What did the 2024 election in Wisconsin tell us about where this state is headed politically? What did it tell us about how the political map is changing? What did it tell us about the state of polling?
To take stock of the last election cycle as we head into the next one, I discussed these and other questions with Charles Franklin, director of the Marquette Law School poll, now in its 14th year. Our conversation was conducted by email:
Gilbert: Let’s start with the big picture, Charles. Wisconsin has had three presidential races in a row decided by under a point. That’s never happened before anywhere. Wisconsin’s last two Senate races are the closest ones in more than a century. Its last two races for governor were the closest in 60 years. How long can this go on? Wisconsin won’t be 50/50 forever, right?
Franklin: We are very close right now. The five closest presidential races (by percentage margin) of the last 100 years have all come since 2000 (2000, 2004, 2016, 2020, 2024). In fact, those five are the only presidential races decided by less than a 1 percentage point margin since 1856. These last 24 years are astonishingly close. But they aren’t inevitable. Obama won by 14% and 7% margins in 2008 and 2012, and (Tammy) Baldwin won the Senate by 11% in 2018. I don’t know when we’ll have our next “easy win” for president, but it can happen with the right combination of candidates and issues.
Charles Franklin, right, speaks at the Newsroom Pub about how the Marquette University Law School poll performed during the 2024 presidential election cycle.
Gilbert: Assuming Wisconsin remains purple for a while, which party do you think has a rosier election outlook here in the next five years? Let’s take the case for Republicans first. Donald Trump made gains this time in both rural areas and many urban areas. That’s powerful. And your polling data shows something pretty eye-opening, which is that over the past decade, party identification has shifted in Wisconsin from a 6-point Democratic advantage to a 3- or 4-point Republican advantage. Sounds like a pretty favorable picture for the GOP.
Franklin: The big change came with Trump’s win in 2016. He gained throughout the state but especially in the west and north. He slipped only a little in his 2020 loss and gained back only a little in his 2024 win. At the presidential level we’ve been in a close equilibrium for three elections. How could it tip? Republicans could make more gains in the west and north, beyond what they picked up in 2016.
Compared to demographically similar counties in other states, Wisconsin’s west and north aren’t as Republican as we would expect. Future GOP gains in that region could shift the statewide balance significantly. A successful Republican win of the governorship in 2026 could also help strengthen the party leading up to 2028.
Much hinges on how Trump’s 2nd term goes. A successful term, with better approval than in the first term, could help the GOP here and elsewhere. The challenge for Republicans is that the president’s party usually loses support in the midterm election, and more so if the president is unpopular.
The shift in partisanship over the past 12 years in Wisconsin is an advantage for Republicans. We remain close, but now with a small Republican edge over Democrats. Partisanship doesn’t perfectly translate to votes — candidates and issues matter too. Since 2016, Democrats have done quite well in winning statewide races against a 1 percent GOP advantage in partisanship, and in 2024 won the Senate and lost the presidency by less than a point each, despite the GOP partisan shift. A successful Trump term could win more adherents to his party, while a poor term could drive some away.
These are potential gains for the GOP. Let’s look at Democratic possibilities.
More: Bice: The winners and losers from a topsy-turvy year in Wisconsin politics
Gilbert: So, my case for a good few years for Democrats starts with the fact — as you noted — that the party out of power in Washington usually does well in off-year and midterm elections. In this case, we have a huge state Supreme Court race in April. The candidates backed by Democrats have won three out of the last four court races, helped by strong Democratic turnout. Then there’s the race for governor in 2026. The president’s party (this time it will be the GOP) had lost eight races for governor in a row until that streak ended in 2022, so make that eight out of nine. That’s a good sign for Democrats.
Democrats have a shot at picking up the Third Congressional District in western Wisconsin next time, after coming close this time in a bad environment for their party. And because of the new maps, Democrats have at least a shot at winning either or both chambers of the Legislature. So that’s not a crazy best-case scenario for Democrats. At the same time, it’s kind of hard to square such a scenario with the party’s ongoing decline among rural and blue-collar voters here, and the fact that the state is getting a little redder in party ID. Let’s zero in for a moment on the court race, which is the next big thing. Do you see the landscape favoring either side right now?
Franklin: The recent Wisconsin Supreme Court results have been among the best for Democratic-endorsed candidates, who have won two of three recent open-seat races by over 11 points each, while the Republican-backed candidate won one open seat by less than a point in 2019. The Democratic-endorsed candidate also defeated an incumbent by some 10 points in 2020.
I think a question is whether Democrats will be demoralized by Trump’s win and fail to produce a big turnout in April. In 2019 Republicans rebounded from their loss of the governorship in 2018 and squeezed out a very tight win just six months later. So perhaps this time Democrats rebound from the 2024 presidential loss. However, the flip side is also plausible. Republicans celebrated their presidential win and continued their hold on the legislature in 2024. They certainly see control of the Supreme Court as crucial (but so do Democrats.) I’m sure both party organizations and outside groups will work very hard on the court race. I’m more uncertain if rank-and-file voters will be energized for this after 2024. Turnout in court races has been rising since 2015, setting records in 2016 and 2023. Can that continue?
Gilbert: You never want to bet against big turnouts in Wisconsin! Let’s shift gears and talk about polling, and the 2024 polls. You have been doing this now for more than a dozen years in Wisconsin. Your 2024 polling was within a point or two on both the Senate and presidential margins, and that was an improvement over the two previous Trump elections, which proved to be challenging for pollsters. I don’t want to get too technical here, and I know there is probably more analysis to do, but have you learned anything from the 2024 cycle about polling in the current political environment?
Franklin: In 2023, we switched from sampling random phone numbers to sampling from the registered voter list. That allows us to send email and text invitations, with a link to complete the survey online, and we follow up with phone calls after a couple of days. So now 80% complete the survey online and 20% by phone. This has improved the response rate and sampling from the voter list allows us to be much more precise about where in the state we are sampling registered voters to make sure we are getting the right proportions from the most Republican, most competitive and most Democratic areas of the state. This new method of contacting voters by email and text, rather than a cold call on the phone, seems to be working well.
We’ve also made some technical changes, including using past votes in our weighting of the sample. That technique seems to have increased accuracy of polling in general this year, and in the Marquette Poll. We missed the presidential race by 1.9 points, and the Senate by 1.1 points, both better than our long-term average error of 2.2 points. Our seven presidential polls sometimes had Biden ahead, sometimes Trump ahead and had Harris up by 1 point the week before the election. Trump’s win by 0.9 points was well within the margin of error of 4.4 points. In fact, this race was so close that none of our seven presidential polls in 2024 were outside the margin of error.
That said, Trump voters are still a little harder to reach. The new methods help, but suspicion of the media and of pollsters makes people less willing to participate in polls, and that remains an ongoing challenge.On the bright side, I think the polls did tell the right story all year. The economy was always the number one issue, and more people thought Trump would handle it better. Abortion was second most important, but less than half as many people said it was the most important than said that about the economy. Harris was seen as better on it. And immigration was third most important. Overall, more people thought Trump was better on whatever issue they said was most important.
Beyond specific issues, Biden’s job approval remained in the low 40% range over the past two years, a substantial burden to the Democratic candidate. Harris had an advantage on some personal characteristics, such as having the right temperament to be president, but Trump was seen as having had stronger accomplishments as president.I think that narrative on issues, the condition of the country, Biden’s performance, and Trump’s personal weaknesses, was a coherent and consistent story about the election, and I think that has held up well in the post-election reviews.
Gilbert: One more thought about the polls. Polling has obviously gotten harder. But it has never been as precise as people would like. When I went back and looked at my coverage 20 years ago of the razor-thin 2004 presidential race in Wisconsin, it reminded me that the polling here was wildly erratic back then. On the eve of the election, one major national pollster had Republican George W. Bush up by 8 in Wisconsin and another had Democrat John Kerry up by 7. Both of course were pretty far off, and in different directions. Wasn’t that supposed to be the “good old days” for polling, with much higher response rates? Broadly speaking, has polling really gotten worse since then, or just more complicated?
Franklin: Good point. In general, polling has gotten more accurate over time, not less, despite recent errors. I think the greater variation in 2004 meant people weren’t as surprised about the outcome because they had seen large errors in both directions. The complaint was more “polls are all over the place.” In 2016 and 2020 we saw errors that more consistently favored the Democrat, so the close outcomes were more unexpected. And in 2024, there were concerns the polls didn’t vary enough! Polls are never perfect but are imperfect in different ways over the years.
Gilbert: Let’s go back to the election map for a moment, because we’ve gotten very used to seeing Republicans gaining with rural voters and Democrats gaining with suburban voters. But something else happened in 2024 that bears watching: Trump made some urban gains, including with voters of color, especially Latinos. In Wisconsin those gains were very modest, and it’s easy to overstate them. Trump’s share of the vote in the city of Milwaukee went up a mere 1 point, from 19.6% to 20.8%. But you can’t dismiss the importance of marginal trends in a state this close. Trump had a net improvement of almost 7,000 votes in the city of Milwaukee (in his losing margin). He had a net decline of almost 5,000 votes in Madison. But in smaller Democratic cities, he had a net gain of around 6,000 votes. That is a net GOP gain of around 8,000 votes in these key cities, or about one-fourth of Trump’s statewide margin. So, it’s not the reason Trump won Wisconsin, but if Republicans can keep making even tiny urban inroads, that could matter in the next razor-thin race, right?
Franklin: The Trump inroads with non-white voters, especially non-white men, are an important change, even as non-white women have moved much less. The time for Democrats to take non-white support for granted is surely over. In the smaller cities, we also saw some GOP gains this year, even if the cities remained Democratic. The growing partisan divide between cities and countryside has been a staple here, but this year’s urban shifts show it isn’t necessarily going to continue.
The question I have is whether these small changes will become long-term trends, or are we simply seeing the reaction to this year’s political climate? Voters were pretty unhappy with the economy and the Biden administration, so they shifted in Trump’s direction. In 2026 and 2028, will they be happier with the direction of the country and continue to shift, or will they again be dissatisfied only now with Republicans, and shift back?
Gilbert: As we wrap up here, what if anything really surprised you about the 2024 election in Wisconsin? Any other final takeaways as we head into another election cycle?
Franklin: The split outcomes for Senate and president was rare nationally until this year, when it was a more common result. Continued Republican majorities in the legislature were also interesting, with GOP candidates running ahead of Trump. Certainly, Democrats gained seats, but we still see divided government for the next two years. The state Supreme Court race in April will be very consequential and may give us a hint of how voting shifts after 2024. It is a long while to the 2026 governor’s race, but I’ll be looking for indications of how that election is shaping up. The legislative session this spring will raise issues that point to how the state is moving.
Gilbert: So, Wisconsin continues to be a place where even amid the political realignment of the Trump era, we remain — more than anywhere else in the country — in a kind of 50/50 standoff, where the swings are mostly tiny, but where tiny swings have massive consequences. Which pretty much sets the stage for the next two crucial elections here, in April 2025 and November 2026, that will determine control of all three branches of state government.
Craig Gilbert provides Wisconsin political analysis as a fellow with Marquette University Law School’s Lubar Center for Public Policy Research and Civic Education. Prior to the fellowship, Gilbert reported on politics for 35 years at the Journal Sentinel, the last 25 in its Washington Bureau. His column continues that independent reporting tradition and goes through the established Journal Sentinel editing process.
Follow him on Twitter: @Wisvoter.
This article originally appeared on Milwaukee Journal Sentinel: Gilbert: A conversation with pollster Charles Franklin about the 2024 election
