But if Biden's in the low forties as he's sort of been in the most recent polls, and you have Democratic candidates who are maybe pretty strong — let's say Arizona Senate as an example, you have a really strong incumbent Democrat; you have a really bad Republican challenger. And the goal there obviously is you want the incumbent Democrat to outrun Biden's approval number by people who are like, "Well, I don't like Biden, but Mark Kelly's done a pretty good job. Blake Masters seems crazy. I guess I'll vote for Mark Kelly anyway, despite my unhappiness with Biden." Have you seen races where that's been working? How many points can you get off of something like that? Usually you can only run so far ahead of that sort of baseline approval, but is there a difference in when you have that quality showing that you can do it in certain cases?
Tom Jensen:
Yes, thank you for asking that because something I want to make really clear about is, like, I don't want to be here saying Democrats are doomed. I actually don't think that's the case at all. There are so many really close races across the country for Congress and state legislative races, and all that sort of thing. I think there was this thought that with the new redistricting cycle that there really weren't going to be very many competitive districts. But there are a lot of competitive districts, and there's a lot of competitive races in places where you wouldn't necessarily expect them to be.
For instance, we did a poll a couple months ago for Wiley Nickel in North Carolina's 13th Congressional District, which is likely to be one of the most competitive in the country. And that's a district that Biden won by one and a half points. So you would think that in a really bad political climate for Democrats, somewhere like there, you'd end up losing by four or five points, and it would sort just be off the map. And he's somebody who's keeping the race tied. It definitely helps that his opponent, Bo Hines, is really weak. But that's just sort of one example of how there's a lot of places where Democrats are kind of holding their own.
There's also places where Biden won by nine or ten points that are now toss up races. But that's not all moving in a direct line where you might think that, oh, if there's districts that Biden won by nine or ten, where Democrats are having a hard time, then it must be that Democrats are going to lose all the races in districts that Biden won by less than five. Not the case at all. We have people running really good campaigns, focusing on how extreme Republicans are. And when I talk about all these voters who are undecided who dislike Biden, the reason they're not already voting Republican when they dislike Biden is because, in a lot of cases, they like the Democratic candidate better than the Republican candidate. And we are running really good campaigns, I think, across the country, focusing on just how extreme the Republican party has gotten, and do we really want more of that?
Just to use another North Carolina example about sort of the candidate quality. I mean you would certainly think that in the US Senate race in North Carolina, Democrats wouldn't have a chance given the fact that the last time there was a new Democratic president in a US Senate race in a midterm, Richard Burr won by 12 points in 2010. And here you have Cheri Beasley in a toss up race against Ted Budd. And the reason for that is Cheri Beasley. She has a net favorability rating that's 15 points better than Ted Budd's, and that's keeping this race competitive.
So I'll be upfront that I'm going to mention a couple clients here, but I really hope that people will think about, over these last few weeks of the campaign, sending House Majority PAC some money, sending Senate majority PAC some money, sending the DLCC some money, because we are holding our own in a lot of these races, and it's not inevitable that they're going to move against us at the end, but we really need the resources to make sure that we're getting the word out, especially about how bad these Republican candidates are, to sort keep that everything falling apart at the end from becoming inevitable.
David Beard:
So Tom, that's a perfect segue into the last topic that we'd like to discuss with you, and that is the messages that both sides are using. No doubt when you are polling for your clients, one of the things that you are looking at is which topics, which messages, which lines of attack are going to be most effective, and how. And of course the area that we have seen Democrats across the country, in so many different states and districts, focus on this year is abortion. And what I'm particularly curious to know from you is, has anything unusual or interesting, or notable, popped out when you have asked questions about this issue? And does it vary from district to district or state to state?
Tom Jensen:
Well, it's really remarkable how much abortion moves the needle. I mean, I think Democratic campaigns are absolutely right to be making this a primary focus of the campaign this year. We did a poll for Pat Ryan the week after the Supreme Court decision came down for his special election against Marc Molinaro. And we did our initial horse race, and Molinaro was up by three. And then we did a message testing question, just one message testing question about abortion, contrasting what the two candidate stances were. And Ryan went from being down by three to I think being up by eight, just based on contrasting the two candidates on abortion.
And I have to tell you something that's really been a reality in our polling over the last five or six years is that as voters have gotten less and less open to changing their minds about anything, we've done so many polls in the last few cycles where we might test like six negative messages on a Republican candidate, and then when we do the re-ask at the end asking people who they support now, there's absolutely no movement at all. Like you were down 43 to 47, and then we said everything bad there was to say about the Republican, and you were still down 43 to 47. And that wasn't even necessarily a case that the messages that the Democrats were trying to run with were bad messages. People just were not open to moving their minds because they're so entrenched in the Democratic column, are so entrenched in the Republican column.
And I can't tell you how many polls we've done over the last three or four months where abortion moves the needle by seven or eight points, which is just much more than we've seen for just about anything in recent years. If this campaign could be run in a vacuum about abortion, Democrats would be looking at something like a 54 to 46 Senate and an increased House majority. Of course, the reality is that there are more issues than just abortion on the table. But I think we saw, especially in all those special House elections over the course of August across the country, where we outperformed how Biden had done in all of those districts, I think that really shows the extent to, which especially in the immediate weeks after the Supreme Court decision, that was just such a salient and effective issue for us.
It's too bad that we couldn't have just gone ahead and had the whole midterm election the day that Pat Ryan won that special election because I think that we've had a hard time keeping abortion quite as top-of-mind relative to other things over the last few months as we were over the summer, but it's definitely been a remarkable game changer. And I'll be honest, very honest, that for most of the time that I've been doing this job, I did not think that abortion was as much of a winning issue for us as it really turned out to be.
David Beard:
Well, I think one thing is that the Dobbs decision has really changed that. I think there were a lot of instances, where because the status quo protected abortion, the issue actually didn't help because our side wasn't motivated to do something because they know they had Roe as a backstop, and the other side was always motivated by the idea of overturning Roe, or trying to cut away at it any way that they could. And now because the things are flipped, the other side sort of feels good about themselves, which is sort of a terrible weird thing to say. But we are the ones that then have to claw any way that we can to try to restore abortion rights. So it's so much more of a motivating factor now.
Tom Jensen:
Yeah, absolutely. And the big impact that this has had politically is giving our people a reason to be excited to go out and vote this fall in a way that did not really happen during the two Obama midterms. I mean, I think in both 2010 and 2014, of course you had people voting Republican who had voted Democratic in the presidential year. But I think our biggest problem in both of those years was the bottom kind of just dropping out on our turnout because we did not have a super-animating issue in the way that the Republicans’ hatred of President Obama was a super animating issue for them.
And I think the biggest political impact that abortions had for this political cycle is, of course, Republicans are going to turn out to vote in this election because they hate President Biden. But the Democratic base has a reason to really turn out and vote too, because I think we're seeing now that all the rhetoric about what Republican control could cause to happen to our rights is now, as you know, it's not speculative, it's real, and the Dobbs decision could just be the tip of the iceberg. I think that gives our people a motivation to go vote in a way that didn't happen in the last few Democratic presidential midterms.
David Beard:
Well, we have been talking with Tom Jensen, the Director of the Democratic polling firm, Public Policy Polling. He's been sharing his insights on how the polling industry works, how his firm operates, and what he has been seeing for 2022. It has been really, really fascinating. Tom, thank you again so much for joining us.
Tom Jensen:
Thank you.
David Beard:
That's all from us this week. Thanks to Tom Jensen for joining us. The Down Ballot comes out every Thursday, everywhere you listen to podcasts. You can reach out to us by emailing thedownballot@dailycoast.com, and particularly this week, you should email us and ask listener questions. Next week, we'll be going through listener questions from you about the 2022 election, so you can email us, like I said, at thedownballot@dailycoast.com. You can also find us on Twitter @DKElections.
If you haven't already, please subscribe to The Down Ballot on Apple Podcasts, 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.
