Less than one week after the Wisconsin Republican Party forced the voters of their state to choose between exercising their democratic rights and protecting their health by fighting common-sense efforts to expand mail voting and delay in-person voting, the results came in and the Wisconsin GOP drowned in a tidal wave of karma.
Their preferred candidate, incumbent Supreme Court Justice Dan Kelly, not only was defeated by progressive Judge Jill Karofsky—he was thumped. When all the precincts reported, the margin of victory was 11 points. As any election junkie will tell you, double-digit wins in Wisconsin are, these days, decidedly rare.
So, is this a sign for November?
At the risk of being overly equivocal, I will admit that there are cautionary notes that make victory laps with an eye on November premature. That said, the results from this election should, at long last, lay to rest the absurdly stubborn conventional wisdom that Democrats would do well to abandon Wisconsin and focus on more “winnable” states in the Sun Belt.
There is no reason not to target the Sun Belt, of course, but Wisconsin, like any number of Trump 2016 states, is eminently winnable for the Democrats, and there were plenty of data points in this week’s results to illustrate that fact.
Indeed, while Republicans may try to convince themselves that what was visited upon them was the result of a Democratic-heavy presidential primary turnout, the evidence suggests that other factors—most notably an ongoing and consistent shift in suburban voting patterns—are at play.
In the eyes of some conservatives, the die was cast for their candidate when they realized that the 2020 Supreme Court election would coincide with the presidential primaries. After all, with an uncompetitive Republican presidential primary on deck, it was reasonable to assume that the bulk of the voter energy would be on the Democratic side.
Democrats, undoubtedly, could counter with the fact that much of the intrigue of the presidential primary was drained by a combination of former vice president Joe Biden’s considerable lead post-Super Tuesday and the COVID-19 crisis. And, of course, they would be correct.
That said, there is still some evidence that there was a clear enthusiasm gap that possibly bled over into the Supreme Court case. As pollster John Couvillon noted after Monday night, there was a marked difference in the partisan breakdown of the 2016 and 2020 primary electorates in Wisconsin. (A valid comparison, because both presidential primaries had accompanying state Supreme Court races.)
In 2016, 52% of voters claimed a Republican presidential primary ballot. In the concurrent Supreme Court election in 2016, as it happened, the Republican-endorsed candidate won 52% of the vote in a narrow victory. In this election, a grand total of 60% of the people who pulled a ballot preferred the Democratic primary. It’s not a leap of faith to suggest that an electorate that substantially different wouldn’t have an impact on the downballot contest.
However, there is data to suggest that the presence of the Democratic presidential primary did not yield as substantial a shift in the electorate as one might suspect.
For example, in last year’s Supreme Court election—which did not have the benefit of a presidential primary—the two Democratic strongholds in the state (Madison’s Dane County and Milwaukee County) combined for roughly 25% of the vote. If Democratic turnout was excessively goosed by the Democratic presidential primary, we would’ve expected that percentage to go up markedly. Indeed, as Reid Epstein of The New York Times noted in their election night coverage, the expectation was that in order for Karofsky to have “a good shot” at victory, Dane and Milwaukee would need to represent a combined 30% of the total vote.
The final figure for the two counties? 25.5%, a fractional improvement over 2019. What’s more, the largest red county in the state (Waukesha County) was a larger proportion of the electorate in this election than it was in the 2016 general election. (An election in which, of course, Republican Donald Trump narrowly carried the state.)
Going across the state, it does not appear to be true that the blue-leaning counties of the state cast an unnaturally high ratio of the vote. If one looks at the counties that saw the greatest turnout increase over the 2019 Supreme Court election, conservative Dan Kelly won seven of the 10. If one looks at the 10 counties that saw the least growth in turnout, they split evenly between Kelly and Karofsky.
An even more fascinating pattern holds, however, when making the more apples-to-apples comparison between the 2016 and 2020 Supreme Court races. In that scenario, every county in the state saw greater turnout four years ago than this year. This is almost certainly due to two factors: the fact that there were two competitive presidential primary battles in 2016, and COVID-19.
Now, given that all the action in 2020 was on the Democratic side (presidentially speaking), one would expect that the Karofsky counties would come the closest to matching their 2016 numbers since there was still an incentive for Dem-leaning voters to turn out. But that didn’t happen. Seven of the 10 counties with the best retention of their 2016 turnout levels were won on Monday night by the conservative candidate.
But one huge factor in progressive Jill Karofsky’s favor was that one of the few counties that saw a monster turnout that rivaled 2016 was Madison’s Dane County. Dane County retained 90% of its 2016 turnout (fifth best in the state), and it gave progressive Jill Karofsky an eye-popping 81.4% of the vote.
In fact, performance in another critical blue county suggests that the conservative thumping could’ve quite easily been worse. One of the lowest-performing counties on the 2016 v. 2020 turnout metric was, perhaps unsurprisingly, Milwaukee County. Given the highly publicized issues with a shortage of polling workers, the county was only able to staff around 3% of their normal voting precincts.
So while Dane County retained 90% of its 2016 turnout, Milwaukee County was able to only retain 71.6% of their turnout, one of the lowest comparative turnouts in the state. Green Bay’s Brown County, another county that struggled with staffing the polls amid the COVID-19 crisis, performed even worse with a retention rate of just 69%. Had Milwaukee County even retained its 2016 turnout at the statewide average (which was 79%), it could’ve padded Jill Karofsky’s margin by an additional 5,000 to 7,000 votes.
But what really has to be considered among the greatest factors in the conservative defeat, and one thing that has been a national issue for them, was the continuing GOP decay in the suburbs. Three of the top 15 counties in retention of their 2016 turnout were the somewhat famous (or notorious, if you are a Democrat) “WOW” counties to the north and west of Milwaukee: Waukesha, Ozaukee, and Washington. All of them had substantial turnout in this election (ranging from a retention rate of 83%-87%), and conservative Dan Kelly carried all of them.
But, even in nominally friendly territory, the conservatives face a crisis of sorts. To win Wisconsin, they need to do the following: dominate in the rural counties, hold their own in the nonurban but historically blue southwestern part of the state, and annihilate the Democrats in the WOW counties. The problem for Republicans is that while they are still winning in the WOW counties, they are losing that sheer dominance.
Consider the following shifts in the WOW counties between the 2019 Supreme Court election and the one that just passed:
Shift in Conservative Performance, 2019-2020
COUNTY |
2019 Share of VOTE
(R-Aligned CANDIDATE) |
2020 SHARE OF VOTE
(R-ALIGNED CANDIDATE) |
Ozaukee |
62.7% |
56.0% |
WASHINGTON |
74.8% |
67.9% |
WAUKESHA |
68.5% |
61.5% |
As you can see, in a consistent fashion, all three of these pivotal right-leaning counties saw an erosion in support for the conservative-aligned Supreme Court candidate. In fact, the rate of erosion was almost down to a tenth of a percentage point, with a range of 6.7%-7.0%. This is also a bigger erosion in support than the statewide average (which was 5.4%), so it really can’t simply be attributed to some kind of “uniform swing.”
The obvious counterargument here for Republicans is that 2020 was held with a Democratic presidential primary, and therefore, even on conservative turf, we would expect an enthusiasm gap to develop. There are two flaws in that theory. For one thing, as noted earlier, all three of these counties were well ahead of the state average in terms of retention of their turnout (based on the 2016 Supreme Court election). For another, if we look at an apples-to-apples comparison that doesn’t include the 2020 primary (support for Scott Walker in his last two gubernatorial bids), we see a similar trajectory.
Shift in Vote Share for Gov. Scott Walker, 2014-2018
COUNTY |
2014 Vote
(Scott Walker) |
2018 Vote
(SCOTT WALKER) |
Ozaukee |
70.0% |
62.7% |
WASHINGTON |
75.9% |
72.2% |
WAUKESHA |
72.5% |
66.1% |
Granted, here we see a bit more variance—Ozaukee being ahead of the curve is unsurprising, given that it is one of the most highly educated counties in the state, and education levels as a factor in the suburban movement away from the GOP is a well-known phenomenon. But the trends are unmistakeable.
And the suburban fade extends beyond the WOW counties. We tend to forget that the two Democratic strongholds in the state (Dane and Milwaukee) have significant suburban centers around their central cities (Madison and Milwaukee). And, quite clearly, they are shifting as well.
Take Milwaukee County. We know that in his three bids for governor, Scott Walker’s performance in his home county (he is a native of Wauwatosa, a town along the county line adjacent to Waukesha County) got progressively worse. He went from 38% support in 2010, down to 36% support in 2014, and finally dropping to just 31.5% support in 2018.
Digging into the numbers in Milwaukee County, we find something quite interesting. While Scott Walker did comparably badly in the city of Milwaukee both times (he did, indeed, do marginally worse in 2018), the real damage was done outside of the city.
In 2014 when Walker narrowly defeated Mary Burke, he actually carried the parts of Milwaukee County outside of the city of Milwaukee by a modest margin (53-46). That translates to a lead of around 11,100 votes.
In 2018, when Walker fell to a narrow defeat at the hands of Democrat Tony Evers, Walker’s performance in suburban Milwaukee was a reversal of his 2014 performance. In Milwaukee County outside of the city of Milwaukee, Tony Evers defeated Scott Walker by a 53-45 margin. That translated to a raw vote edge of 13,000 votes.
In other words, Evers won a net gain of 24,100 votes in just the relatively small (160,000-180,000) suburban parts of Milwaukee County. Not an insignificant gain when one considers that his total margin of victory was 29,300 votes.
Finally, the speed at which the other blue oasis in Wisconsin is getting away from the GOP is truly stunning. Daniel Kelly won only 18.6% of the vote in Dane County in this election. For comparison, in Scott Walker’s relatively narrow win in his initial bid for governor in 2010, he won 31% of the vote in Dane County. Heck, even when he was getting annihilated by double digits in the 2008 presidential election, John McCain managed to hang onto 26% of the vote in Dane County.
The problem for Republicans is those kinds of drops in large counties like Dane necessitate a larger counterbalance in smaller counties. Kelly in the 2020 Supreme Court election did roughly five points worse than Scott Walker did in his losing 2018 reelection bid. That means that if the GOP loses five points in Dane, they need to make up roughly 15,000 lost votes elsewhere.
Can they do that? Of course they can. They struggled in southwestern Wisconsin in this election, losing almost every county to the south and west of Madison. But that was a location where Trump ran particularly well. Furthermore, while the counties in the more rural and exurban north and north-central part of the state have turned red in recent elections, there is still further room for Democratic support to erode. In no way should Karofsky’s excellent result breed complacency. The math is still very much there for the GOP to hold Wisconsin in November.
But for Republicans, a tough though certainly plausible challenge is starting to exist. Even in what has been amenable turf, they are seeing some decline in suburban support. At the same time, the Democrats somehow have not managed to cash out all of their gains in places like Dane and Milwaukee Counties. That means that Republicans will need to not only replicate their 2016 performance elsewhere in Wisconsin, they will probably need to improve on it in order to win.