Thursday, April 7, 2016

How did Bradley win? Look to the Northeast

Now that I've had nearly 48 hours to reflect on the sickening outcome in the Supreme Court race, where homophobic Walker appointee Rebecca Bradley defeated JoAnne Kloppenburg for a 10-year seat on the court. And in taking a look inside the numbers, here's what jumped out to me.

And let's start the analysis by going back to the results of the Feb 16 primary, where Bradley got a surprisingly-low 44.6% of the vote, and Kloppenburg nearly beat her outright. In addition, the 3rd-place Joe Donald endorsed Kloppenburg soon after the primary, which would seem to make Kloppenburg a slight favorite if everything else held (and this was before people were aware what an amoral homophobic train wreck Bradley was as a person).

The upshot of this was that Bradley needed to improve her vote share by at least 5.4% on her February primary performance to get over 50%. So let’s use that 5.4% increase as a benchmark that Bradley had to beat in order to win. So when I laid out each county’s results on February 16 and compared it to April 5’s unofficial results, I found some interesting geographic differences in the data.

First of all, it wasn’t the Milwaukee area that was Bradley’s key to victory. If anything, Bradley underperformed in the media market for Wisconsin’s largest city. Sure, the sheep in the Republican WOW Counties (Waukesha, Ozaukee, Washington) voted for her, but not much more than they did for Mitt Romney in 2012, when Romney lost the state by 7%.

Romney 2012 vs. Bradley 2016, WOW Counties
Waukesha- Romney 66.76%, Bradley 68.50%
Ozaukee- Romney 64.63%, Bradley 64.92%
Washington- Romney 69.55%, Bradley 72.06%

In fact, Bradley’s vote share went down in 2 of the 3 WOW Counties compared to the February primary (Waukesha and Ozaukee). Bradley didn't gain all that much in the rest of the Milwaukee media market either. Out of the 11 counties that go from Sheboygan-Fond du Lac to the state line, Bradley only beat her +5.4% benchmark in 3 of them - Jefferson (+5.79%), Kenosha (+5.93%) and Milwaukee (+6.38%). Yes, slightly overperforming in Milwaukee County (due to a mysteriously high amount of Dem primary voters that voted for her) likely didn’t put Bradley too far behind, but it still probably translate to a loss if repeated statewide. This likely helps to explain why early exit polls were projecting a Kloppenburg victory, because Bradley wasn’t improving in the largest GOP areas.

So how did Bradley end up with approximately 52.4% of the vote? Because of a huge swing to her in the Northeastern part of the state, as well as some areas that Kloppenburg still won, but weren’t the big margins she needed. The Green Bay/Appleton media market was extremely kind to Bradley, and should take the “credit” for her victory. This area saw Bradley boost her vote share by large amounts, well above the benchmarks, and included the areas with the largest population.

Bradley Feb vote share vs. April vote share
Door Co. +13.29%
Shawano +12.99%
Kewaunee +12.94%
Langlade +12.53%
Oconto Co.+11.34%
Marinette +10.55%
Brown Co. +10.40%
Manitowoc +10.30%
Outagamie +9.97%
Calumet +9.70%

You might want to keep this list in mind when planning Summer and Fall vacations. Just sayin’.

Another area that Bradley did better noticeably better compared to February was in South Central and South Western Wisconsin. This seems counter-intuitive, because Kloppenburg won 11 of 12 counties on and south of a line from Beloit from La Crosse (only Grant County voted for Bradley,and barely). But Bradley was held to less than 40% of the vote in many of these counties in February, so keeping these counties relatively close meant that not only did Bradley beat her +5.4% benchmark, but her huge gains in Northeast Wisconsin outpaced Kloppenburg’s wins in the other corner of the state.

Bradley gains in SW/SC Wis counties
Crawford +15.46%
Grant Co. +11.82%
Sauk Co. +10.87%
Rock Co. +10.38%
Lafayette +10.27%
La Crosse +10.26%

The Grant and La Crosse County gains are especially interesting, because UW campuses are located in both of those counties. You would expect young people to be repulsed by a homophobic fundie like Bradley, but these figures bear out a certain theory making the rounds. The theory says that college students voting may not have been as familiar with the Supreme Court race, and either didn’t vote in it, or voted Bradley because they didn't know the difference between the candidates (which makes some sense- did you pay attention to State Supreme Court races when you were 19?). This pattern repeated throughout the state, and in addition to the underperformance in Milwaukee County, indicate the largest non-rural failures for Dems and the Kloppenburg campaign in the Supreme Court race.

Notable gains by Bradley in counties with UW campuses
Portage Co. +11.94%
Grant Co. +11.82%
Eau Claire +11.77%
Brown Co. +10.40%
Winnebago +10.32%
La Crosse +10.26%
Dane Co. +8.49%

The Dane numbers do come with a caveat, as Bradley still got less than 28% of the vote in my home county, no different than Romney did when he was losing Wisconsin in 2012. But if students weren’t aware of the race, and failed to fill out the Court race, Kloppenburg would likely have gained even more.

But more than college voters’ ignorance (and not being taught by Dem and Kloppenburg to pay attention), a look at these county-wide stats shows that just like with Scott Walker’s elections in the 2010s, Northeast Wisconsin remains a key advantage for Republican candidates. It also shows that where the race was most heated and “local”, in Dane County and in the Milwaukee area, that Kloppenburg got the vote totals she needed to win. If I’m the Democrats in Wisconsin- Tuesday’s election makes it pretty obvious where the work needs to be done.

Which makes Outagamie County Exec Tom Nelson’s entrance into the race for the open 8th Congressional District a big indicator. Nelson is a popular 2-term at the local level, as well as a prior State Representative and the 2010 Dem nominee for Lieutenant Governor. He makes for an interested test case as to whether the Northeast’s slant to the GOP is simply a “lower-turnout, state election” reason (bad enough), or a more permanent shift that makes capturing the state tougher for the Blue Team. 


  1. Must read analysis here:

    1. It would have been so easy to take both Bernie and Hillary's call for Kloppenberg's election and get out the message. The undervotes for JK were appalling. JK's campaign manager was the old DPW Comm chief. Surprise.

    2. Yep, Melissa Mulliken. And old Boomer operative who ran a very conventional, low-key campaign with lots of invite-only events. Kind of like another Boomer woman who lost statewide on Tuesday, come to think of it

    3. This is Melissa Mulliken, Judge Kloppenburg's campaign manager. MAL and Jake have it wrong: it is Melissa Baldauff who was the former DPW Communications person.

    4. Melissa Mulliken- thanks for reading and responding. I never said you were a DPW Communications official (I just was naming you), but I apologize if that wasn't clear.

      What is clear is that you're now 0-for-2 in important Court seats at times when a win would have put Walker down, likely for good. It also made the the nominal amount of dollars I threw to those campaigns a waste.

      Maybe it's time for someone else to do the job next April, or at least the strategy part. Just saying.

    5. DNC take note, again came unprepared for the gun fight with a fist. The NeoCon wins in this state are NOT just a fluke. The DNC needs to earn the votes of the people and of that they are doing very poorly. The last two elections are more proof that Wisconsin is no longer a progressive state, unless we compare ourselves to the heart of Dixie.

  2. Oh man. I am ashamed on my Green Bay--and we often visit Door County! Jesus, don't the artists up there vote? (Also, linked to your blog in my latest post--trying to send people to the good stuff.)

  3. Except... I'm not sure any of this really matters unless you take into account the presidential primary context.

    More voters turning out for the GOP primary than for the Dem primary explains the Bradley over Kloppenburg margin, to the decimal point.

    Given that dynamic, the rest of it pales into insignificance.

    1. That's a bit simplistic, Joy. It doesn't explain how Bradley got 15,000 more votes in Milwaukee than GOP primary vites, nor does it explain Kloppenburg getting more votes in other counties where the vast majority voted in the GOP primary.

      You're missing a third dimension of voters, who voted in the GOP or Dem primary because it held more interest to them (just like how 73% of prez primary turnout was Dem because it was getting more attention), but had no party loyalty, or other party loyalty. Those people were open season for the Supreme Court.

      And the huge dropoff number where 125,000 people didnt vote at all in the Supreme Court race is especially alarming, and inexcusable. Why wasnt this pushed in the month before the election, before the noise of the presidential campaign got here?

    2. Sorry, that's "73% of 2008's prez primary turnout was Dem."

  4. Is there a data source indicating the method of votes cast between electronic machine, paper ballot, and absentee in-person/ mail?

    1. I don't have one, but you are correct that something still doesn't add up. Maybe there should be a check some of those Milwaukee districts where Kloppenburg underperformed so badly, just to make sure that's a legit undervote and vote switch.

      Do the same in GB, esoecially where all those college kids were lined up at UWGB for hours