Wisconsin Primary Elections: Live Coverage And Results

Exit polls in Wisconsin show Cruz 10 points ahead of Trump. Based on the formula I described earlier, that would point toward an 8- or 9-point win for Cruz when combined with pre-election polls. The margin of error is high enough, though, that while a Cruz win is likely, it isn’t quite certain. Hence, ABC News and other networks aren’t likely to call the race until some actual votes come in.

https://twitter.com/NateSilver538/status/717514811265916929

If you’re a fan of our elections podcast (or, even if you’re a hater, why not?), give your podcast app a refresh. We just uploaded a special “FAQ” this afternoon, addressing some of the most frequently asked questions we get about the election. These are mostly structural questions — about how the election works, delegate rules, contested conventions and so forth. Here are the FAQs we tried to tackle with as much clarity as possible.

• Does turnout in the primaries indicate anything about turnout in the general election?
• Does head-to-head general election polling mean anything at this point?
• Why is there so little polling in some states and lots of polling in others?
• What makes a poll good or bad?
• Why do you always complain about the lack of polling? Why not just commission a poll yourself?
• What happens to a candidate’s delegates after they drop out?
• What exactly is a contested convention?

Take a listen here or subscribe in iTunes. And we’ll continue to answer more of your questions over the course of the election, so be in touch.

I don’t know if it’s an overreaction to what will probably be a Cruz victory in Wisconsin, but Trump’s chances of winning the nomination have tumbled even further in the prediction markets. The chances of a Trump nomination have fallen to below 45 percent. The odds haven’t been that low since the middle of February.

Here’s something to ponder. Most of the focus tonight will be on Cruz and Trump. But what if Kasich has a poor night while Cruz has a good one? This is relevant because we’re headed to a bunch of Northeastern states later this month where Kasich seemed set to be more of a factor in the race. Would an under-performance by Kasich be a bad sign for Trump because it indicates voters are capable of behaving tactically to stop him? Or would it be a good sign for Trump because it suggests Kasich’s campaign is flagging and there’s no clear alternative to Trump in the Northeast?

This is preliminary, but here’s some information that deserves more thought in light of Trump’s troubles in Wisconsin. The January Marquette University Law School poll found that 57 percent of registered voters in the state support allowing undocumented immigrants to stay and apply for citizenship. Pew reports that only 46 percent of respondents in a national survey supported a “path to citizenship” (although 72 percent supported allowing undocumented persons to remain in the country under some conditions). This certainly deserves further scrutiny, but it’s possible that the state’s somewhat more liberal immigration views, combined with other factors, have worked against Trump.

It was just four years ago that the Wisconsin primary pretty much brought an end to the Republican primary season. Back then, Romney beat Santorum by a 44-to-37 percent margin. A week later, Santorum dropped out of the race. This year the race will probably slog on, assuming Cruz takes a commanding share of the state’s delegates. If, however, Trump were to somehow pull off a surprise victory, it’d be difficult to see how Cruz would have a realistic path to the nomination, given that he needs to win states with somewhat similar demographics as Wisconsin. In that case, Wisconsin could once again be the final blow of the primary season.

One issue making national headlines is the debut of a new Wisconsin law mandating specific forms of voter ID. This may make a difference in the college-age Sanders vote. The Journal Times, headquartered in Racine, Wisconsin, included this account of a student who had to double-back to get appropriate ID:

UW-Madison senior Dylan Edwards was turned away from his Downtown polling place Tuesday morning because he only had a driver’s license from his home state, Pennsylvania. Like thousands of UW-Madison students, Edwards needed to get one of the university’s separate voting ID cards Tuesday morning. It took him about five minutes to wait in line and get a voting ID at an office in the Gordon Dining and Events Center two blocks from his polling place.

Edwards did go back to vote. College students occupy a special place in the voting universe, as they can vote either in their home districts or where they go to school. An academic analysis points out the varying ways states implement those rules: some have few restrictions while in others students are expected to show an investment in the civic community, through taxes and willingness to do jury duty. But Wisconsin appears to be reluctant to make sure that students and other voters know what its new rules are. As The New York Times editorial board wrote yesterday:

The Wisconsin law requires the state to educate voters about acceptable forms of ID and how to secure them — a particularly important public service for the roughly 300,000 state residents estimated not to have the proper ID. But despite requests from the state’s nonpartisan Government Accountability Board for $300,000 to $500,000 for that effort, the Legislature provided no funding. Instead, Governor Walker signed a bill in December to dismantle the board.

If Sanders, in particular, performs below expectations in Wisconsin, it will probably add to existing calls to evaluate the implementation of the ID law.

Somewhat to my surprise, Wisconsin ranks only seventh in the country in per-capita beer consumption, according to the National Institute on Alcohol Abuse and Alcoholism, trailing North Dakota, South Dakota, Nevada, Montana, Louisiana and New Hampshire. One must point out, however, that these statistics are potentially confounded by beer sales to non-residents, such as tourists in Las Vegas (Nevada) and New Orleans (Louisiana) and out-of-state residents crossing the New Hampshire border to buy low-tax beer there.

Wisconsin has generally been portrayed in the media as a white, rural place, and that’s true for most of the state. But Milwaukee is a major state population center, with just over half a million residents in the city limits and about another half-million in the suburbs. The classic map is this one: deep blue city, with deep red suburbs. The metro area is indeed very segregated politically, with Democrats and Republicans living near those who vote like them. The city and metro area is also among the most racially segregated in the country. As Erik Loomis of the blog Lawyers, Guns and Money points out, a new work of popular scholarship, Evicted, by Matthew Desmond, chronicles poverty and misery among Milwaukee’s black residents. There’s a substantial “graduation gap” between whites and African-Americans. Educational inequality, segregation, and criminal justice issues have led to the (somewhat controversial) claim that Wisconsin is one of the worst states to be black. These questions have implications for tonight’s primary. And when we describe Wisconsin as a white state, it’s important to remember that while this is true overall, the state is home to a more diverse city that has all the usual Rust Belt city problems – and then some. Racial inequality, and racial divisions, are key to understanding the state’s politics.

Trump is reportedly headed to California later this week. The state doesn’t vote until June 7, but he may be making a smart move. The state is winner-take-all by congressional district and its 172 delegates could easily determine the outcome of the Republican nomination. How do the polls look there? They’re pretty weird. California is a state where Rubio polled quite well before he dropped out. It doesn’t look like an especially good state for Trump, but it’s also not inherently all that strong for Cruz or Kasich. Trump is at 35 percent in our polling average, with Cruz at 30 percent, and a fairly large undecided vote. With so much time to go, our “polls-only” forecast gives Trump a 62 percent chance of winning California, Cruz a 33 percent chance, and Kasich a 5 percent chance. Our “polls-plus” forecast, by contrast, narrowly favors Cruz in California. That’s partly because it still looks at endorsements, which favor Cruz more than Trump. But also, “polls-plus” looks at the spread between a candidate’s national polls and his state polls. When a candidate’s state polls lag his national polls, as Trump’s do in California, that’s a bearish indicator. The silver lining for Trump is that all but 13 of California’s delegates will be awarded by congressional district. And as FiveThirtyEight’s David Wasserman explained today, there’s reason to think Trump could perform well in highly Democratic districts in the Los Angeles and San Francisco metro areas, where there are plenty of delegates available.

One clear indicator of how the Democratic race is going state-by-state is the percentage of the electorate made up by black voters. Early exit polls suggest that just 9 percent of Wisconsin Democrats voting today were black. That’s about the same as it was eight years ago, so it won’t be surprising if Sanders does well tonight. The problem is that in 2008, at least 15 percent of voters in the biggest delegate prizes remaining this month (Maryland, New York and Pennsylvania) were black.

Something that jumped out at me in some of the early exit polls: four in 10 GOP primary voters said that they were “scared” of what Trump might do in office. Those numbers went up to six in 10 for Cruz and Kasich supporters. The Wisconsin GOP scene has been more vehemently anti-Trump than many other states, so this tracks. Another notable item in the exit surveys was that about half of the Wisconsin voters said that they wanted someone who did have experience in elections, which is fairly high for this outsider-centric election year. And voters also seemed to be bucking against the sentiments that Trump has expressed recently about the U.S. involvement in affairs overseas — withdrawing from NATO, letting Russia do the fighting against ISIS. Half of Wisconsin voters said they thought the U.S. should be more active in world affairs. So, overall, the voters of the state seem to be hewing more to party orthodoxy than to Trumpism.

Getting back to Nate’s talk of consolidation of the anti-Trump vote, I think it’s going to be interesting to see if voters got the message to get behind Cruz in Wisconsin. Cruz was able to snag Gov. Walker’s endorsement and, if the polling is to be believed, is on his on way to victory. That consolidation doesn’t matter that much in New York because of the way delegates are allocated, but it does matter in many other winner-take-all congressional districts or on the statewide level. If anti-Trump voters can get behind Cruz, it would significantly impede his path to 1,237 delegates.

Cruz may be the strong favorite to win Wisconsin tonight, but there’s still a way Trump could end up having a very respectable night. If there’s any region where Trump could “steal” valuable delegates from Cruz, it’s blue-collar northern Wisconsin – think iron ore, lumberjacks, and shuttered paper mills. Wisconsin awards 24 of its 42 delegates on a winner-take-all by congressional district basis, with three delegates going to the winner in each district and 18 going to the statewide winner. There are only three districts where fewer than a quarter of residents over age 25 hold a college degree: the 3rd District (23 percent), the 7th District (21 percent), and the 8th District (23 percent).

These districts — particularly the 7th, which takes in Wausau and Superior — should be winnable for Trump even if Cruz decimates him in the more white-collar Milwaukee suburbs and prevails statewide. If Trump could win three, six, or even nine delegates out of Wisconsin, it would make a huge difference in his march to 1,237. Northern Wisconsin has demonstrated a stronger attraction to political outsiders than other parts of the state — an auspicious sign for Trump. Between 1969 and 2011, the congressman from the 7th District was Democrat Dave Obey, a legendary appropriator. When he announced his retirement in May 2010, he famously declared “There’s no way in hell a progressive district like mine is going to a conservative Republican.” Six months later, the 7th District voted to elect a conservative Republican and former MTV reality TV star named Sean Duffy. Since then, Duffy has won re-election twice.

In a primary season as captivating as the one we’re in, it’s easy to overlook local politics. Not in Wisconsin. Today voters there weighed in on a state Supreme Court election that was “non-partisan” by definition only. The race between Justices Rebecca Bradley and JoAnne Kloppenburg featured seven-figure spending by partisan groups and arguments from both sides that the candidates are too political to sit on the court. Bradley was given an interim appointment to the state Supreme Court by Gov. Scott Walker after the death of one of the justices last year. Kloppenburg sat on the panel that dismissed citations against Walker protesters in the lead-up to his recall election in 2011. Bradley named her judicial role models as Scalia, Alito, and Thomas. Kloppenburg named hers as Sotomayor, Ginsburg, and Kennedy. Voters can take the hint. The legacy of Wisconsin’s polarizing governor was likely on the minds of voters today, along with the heated debates on display in local and federal courts. In a primary that has seen relatively low turnout among Democrats, it wouldn’t be surprising if the local dynamic of today’s vote helped Wisconsin defy that trend.

On this week’s elections podcast, we engaged in a bit of a thought exercise: If you stepped into an elevator with someone who had been paying zero attention to this year’s election, how would you describe Trump’s place in the race? We wanted to use the question as an excuse to step back and assess his coalition, his polling, his base — without getting caught up in the outrage of the moment. Here’s how Nate answered:

Usually the strategy that candidates take is to try and be as broadly appealing as possible. Donald Trump instead applied a strategy of trying to win the narrowest plurality possible that would allow him to win the Republican nomination… He’s won actually 20 percent of the popular vote if you count both parties. 37 percent of about a third of the Americans that are Republicans.

To me the thing the media profoundly missed until recently is that Donald Trump is profoundly unpopular with most Americans. And that’s not a new story. It’s gotten a little worse, but it’s been true for a long time, and he might with the party nomination despite that.

How’s that happened? Number one, among that 37 percent, they tend to feel very strongly for Trump. They are hard to peel off. Number two, the Republicans took a long time to get organized — in fact, they kind of failed at that task. So a 37-percent plurality can be enough to win an awful lot of the states and come close to the delegate majority. Number three, electability was not a major issue in the election.

How would you answer this question? Tell us in the comments at the right, or if you can fit it into 140 characters, tweet me. Remember, try to focus on the numbers/demographics that describe Trump’s role in this race, not the narrative around the things he says (though of course those are often hard to dissociate).

Politics in Wisconsin have become emblematic of national divisions. This is significant in the primary because the divisions in each party – the Trump-Cruz divide and the Clinton-Sanders one – map pretty well onto Wisconsin’s geography. Trump’s areas of strength are in the Northwest part of the state – much of it in the large congressional district that is now represented by Sean Duffy, a Republican, but was represented by a prominent Democrat, Dave Obey, for more than four decades. Cruz, in contrast, is popular in the more densely populated, deep-red Milwaukee suburbs, a far more affluent area which is represented in Congress by Jim Sensenbrenner, a Republican. For Democrats, the Madison-Milwaukee area maps pretty clearly onto the Clinton-Sanders question. Madison is a bastion of white progressivism, while Milwaukee is a majority-minority union town. But while much of the Wisconsin political landscape is representative of what’s happening nationally, politics in the state have resisted nationalization. The popular narrative of Russ Feingold’s Senate loss in 2010 was that he had stopped representing the state and become more like the national Democratic Party. Now Feingold seems poised to defeat Ron Johnson, who was elected in the Republican wave that year, on similar grounds. And while liberal Democrat Tammy Baldwin is just as strong a contrast to Johnson, she’s developed a reputation for staying behind the scenes with “workhorse” issues like manufacturing jobs and care for veterans. And Scott Walker’s short presidential bid opened up questions about the national potential for Wisconsin leaders. Paul Ryan, on the other hand, remains in the national spotlight as House Speaker and a name floated for a possible floor nomination. Wisconsin’s recent tendency to both reflect and resist national political trends has some implications for tonight’s contests. The splits within parties might represent the big picture. But the results are likely to buck national trends. Some have suggested that this will be a “turning point,” at least for the GOP, but it’s also possible that it’ll just be another example of the state’s insistence on a unique political trajectory.

One question I’m interested in tonight is whether Trump picked up any undecided voters. In the average poll 21 days before today’s primary, Trump averaged 34 percent of the vote. In all the previous contests, Trump has just picked up 0.2 percentage points in the average state from his average poll within 21 days of the primary. That is, he tends to hit his polling average and gets nothing more. In only five of 26 times did he outperform his average by more than 4 percentage points, which is the margin by which Cruz leads Trump in the FiveThirtyEight Wisconsin polling average.

From the very beginning, one of the big questions about the Republican race was how much second-choice support Trump would pick up as other candidates dropped out of the race. We haven’t really had a good test of that yet since Rubio ended his bid on Mar. 15. In Arizona, which voted the following week, so many ballots were cast early that Rubio still won a significant share of the vote, and in Utah, Mormons were going to be deeply opposed to Trump whether or not Rubio was on the ballot. So we’ll get a read of that in Wisconsin tonight. In the meantime, we can also look at national polls. The FiveThirtyEight national polling average is designed to be pretty conservative, rather than overreact to individual polls — nobody should be in a hurry to update a national polling average since there’s no such thing as a national primary. But three weeks of data since Rubio dropped out is pretty useful, and it suggests that almost all of his voters have gone to Cruz and Kasich.

Cruz and Kasich appear to have won most Rubio voters

	FIVETHIRTYEIGHT NATIONAL POLL AVERAGE
CANDIDATE	MAR. 15	APR. 5	CHANGE
Trump	38.0	40.3	+2.3
Cruz	21.7	29.3	+7.6
Kasich	10.7	18.4	+7.7
Rubio	14.8	Dropped Out	-14.8

Rubio had about 15 percent of the vote when he dropped out. Since then, Kasich and Cruz have each gained about 8 percentage points in national polling while Trump has added just 2 points. That’s consistent with earlier evidence from exit polls and other sources showing few of Rubio’s voters going to Trump. And over the longer term, Trump has gained only 5 points in our national polling average since Jan. 31, the day before Iowa voted — at which point there were still a dozen Republican candidates running. I’m not going to troll everyone by saying that Trump has a ceiling — the issue is more complicated than that — but there does seem to be something of a #NeverTrump vote.

With voting in Wisconsin open until 9 p.m. Eastern time (8 p.m. locally), it’s going to be a while before the networks release their exit polls to the public. But here’s a formula that might be useful once that happens — or if you’re making inferences about what the exit polls say based on the teasers that have been reported so far. So far this cycle, a blend that weights the exit poll at 75 percent and the FiveThirtyEight “polls only” forecast at 25 percent would give you the best projection of the actual results. In other words, you should afford those exit polls a fair amount of weight. (You can get a slightly more accurate projection by using the FiveThirtyEight “polls-plus” forecast instead of the polls-only forecast, but let’s not overcomplicate things.) Nonetheless, there’s no substitute for actual results. The exit poll + pre-election poll blend still had a margin of error of plus-or-minus 6 percentage points when predicting an individual candidate’s vote share. And the margin of error in predicting the difference between two candidates is even larger — about 12 percentage points in a two-candidate race — since when one candidate gains votes it comes at the other candidates’ expense.

Wisconsin really gets stereotyped — cheeseheads, milk maids, overalls — but what many people don’t realize is that it’s actually quite a politically active place (which, when phrased like that, almost sounds like something your Sunday school teacher warned you about doing before marriage). Wisconsin has a higher-than-most rate of participation in presidential elections, routinely getting over 60 percent turnout in presidential elections; for a comparison, turnout in the 2012 election was about 53 percent nationally. It’s got a history of lefty-ism (see the socialist mayor of Milwaukee once upon a time) as well as far righty-ism (see Joseph McCarthy). A tradition of being on the forefront of various modes of political thought, in other words. In this year’s contentious Republican race, the state also finds itself at the center of the Never Trump/Stop Trump/Stump Trump/Dump Trump movement. Trump is trailing Cruz in the state, with our polls-plus model giving Cruz an 89 percent chance of winning the state (B+ ain’t bad, ya know?). So why is Cruz doing so well there? One reason might be that the institutional powers in the state’s Republican party/establishment are not exactly Trump fans. Paul Ryan, speaker of the House and Janesville, Wisconsin’s fittest, most Jack Kemp-loving resident, isn’t so sure about Trump — he recently dressed-down Trump for suggesting that there might be riots at the Republican National Convention. Gov. Scott Walker is a former contender in the GOP race who came out for Cruz a couple of days ago on a Wisconsin talk radio show. People like Ryan and Walker have real sway when it comes to turning out party faithful in the state, who might be more inclined to go with the guys they know and trust rather than Trump. Maybe, in short, the establishment still matters a little bit in Wisconsin!

Thankfully, the election gods have blessed us with another important primary night — it’s been a while. The eyes of every election nerd are on Wisconsin, where the polls will close at 9 p.m. EDT. (We’ll have exit poll data before then, of course.) So what should we be expecting as the results start rolling in and onto our computer screens? It’s all about margins. On the Republican side, Donald Trump is in trouble. Our polls-only and polls-plus models give Cruz an 80 percent and 89 percent chance of winning, respectively. Cruz led Trump 40 percent to 30 in the last Marquette University Law School poll, the most reputable poll in the state. A double-digit margin could mean that Cruz (who would get 18 delegates for winning statewide) locks Trump out of winning any of the 24 delegates awarded to the winners of each congressional district (three in each district). Not only will that hurt Trump’s chances of reaching 1,237 delegates overall, the number he needs to clinch the nomination, but it would suggest that Trump, who still hasn’t won a majority of voters in any statewide primary, hasn’t picked up much support since Marco Rubio left the race. On the Democratic side, Sanders has a 72 percent chance of winning in both our polls-only and polls-plus models. And while winning is never bad, the 3-percentage-point margin both models project for him is far below the pace he needs to set. Sanders, instead, probably needs something like a 16-percentage-point blowout in Wisconsin to be “on track” to close Clinton’s advantage of 224 pledged delegates. That’s because all delegates are awarded proportionally in Democratic primaries, and Wisconsin has few black voters. If Sanders cannot run up the margin in a state as white as Wisconsin, he’ll be unlikely to mount a comeback in the remaining states with bigger delegate prizes, like New York, Pennsylvania and California. With all this in mind, sit back and enjoy. We’ll start getting results at 9 p.m., but it’ll likely take a couple of hours for most of the precincts to report their votes. We’ll be with you the whole way.

Tweet us a question or comment

Leave a comment, and send us questions @FiveThirtyEight.