Sunday, April 10, 2016

So was Supreme Court race just turnout= Bradley?

This is a pretty good "poli-sci"-type article from the La Crosse Tribune that came out Thursday, in the wake of this week's Wisconsin presidential primary and Supreme Court race. It talks some about the results of the races, but spends a lot of time concentrating on turnout figures from throughout the state. Here's a good graphic showing how turnout varied throughout the state on April 5, with the lightest shades showing 30% turnout, to the darkest shades showing 60% turnout.

In addition, the La Crosse Tribune story allows us to see which party's presidential primaries were voted in for each county, and then we can translate those figures across for the Supreme Court race, to see if those numbers truly match up, and explain the outcome of the race (a common meme that has been bandied about). In La Crosse County, this pattern did seem to hold.
Despite some embarrassing revelations during the course of the campaign, Wisconsin Supreme Court Justice Rebecca Bradley, a freshly minted Walker appointee, kept her seat with a 52 percent to 48 percent victory, outpolling state appeals court Judge JoAnne Kloppenburg by 91,247 votes.

Statewide, about 94,000 more voters cast ballots in the Republican primary than in the Democratic primary, a number remarkably close to the margin of victory in the nominally nonpartisan high court race.

La Crosse County turned those numbers on their heads: 24,105 people cast ballots in the Democratic primary, 19,989 in the Republican primary. Kloppenburg was a big winner here, by a 55 to 45 percent margin.
So let's look at other key counties in the state and see if the pattern holds. A great thing about that La Crosse Tribune article is that it not only shows total turnout, but breaks down the turnout between the 2 parties, so we can determine the share of the primary votes that each party's race got. Let's start with the top Dem counties for share of the overall vote, then compare that to Kloppenburg's vote totals. I'm going to remove Menominee County from this analysis, as they had the lowest turnout in the state, and only accounted for 720 votes total in the Supreme Court race.

Top "Dem turnout" counties, April 2016 primary
Dane County 70.7%, Kloppenburg 72.2%
Milwaukee Co. 62.8%, Kloppenburg 55.3%
Ashland Co. 59.2%, Kloppenburg 62.7%
Bayfield Co. 56.9%, Kloppenburg 62.3%
Iowa County 56.3%, Kloppenburg 57.5%
La Crosse Co. 55.3%, Kloppenburg 55.1%
Douglas County 54.8%, Kloppenburg 54.8%
Eau Claire Co. 54.3%, Kloppenburg 55.6%

There's that gap in Milwaukee County that I've talked about before, where Kloppenburg underperformed the Dem turnout by 7.5%. Otherwise, Kloppenburg's vote share matched and sometimes exceeded the Dem turnout, including a huge number of votes in Dane County.

Now let's see if the reverse is true in Republican-leaning counties. I am going to remove Florence County for similar reasons that I removed Menominee County in the Dem list- they only accounted for 1,500 votes, and make for a very small sample size to draw from.

Top "GOP turnout" counties, April 2016 primary
Washington Co. 74.0%, Bradley 72.1%
Waukesha Co. 70.7%, Bradley 68.5%
Green Lake Co. 69.6%, Bradley 66.2%
Ozaukee County 67.7%, Bradley 64.9%
Taylor County 67.4%, Bradley 64.1%
Dodge County 65.1%, Bradley 62.2%
Oconto County 64.6%, Bradley 62.5%
Fond du Lac Co. 64.4%, Bradley 62.7%

So in the places that gave the highest share of its presidential votes to the GOP primary, Bradley was consistently underperforming by 2-3%. That may not seem like much, but that would have taken the race into recount territory, if not a close Kloppenburg win.

So maybe there was a lot of vote switching, where Dem counties went Bradley or GOP counties went Kloppenburg? No. There were only 6 of those counties that came close to hitting that criteria (and 2 of them were 50-50), and it was the opposite trend- more 50-50 and GOP counties voted Kloppenburg.

50-50 + GOP counties that voted Kloppenburg
Columbia Co. 51.0% GOP, Kloppenburg 52.1%
Richland Co. 51.1% GOP, Kloppenburg 52.4%
Jackson Co. 51.2% GOP, Kloppenburg 51.1%
Lafayette Co. 53.5% GOP, Kloppenburg 51.4%
Green County 50-50, Kloppenburg 56.7%

50-50 and Dem counties that voted Bradley
Buffalo Co. 50-50, Bradley 55.5%

So let's go back to my theory from a few days ago, which says Northeastern Wisconsin was the main reason Bradley won, and see what the patterns in that part of the state look like. Let's start with the three most-populous largest counties in that area (Brown, Outagamie and Winnebago), then look at the rest of the region.

Northeastern Wisconsin, April 2015 primary
Brown County 56.5% GOP, Bradley 58.0%
Outagamie Co. 56.3% GOP, Bradley 58.2%
Winnebago Co. 54.2% GOP, Bradley 56.2%

Door County 54.2% GOP, Bradley 52.5%
Manitowoc Co. 60.9% GOP, Bradley 62.8%
Kewaunee Co. 61.5% GOP, Bradley 62.6%
Calumet Co. 62.0% GOP, Bradley 63.6%
Marinette Co. 62.8% GOP, Bradley 60.6%
Shawano County 63.6% GOP, Bradley 64.8%
Oconto County 64.6% GOP, Bradley 62.5%

Here we go! Bradley overperformed in much of Northeast Wisconsin, especially the areas with the most people. This is where she won and Kloppenburg lost, and it means that there were likely some Dem/Bradley votes in these parts, or Dems not voting in the Supreme Court race, and that is a failure on the Klopppenburg/DPW side.

It begs the question- WHY? Is it a Catholic/religion thing? Is it a gun thing? Is it bad DPW branding? This is what needs to be worked on by the state party between now and November, not just for the presidential and U.S. Senate races, but also because that area has a winnable open House seat and some key swing Legislative races.

So thanks to the La Crosse Tribune for their in-depth article, because it really opens the door to further explaining how a fool like Bradley got voted onto the State Supreme Court. It wasn't as simple as "more GOPs turned out", but had more to do with a stronger performance by Bradley in the GB/Appleton media market, and Kloppenburg's underperformance in Milwaukee County.


  1. I have lived in Outagamie Co my whole life. I think the biggest difference between NE Wisconsin and the rest of the state is the TV newsmedia. This area is obsessed with Packers and teh weather. Rarely does the local TV news report on state politics. However, hey did with the 2015 Budget debate. They did with Walker campaign. The result was that voters became informed. Informed voters are terribly dangerous to the WisGOP since the more informed Wisconsin voters are they more likely they will be anger. They love their schools but rarely does the TV media report about the budget cuts to schools. They hate fiscal irresponsibility but you will see Aaron Rodgers in a Viking uniform before you see any budget journalism on local TV news. The typical local news reports breakdown this way: in a 30 min broadcast, 15 minutes are spent on weather. 5-10 minutes on sports. The remaining time is usually fluff stuff or reports of housefires. It is sad. NE Wisconsin never found out how bad Bradley is as a candidate or they would never have voted for her.

    1. That's good insight, thanks for sharing it.

      Remembering college people from GB and knowing people who grew up there, the Packer thing is ridiculous in those parts (and I'm a 35-year Packer fan). I could see Demetri Goodson's suspension getting more attention this week than the Supreme Court race, which means all the locals hear about state politics is right-wing ads and Jerry Bader BS.

      That's where Dems have to step up and puncture that bubble.

    2. NE Wisconsin's Media profile is STRONGLY Catholic, Gun, and Conservative. The Conservative money continually hammers this area with ads supporting their candidates and the so called "independent group" ads with negative content about the Liberal side. It is relentless and very one sided. The Democrats need to do a much better job of countering these ads.

  2. I think that Kloppenburg's under-performance in Milwaukee County may be due to socially-conservative blacks voting for Bradley because WI Supreme Court races are officially non-partisan, so they're not turned away from a bible-thumper like Bradley by having the Republican label actually alongside her name on the ballot. Basically, socially-conservative blacks who would never vote in a Republican primary for any office, but voted for Bradley because she agrees with their anti-LGBT and other forms of bigotry, helped Bradley win. If that's the case, it would be Clinton/Bradley voters, not Sanders/blank or Sanders/Bradley voters

    I would need to see precinct-by-precinct data from Milwaukee County to determine whether or not that was actually the case.