Leave a comment, and send us questions @FiveThirtyEight.
We’re calling it a night after one of the chalkier days of the campaign so far. All the favorites won, although in some cases the margins were a bit larger than we might have expected.
Trump had a very impressive victory in Arizona. Because it’s a winner-take-all state, the margin doesn’t matter from the standpoint of the delegate math. But it’s a good sign for how future states might go for Trump — especially California, which is among the most important states in Trump’s chase for 1,237 delegates.
However, Ted Cruz will win the Utah caucuses, and will very probably do so with more than 50 percent of the vote, meaning he’ll get all 40 delegates there. Meanwhile, it appears as though all of American Samoa’s delegates will go to the convention in Cleveland uncommitted. Thus, we expect the delegate count on the evening to be Trump 58, Cruz 40, uncommitted 9. That’s pretty good for Trump, although slightly fewer delegates than our “expert” panel expected earlier this week. Somewhat in contrast to last week’s primaries, therefore, when the headlines about close races didn’t quite express how well Trump had done in collecting delegates, this night is more of a morale/momentum boost for Trump than something that will dramatically increase his chances of getting to 1,237.
On the Democratic side, Hillary Clinton won Arizona easily, while Bernie Sanders won Utah and (although it hasn’t been called officially yet) very probably will win Idaho — in both cases perhaps by overwhelming margins. Thus, it’s probable — likely if I had to guess — that Sanders will win more pledged delegates on the evening.
Not all the news is good for Sanders, however. He was expected to win more delegates on the evening based on our demographic targets — and more importantly, he’s far enough behind Clinton that he needs to not just meet but blow out his delegate targets the rest of the way to have a shot at eventually catching Clinton. Alaska, Hawaii and Washington will vote on Saturday, states where we expect Sanders to perform well.
As a slight counter to that though, Harry, Trump’s margin in Arizona — which apparently held up pretty well in election day votes, although they’re taking forever to count — is pretty darn impressive. No, that doesn’t matter for the delegate math since Arizona is winner-take-all. But it’s a sign that the other campaigns have a lot of work to do if they want to win large, relatively diverse states.
Going strictly by the math (which may be dangerous if Trump picks up momentum after tonight), this was not a great night for Trump. Our expert delegate panel expected him to pick up 64 delegates tonight in the same projection that had him falling short of 1,237 after all the votes were cast. Our delegate target for Trump to reach 1,237 before the convention had him winning 70 delegates tonight. In reality, Trump’s likely only going to pick up a total of 58 delegates from American Samoa, Arizona and Utah.
It looks as though Sanders has a good shot to get more delegates than Clinton tonight.
Clinton is winning Arizona by 22 percentage points right now, which would translate to a net gain of about 16 pledged delegates on Sanders. And Arizona has more pledged delegates than Utah and Idaho combined. However, Sanders could net more delegates on the night if he wins Utah and Idaho by more than about 30 percentage points. Based on the results we’ve seen so far, he could fairly easily do that.
It turns out my wild dreams do sometimes come true. I was worried that we wouldn’t have results from the Republican caucuses in American Samoa before our live blog shut down for the night, but our friends at the Decision Desk report that all nine of American Samoa’s delegates will be announced as uncommitted.
OK, maybe I’m just grasping for some excitement on a night when Trump has swept Arizona and Cruz will sweep Utah. But, I’m genuinely curious as to whether Kasich can avoid the embarrassment of finishing behind Trump in Utah.
Right now, Kasich leads Trump 22 percent to 19 percent, thanks to a strong first place finish in tiny Sevier County and a solid second place showing in metropolitan Salt Lake County. But Trump leads Kasich 27 percent to 16 percent in booming Washington County, which is actually on the fringes of the Las Vegas metro area and has a decent population of non-LDS voters.
Kasich’s the favorite to top Trump based on the earliest returns, but I’m not ready to predict that Mitt Romney will award him the silver medal here quite yet.
The Boise, Idaho (Ada County) caucus sites just reported their Democratic results on CNN. Technically, it’s just the first ballot — voters have a chance to change their vote — but Sanders is doing extremely well, with 80 percent of the vote to Clinton’s 18 percent out of about 9,100 votes cast.
Earlier, I listed the states where the Clinton tandem has gone 5-for-5. What about the reverse? States where the Clintons have gone 0-for-5?
There are none yet, but there might be one soon. Bill Clinton lost Utah to Paul Tsongas in 1992, and in both his general election bids. Hillary Clinton lost it to Obama in 2008. And it looks like she’s about to lose it to Sanders this year.
Alaska is the only other potential 0-for-5 Clinton state, as far as I can tell.
On the Democratic side in Utah, Sanders looks like he’s heading towards an easy win. He’s up 63 percent to 33 percent in the early returns.
And at the risk of being repetitive: so far, as Dave pointed out, the results are pretty consistent across Utah, with Cruz above 50 percent all candidates to have reported votes so far. There’s nothing in from Salt Lake County, however, so we’ll wait at least a little bit longer.
About 2 percent of caucus sites are reporting across 11 counties in Utah. Cruz is above 50 percent in all 11 counties, and his share ranges from 55 percent (Millard County) to 80 percent (Rich County). As Harry said, this is looking like a clean 40 delegate sweep for Cruz. Tonight is looking like a pretty cut-and-dry night, delegate-wise.
We’ve got the first official returns from Utah, and it’s Cruz in a rout. He leads 62 percent to Trump’s 23 percent. If those results hold, he’ll win all 40 delegates from the state.
So far this year, Clinton’s map has resembled Barack Obama’s from 2008 as much as her own in 2008, but Arizona is one of the states she won both this year and in 2016.
But let’s raise the bar a little higher. In which states have the Clintons gone 5-for-5? That means a win for Bill Clinton in the 1992 Democratic primary, and both the 1992 and 1996 general elections — and a win for Hillary Clinton in both 2008 and 2016.
Arizona doesn’t qualify. Bill Clinton won it in the 1996 general election, but lost it in both the 1992 general election and primary.
As far as I can tell, the only 5-for-5 Clinton states so far are Arkansas, Tennessee and Ohio.
Several other states could potentially join them, however, where the Clintons are 4-for-4 and which haven’t voted yet in the Democratic primary yet this year. They include Kentucky, New Jersey, New Mexico, New York, Pennsylvania and West Virginia.
The Clinton campaign has done about a 180-degree turn over the last month and a half. After New Hampshire, some thought the Clinton campaign was in trouble. Clinton then won Nevada and really hasn’t looked back. Not only has her pledged delegate lead grown, but Trump’s delegate lead has blossomed at the same time. Trump may somehow end up being a strong general election nominee, but at this point he is even more disliked than Clinton, who isn’t adored by the general public herself. Clinton could also benefit from President Obama’s rising job approval ratings, which are now regularly topping 50 percent in the Gallup tracking poll. All in all, things seem to be turning Clinton’s way at the right time.
Dave noted a similarity between the Arizona and Florida votes in at least one retirement community, but I’d take the comparison a step further and say the two states have voted like each other in a number of recent primaries. Trump is getting roughly the same share of the vote (46 percent) in Arizona as he did in Florida. Clinton is getting 60 percent in Arizona and got 64 percent in Florida. In 2012, Romney got 46 percent in Florida and 47 percent in Arizona. In 2008, Clinton got 50 percent in both states.
Is Trump’s huge win in Arizona tonight because of counties on the Mexican border? Not necessarily. He’s doing well in border countries, but not any better or worse than elsewhere in the state. So far, Trump has 43 percent of the vote in highly populous Pima County (which contains Tucson), 49 percent in Yuma County, 45 percent in Santa Cruz County, and 49 percent in Cochise County. All those results are pretty similar to the 46 percent of the vote he has statewide.
Looking ahead on the Democratic side, I’d expect it to be a pretty good weekend for Sanders. While there really hasn’t been any polling in Alaska, Hawaii or Washington, Nate’s demographic model has Sanders favored in all three states by at least 17 percentage points. Obviously those projections come with a margin of error, but Sanders fans can take delight in what should be three victories. The problem, of course, is that Alaska and Hawaii are small states with few delegates up grabs, so it’ll be difficult for Sanders to make too big of a dent in Clinton’s delegate lead.
To follow up on my own question about Utah in the general election, Harry — yeah, usually I’d be super dismissive of a poll showing Democrats with a shot there. I can vaguely remember early polls from past presidential elections showing Republicans running competitively in Connecticut, Barack Obama with a chance to win North Dakota, and so forth. You take enough polls, especially in the early going, and you’ll get a few goofy results.
The thing about Utah, though, is that it’s dominated by one demographic group, Mormons, that doesn’t have a big presence in other states. And they have some plausible reasons to object to Trump — they’re a minority group in many senses, and Trump’s populist politics are potentially dangerous to any minority group. So it’s not that hard to imagine a big shift among that one group that isn’t replicated elsewhere. It’s easier to imagine Clinton winning Utah than winning Nebraska, for instance.
One thing to keep in mind as we watch the returns on the Republican side tonight is that the calendar slows down considerably over the next month. Between now and April 19, we have only the Wisconsin (April 5) and New York (April 19) primaries. That means we’ll be moving to a state-to-state campaign. It’s not until April 26 that we have a night with multiple states voting at one time again.
“Would a third party stand a chance given Trump and Clinton’s negative favorability?” — commenter Zach Dasher
In the abstract, it’s a very good opportunity for a third party candidate to run. The challenges are still manifest, however. For one thing, it’s a little late at this point to be assured of ballot access in all 50 states.
But more importantly, it’s hard for a third-party candidate to build a winning coalition. You might think, “well, there are plenty of independents,” but independents don’t necessarily agree on all that much. Some of them are more Trumpian, some are more Bloombergian, and some are “closet partisans” who are independent in name only. So most of the time, a third-party candidate is going to take votes unevenly from the two major parties. And whichever major party she takes fewer votes from will tend to win with a fairly clear plurality.
As I noted earlier, the first unofficial returns from Utah suggest that Cruz is going to win by a lot. Perhaps more interestingly, the early returns also suggest that Trump will run in a very distant third place. Nate, I’m usually not one to try and predict general election results from primary outcomes, but Trump seems to have a big problem with Mormon voters, and they make up the majority of voters in Utah. It’s not just that they prefer Cruz and Kasich to him; it’s that they really don’t like Trump or what he stands for. While I think Trump would still carry the very Republican state of Utah in the general election, I’m not sure I can simply dismiss out-of-hand polls showing him trailing Clinton.
We haven’t seen any official results from Utah yet, but reports from caucus sites suggest Trump will do quite poorly there. Do you think Democrats could win Utah in November, as a poll this week suggested?
Here’s an interesting tweet from Cruz campaign manager Jeff Roe:
However, in Maricopa County, the earliest election day returns show Trump leading Cruz 5,011 to 2,847, a tiny bit tighter than the early votes but not vastly different as was the case in Louisiana a few weeks back. My guess is that Trump’s margin over Cruz will continue to shrink tonight, but that he’ll still end up with a solid double digit win.
There are 131 pledged delegates at stake tonight for Democrats. It’s going to be hard for Sanders to win a majority of those given that Arizona has most of the delegates and Clinton is winning big there, but it’s not impossible if he crushes it in Utah and Idaho. So let’s say he almost does it. Clinton gets 66 delegates on the night and Sanders gets 65.
That would get Sanders up to 920 pledged delegates, while Clinton would have 1,242, with 1,889 pledged delegates still outstanding. Skipping a little bit of math, but Sanders would need 59 percent of the remaining total to tie Clinton in pledged delegates. That’s really difficult to do; it would be equivalent to beating Clinton by 18 percentage points the rest of the way out. Merely breaking even in delegates isn’t nowhere near enough for Sanders at this point.
Tonight’s likely result of a split verdict — Trump winning Arizona and Cruz stomping him in Utah — would seem to suggest a close fight between the two from here on out. But the problem for Cruz is the same one we’ve been talking about for over a month: There just aren’t a ton of great states for him later in the calendar. Almost every state with a high share of evangelical Protestants have already voted, and the two states with the highest shares of Mormons — Trump’s worst group by a mile — will have already voted after tonight (Idaho and Utah).
In fact, between now and June, the only states left where more than 3-in-10 residents are either evangelical Protestants or Mormons are Indiana (32 percent), Montana (32 percent), Oregon (33 percent) and West Virginia (41 percent). Together, these states account for just 146 of the 905 delegates at stake in remaining contests. And, West Virginia is probably Trump’s prototypical state, not Cruz’s. To me, these are the most damning statistics when it comes to Cruz’s bid to prevent Trump from winning 1,237 delegates.
There’s been an ongoing question during the Democratic campaign about which candidate is winning the Latino vote. In Nevada, the exit polls suggested Sanders carried the Latino vote, while an examination of precinct-level results argued Clinton did. In Florida and Texas, it was clear Clinton won Latinos. We don’t have any exit polls tonight, but it would seem that Clinton is again winning the Latino vote. In the votes counted so far from Santa Cruz County (83 percent Latino), Clinton is winning 70 percent of the vote. In Yuma County (62 percent Latino), she’s taking 67 percent of the vote.
Here’s part of the reason I think Trump’s apparently easy win in Arizona is important. I know Arizona looked like a good state for Trump all along. And I know we’re all into the delegate math here at FiveThirtyEight. It’s our thing.
But the delegate math suggests that Trump’s quest for 1,237 delegates could be close. He might just get over the finish line, or he might finish 10, 50, 100 delegates short. Or maybe something goes wrong and he finishes 200 delegates short instead, in which case he’d have a lot more work left to do.
To the extent Trump doesn’t clinch outright, however, and needs help from unpledged delegates or even delegates on the convention floor, there’s going to be a question of whether Trump has a mandate from Republican voters. It’s not so hard to deny him the nomination if everything is a big mess. It’s harder if Trump’s winning the strong majority of high-profile states, no one else is close, and Trump is just happening to come up a few delegates short. In my view, Cruz not only has to keep Trump below 1,237 — he probably has to create some real doubt about what Republican voters really want. He didn’t do that in Arizona tonight.