KEY POINTS FROM THIS ARTICLE
— In Wisconsin, Susan Crawford, a Democratic-aligned candidate, won a seat on the state Supreme Court by a 10-point margin—a showing that lines up with some other recent performances by liberal judges there.
— With Crawford’s win, liberals are positioned to retain their majority on the important court for at least the next few years.
— The places in Wisconsin where turnout was relatively high voted, collectively, several points bluer than the state as a whole in the 2024 presidential election.
— A pair of House special elections in Florida did not end up being all that close, but Republican underperformances compared to last year’s presidential race mirrored what we saw in special elections during 2017 and 2018.
Wisconsin: Another 55-45 win for liberals
In Wisconsin, close to $100 million was spent in this year’s state Supreme Court race, making it the most expensive judicial race in state history—the record was held for the last two years by 2023’s contest, the most recent such race before this week’s. Despite the rising price tag, the needle hardly seemed to move: Justice-elect Susan Crawford became the fourth Democratic-aligned candidate since 2018 to win with about 55% of the vote.
Crawford’s win over former state Attorney General Brad Schimel, a conservative aligned with Republicans, will preserve the 4-3 majority that liberal justices secured with their 2023 victory. Importantly, under normal circumstances, conservatives will not have a chance to flip the court back until 2028: conservatives will be defending seats in the next two contests, set for 2026 and 2027. We say “under normal circumstances” because if Republicans win the governorship in 2026 and there is a surprise liberal vacancy, conservatives could theoretically regain the majority via a gubernatorial appointment—though that special scenario would also involve them winning both the 2026 and 2027 court races, which seems like a heavy lift given their recent difficulties in these races.
The 2026 race—and, not to spoil anything—seems like it will be another similarly heated campaign. Justice Rebecca Bradley, a conservative whose seat is up next year, did not hold back at Schimel’s campaign headquarters last night when asked about her relationship with the liberal majority on the court.
But looking backwards, something that seemed evident in early vote trends was that turnout in this year’s contest would fall between the 2023 state Supreme Court race and the 2024 presidential general election. That was indeed the case. As we enumerated last week, Janet Protasiewicz and Daniel Kelly, the two candidates in the 2023 contest, combined to take about 55% of the votes that Donald Trump and Kamala Harris totaled. Last night, Crawford and Schimel combined for just over 70% of the presidential two-candidate total.
Map 1 follows a similar format that we used for one of last week’s maps—it considers turnout in yesterday’s election as a percentage of last year’s presidential turnout. Orange counties saw rates above the statewide number (as of this morning, it stood at around 70.2%, although that figure will probably inch up as late votes trickle in), while blue counties are below the statewide number.
Map 1: 2024 presidential vs 2025 state Supreme Court Turnout
(Right-click to enlarge)
Last week, we suggested that Democrats would be helped in the race because of their “higher propensity” coalition. This was probably a factor in Crawford’s win, as orange counties would have, collectively, backed Harris last year by a nearly 53%-47% margin when she otherwise lost by a point across the entire state. Dane County led the way on turnout, too: yesterday, it cast 79.7% of its 2024 presidential vote, nearly 10 percentage points higher than the statewide figure. Madison itself cast 81% of its 2024 vote, and darker shades of orange are common throughout Dane County. The “WOW” counties around Milwaukee County also, not surprisingly, were all some shade of orange. Though Milwaukee proper was 7 percentage points below the statewide number, accounting for the blue color in the county overall, its lag was somewhat offset by some adjacent suburbs, such as Wauwatosa and Whitefish Bay, which saw higher engagement.
Map 2 uses the 2023 state Supreme Court race as a baseline and looks at the change in percentage margin.
Map 2: 2023 to 2025 state Supreme Court race “partisan” swing
(Right-click to enlarge)
Statewide, Crawford’s 10-point margin represented a slight decline from Protasiewicz’s 11 points two years ago. Crawford underperformed Protasiewicz in 60 of the 72 counties, with the strongest movements in the conservative direction coming on the peripheries of the state: counties that line the western border, bump up against Lake Superior, or border Michigan in the northeast tended to see red swings of 5 percentage points or more.
The deepest blue county on the map is actually not far from some redder ones: Menominee County, which includes an Indian Reservation, saw a 23-point shift to Crawford. However, this county can sometimes have “noisy” results because of its small population. Most of the dozen blue-swinging counties on Map 2 are centered on the Milwaukee metro area. One of the worst signs for Schimel last night was that after Waukesha County—his home county—had reported about 85% of its vote, he was carrying it by a slightly smaller margin than Kelly did in 2023. Though, as mentioned earlier, relative turnout there lagged the statewide vote by several percentage points, Crawford actually expanded on Protasiewicz’s percentage margin in Milwaukee City by almost 4 points as part of running a little ahead of her countywide.
Since 2018, victorious Democratic-aligned state Supreme Court candidates have all cracked 80% in Dane County—in 2019, when conservatives pulled off a narrow win, the liberal candidate “only” took 79% there. Crawford took 81.7% in the blue bastion last night. Though she performed a little worse than Protasiewicz countywide, Crawford performed about a point better in Madison.
According to AdImpact, the spending dynamic was most tilted in Crawford’s favor in the Milwaukee media market while the red-swinging “peripheral” counties were included in media markets where Schimel had an advantage.
So, will the high turnout coalition that’s won Democrats 4 of the last 5 state Supreme Court races be enough for them to expand their majority on the court? Check back this time next year and we’ll know.
The Florida Specials: “It is Happening Again”
A theme of the first two years of Donald Trump’s first administration was Democratic overperformance in special elections. In the oft-memed words of the Giant in Twin Peaks, “It is happening again.”
In a pair of House special elections in dark red Florida districts on Tuesday night, Republicans won, as expected, and the races did not end up being close. In the more closely-watched race, Rep.-elect Randy Fine (R) defeated Josh Weil (D) 56.7%-42.7% to hold FL-6, which includes some coastal communities like Daytona Beach, for Republicans. In the panhandle, Rep.-elect Jimmy Patronis (R) defeated Gay Valimont (D) by a similar margin, 56.9%-42.3%.
The striking thing about the results, though, is that they represented a big underperformance of Donald Trump’s 2024 showing in the districts—last November, he won FL-6 by 30 points and FL-1 by 37 points.
So despite FL-6 getting more attention pre-election, FL-1 represented the bigger underperformance for Republicans.
We mentioned that this was possible in last week’s Crystal Ball, both because FL-1 has a higher four-year college attainment than FL-6 (probably a positive indicator for Democrats in a lower-turnout environment) and, perhaps salient in a time of major government job cuts and uncertainty, a stronger federal government presence, particularly in Pensacola, the district’s population center. As it was, Valimont actually carried Pensacola’s Escambia County by about 2.5 points despite her districtwide loss: Florida Democratic consultant and analyst Matt Isbell said this is the first time a Democrat has carried the county in a federal race since then-Sen. Bill Nelson’s (D-FL) landslide reelection in 2006. Republicans also made a big deal about potentially losing FL-6 in the runup to the election, which may have activated voters in that district in a way that they weren’t in FL-1. There were about 25,000 more raw votes cast in FL-6 (about 195,500) than in FL-1 (about 171,000) last night. Those turnouts are decent for a special election—see Bloomberg’s Greg Giroux’s history of special elections for raw vote comparisons—but they do pale in comparison to presidential or even midterm-level turnout: FL-1 had 291,000 votes cast for House in 2022 and FL-6 had about 301,000, a vote total that was likely a little depressed because the Democrats did not field a candidate, leaving only a Libertarian to face then-Rep. Mike Waltz (R).
Respectively, these winning Republicans ran 16 points (FL-6) and 22 points (FL-1) behind Donald Trump’s districtwide showings in 2024. That is very reminiscent of what happened in a handful of federal special elections held in 2017 and 2018. Table 1 shows the nine races featuring two-party competition that we included, 8 House specials and a special Senate election in Alabama shockingly won by Democrats.
Table 1: Federal special elections in 2017-2018
On average in these 9 races, the Democrats ran 13 points ahead of Hillary Clinton’s 2016 performance. The results in the Florida specials last night would fit very neatly on this table.
The Downballot, the helpful, liberal elections analysis site formerly known as Daily Kos Elections, has compiled special election data for several cycles, including both federal and state legislative races.
In 2017-2018, Democrats ran on average 10.6 points ahead of Clinton’s showing in 2016. That was a smaller 4.8 points in 2020, and then Democrats ran 3.7 points on average behind Joe Biden in the 2022 cycle, when Republicans flipped the House. In 2023-2024, Democrats ran 3.5 points ahead of Biden on average.
This isn’t a perfectly predictive indicator—indeed, it suggested a better showing for Democrats than they got in 2024, for sure. The Democrats trading a larger number of lower-propensity, non-college educated voters for a smaller number of college-educated, higher propensity voters may make Democrats performing better in special or off-year elections a more consistent feature of American politics.
But it is notable that the biggest overperformances in the Trump era came in advance of 2018, also the Democrats’ best electoral showing in this era. The 2025 Downballot average so far, including the two races last night, is that Democrats are running 11.4 points ahead of Kamala Harris’s 2024 margin, very similar to the 2017-2018 number.
This is why we think it feels a lot like 2017 right now.
In a year devoid of much electoral action, this was probably the biggest election night until the Virginia and New Jersey gubernatorial elections in November. We haven’t rated the 2025-2026 gubernatorial races yet—stay tuned, we promise—but we knew that Republicans replicating their impressive 2021 showings in each of those races would likely be difficult with Donald Trump in the White House instead of Joe Biden. Nothing that’s happened in elections so far this year has shaken that belief.
One can also see a possible reason why Trump pulled his nomination of Rep. Elise Stefanik (R, NY-21) as ambassador to the United Nations. What happened in the two Florida races suggested the possibility that her Trump +21 district could have been endangered in a special election environment. As it stands now, the House will be 220-213 Republican, with a pair of Democratic vacancies in AZ-7 and TX-18, two strongly Democratic districts at Kamala Harris +22 and +40 points, respectively.
P.S. For some Florida Democrats, who have seen their party wither over the past decade, last night was more frustrating than anything else despite Democratic overperformances compared to 2024. The reason is that both Weil and Valimont raised tons of money, which some operatives rightly point out was not exactly money well spent—in the end, neither race was close. Steve Schale, a veteran Florida Democrat, wrote yesterday that “If I had to take a wild guess, the total spending in FL 1 and FL 6 was 3-4x what the Florida Democratic Party has raised for voter registration and organizing over the last decade.” Another Florida Democrat, Chris Mitchell, made a similar argument: “We’re fueling a system that prioritizes emotional appeal over strategic outcomes. And while small-dollar donors give in good faith, believing their contribution might flip a seat, the only people consistently benefiting are the consultants cashing the checks.” Anyway, just something to keep in mind for political donors on either side of the aisle. Special elections, in aggregate, may suggest something about the political environment, but one-off overperformances in individual races aren’t a substitute for longer-term party building, at least in the view of these Florida Democrats.