Nate Silver's Blog, page 136

It’s Really Hard To Get Bernie Sanders 988 More Delegates

After a trio of landslide wins in Washington, Alaska and Hawaii on Saturday — the best single day of his campaign — Bernie Sanders narrowed his delegate deficit with Hillary Clinton. But he still has a lot of work to do. Sanders trails Clinton by 228 pledged delegates and will need 988 more — a bit under 57 percent of those available — to finish with the majority.

That alone wouldn’t be enough to assure Sanders of the nomination because superdelegates could still swing things Hillary Clinton’s way in a close race, but put aside that not-so-small complication for now. The much bigger problem is that it isn’t easy to see where Sanders gets those 988 delegates.

If you’re a Sanders supporter, you might look at the map and see some states — Oregon, Rhode Island, West Virginia, Montana and so forth — that look pretty good for Sanders, a lot like the ones that gave Sanders landslide wins earlier in the campaign. But those states have relatively few delegates. Instead, about 65 percent of the remaining delegates are in California, New York, Pennsylvania, New Jersey and Maryland — all states where Sanders trails Clinton in the polls and sometimes trails her by a lot.

To reach a pledged delegate majority, Sanders will have to win most of the delegates from those big states. A major loss in any of them could be fatal to his chances. He could afford to lose one or two of them narrowly, but then he’d need to make up ground elsewhere — he’d probably have to win California by double digits, for example.

Sanders will also need to gain ground on Clinton in a series of medium-sized states such as Wisconsin, Indiana, Kentucky and New Mexico. Demographics suggest that these states could be close, but close won’t be enough for Sanders. He’ll need to win several of them easily.

None of this is all that likely. Frankly, none of it is at all likely. If the remaining states vote based on the same demographic patterns established by the previous ones, Clinton will probably gain further ground on Sanders. If they vote as state-by-state polling suggests they will, Clinton could roughly double her current advantage over Sanders and wind up winning the nomination by 400 to 500 pledged delegates.

But things can change, and polls can be wrong — and so it’s worth doing the math to see what winning 988 more delegates would look like for Sanders. Call it a path-of-least-implausibility. If you think Sanders can meet or exceed these targets, then you can say with a straight face that you think he’ll win the nomination. If you think they’re too good to be true, then you can’t. Here’s the Bernie-miracle path I came up with:

SANDERS DELEGATE TARGETSTATE OR TERRITORYNO. ELECTED DELEGATESORIGINALREVISEDPOPULAR VOTE MARGIN NEEDED TO REACH TARGETCalifornia475239274Sanders+15New York247125128Sanders+4Pennsylvania18996101Sanders+7New Jersey1266167Sanders+6Maryland954243Clinton+9Wisconsin864850Sanders+16Indiana834448Sanders+16Oregon613745Sanders+48Puerto Rico603033Sanders+10Connecticut552831Sanders+13Kentucky552833Sanders+20New Mexico341818Sanders+6West Virginia291719Sanders+31Rhode Island241316Sanders+33Delaware211010Clinton+5Montana211316Sanders+52South Dakota201214Sanders+40District of Columbia2089Clinton+10North Dakota181114Sanders+56Wyoming14911Sanders+57Guam73.54Sanders+14Virgin Islands73.54Sanders+14Total1,747896988Sanders+13Sanders’s unlikely path to a pledged delegate majority

To repeat, these are not predictions. On the contrary, they describe a rose-colored-glasses scenario for Sanders that I consider to be very unlikely. To develop them, I started with our original pledged delegate targets for Sanders. Those already look optimistic for Sanders, who has underperformed his delegate targets in most primaries (he’s beaten them in most caucuses, but there aren’t many caucuses left on the calendar).

But for Sanders to get a pledged delegate majority, even our original targets aren’t enough now — they’d leave him 92 delegates short. So I kept tweaking these numbers upward until I got Sanders to 988 delegates. I was a bit more conservative about giving him extra delegates in states with substantial black or Hispanic populations, since Sanders has tended to underperform our original projections in those states. But mostly, I had to be very liberal about those extra delegates.

I assumed Sanders would narrowly win New York, for instance, even though he’s trailed Clinton by margins ranging from 21 to 48 percentage points in recent polls there. Likewise, I had him winning Pennsylvania and New Jersey, where polls also have him down by 20-something points. And I had Sanders winning by a landslide, 15-point margin in California, even though he’s behind in polls there also. (Because Democrats’ delegate allocations are highly proportional, it’s easy to approximate the popular vote from the delegate count and vice versa.)

I assumed Sanders would win Oregon by the same enormous margin that he won Washington, even though Oregon is a primary while Washington held caucuses. I gave a blowout win to Sanders in Kentucky, even though neighboring Ohio and Tennessee easily went for Clinton.

The most recent poll of Wisconsin, which votes next week, has Clinton winning there. I ignored it and assumed Sanders will win by 16 percentage points instead. The demographics do look pretty good for Sanders in the Badger State.

But is Connecticut a good state for Sanders? I’m not so sure: Its demographics are more Ralph Lauren than L.L. Bean. I gave it to Sanders anyway.

I assumed Sanders would win Puerto Rico because it’s a caucus, even though Clinton has much of the party establishment behind her. New Mexico? Nearby Arizona and Texas went overwhelmingly for Clinton, but let’s give it to Sanders.

You get the picture. It’s not hard to imagine Sanders meeting these super-optimistic projections in a few of the states. But he’ll have to do so in all of the states, or else he’ll have even more ground to make up elsewhere. If he loses Wisconsin, for instance, or only narrowly wins it, that’s more votes he’ll need to win in New York or California.

The good news for Sanders is that this scenario would represent such a massive sea-change that superdelegates really might have to reconsider their positions. You might even say it would require a revolution, a profound rejection of Clinton and the status quo.

Listen to the latest episode of the FiveThirtyEight elections podcast.

http://c.espnradio.com/s:5L8r1/audio/2714714/fivethirtyeightelections_2016-03-28-180814.64k.mp3?ad_params=zones%3DPreroll%2CPreroll2%2CMidroll%2CMidroll2%2CMidroll3%2CMidroll4%2CMidroll5%2CMidroll6%2CPostroll%2CPostroll2%7Cstation_id%3D4278Subscribe: iTunes |Download |RSS |Video

One Weird Trick To Lose The 2016 Election: Alienate Women

In this week’s politics Slack chat, we talk Donald Trump, sexism and the general election. The transcript below has been lightly edited.

micah (Micah Cohen, politics editor): Greetings, all! With no debate or election this week, let’s consider a longer-term problem: Trump and women. The Internet spent the past week picking over a startling fact: Women (and I’m talking Republican, Democratic, unaffiliated, etc.) disproportionately don’t like Trump. We’ll get to why in a moment, along with what it means electorally, but first question: The data has shown for a while that Trump has less support among women than among men — why all of a sudden the media interest?

harry (Harry Enten, senior political writer): I think it’s a few things. First, because there are no elections or debates this week, the media is looking for new ground to cover. And with Trump ahead in the delegate count, the pivot to the general election is obvious. Second, I think the whole Heidi Cruz picture thing really put the issue at the forefront. Even Newt Gingrich and Ann Coulter are having a tough time defending Trump on this.

natesilver (Nate Silver, editor in chief): One answer is that the media is mostly dudes. But this is part of a broader issue also, which is that the media has only just now started to write how terrible a general election candidate Trump might be. Trump’s favorables among the general electorate might have gotten a bit worse, but they’ve been epically terrible since he launched his bid (actually, since before he launched it). Women are a big part of that, of course.

I mean, it literally took nine months for the conventional wisdom to recognize that someone could be astoundingly popular with a plurality of the electorate and extremely unpopular with a majority of it.

clare.malone (Clare Malone, senior political writer): The media interest has always been there; Megyn Kelly gets all kinds of credit for bringing up Trump’s treatment of women from the very beginning of the campaign. But there’s been a host of crazy things that he’s said throughout the election, about a host of other groups and the misogyny thing has basically been this baseline element of Trump; it’s the foundation to his looming political tower built by big hands and no loans. Just look at the news that broke today: Trump’s campaign manager Corey Lewandowski was charged with battery against former Breitbart News Network reporter Michelle Fields.

But I think that sexism didn’t bubble back up into the conversation — i.e., the candidates talking about it — until it was about someone’s wife, which is troubling. It’s good that Trump’s sexism is being addressed, but it’s also important to note that it wasn’t addressed by a candidate until it directly hurt that candidate in his personal life. There are some notions of chivalry and protecting your woman that are at play here, so it’s also about the male ego. Wives are supposed to be off limits in politics, though Hillary Clinton might know that that certainly isn’t the case.

natesilver: Yeah, another part of the answer is that Trump puts so much crap out there that it’s been hard for the media to focus on any one part of the story for very long.

micah: So just how poorly is Trump doing among women? Republican women and women overall?

clare.malone: By the way, readers, one person in this chatroom is wearing a “Feminist” T-shirt today. Guess who.

Harry: From The Huffington Post:

As you can see, Republican women in particular are much more unfavorable towards Trump than to Cruz or John Kasich (or the dearly departed Marco Rubio). Those are horrible numbers.

The same poll has Trump with just a 21 percent positive rating with all women nationwide. His negative rating with women nationally is 70 percent.

clare.malone: This might be a historical level of dislike, no?

harry: No major party nominee has had that bad of a rating, based on the data I look at.

micah: So Trump does really poorly with women — it hasn’t stopped him in the Republican primary, why will it stop him in the general?

natesilver: Because winning 51 percent of 100 percent is way harder than winning 35 percent of 35 percent?

clare.malone: The Republican primary is kind of nuts that way. If Republican voters were rational actors, thinking strategically for the general, then Kasich would be the nominee.

natesilver: I mean, there are some decent arguments to be made that Trump could win a general election. We’re overdue to do more exploration of those. I just don’t find the argument that because he beat expectations in the primary, he’ll also do so in the general election to be very well thought out. The reason is that a big part of Trump’s success in the primary is because he figured out that in a 17-candidate field, you can double down on a plurality and win, not worrying that much about what the rest of the electorate thinks of you. That doesn’t work in the general election.

micah: So if 70 percent of women dislike Trump on Nov. 8, he’s basically guaranteed to lose, right?

clare.malone: More women vote than men, that’s for sure. According to our friends over at Rutgers Center for American Women and Politics, American women have cast between 4 and 7 million more votes than men in the last couple of presidential elections.

micah: This has been a pet peeve of mine in how the media talks about women, as if they’re simply another demographic group, like Latino voters or retirees. Women are a majority of voters, making up a roughly equal share of the electorate in every state.

natesilver: Yeah. Apart from the fact that Trump’s misogyny is really gross, it’s also really dumb as an electoral strategy.

clare.malone: Yup, women contain political multitudes — white women, for instance, went for Mitt Romney in 2012.

micah: Of course they contain multitudes! They’re over half the electorate!

natesilver: Yeah, a Machiavellian could argue that demonizing blacks or Hispanics or Muslims or gays is a winning, if incredibly cynical, electoral strategy for the GOP. That doesn’t work when you’re talking about 51 percent of the population. Furthermore, the fact that Trump doesn’t realize this suggests either that (i) he’s just making shit up as he goes along instead of being some sort of brilliant tactician; (ii) he’s a sexist down to his core and can’t help himself; or (iii) both.

harry: At the end of the day, elections are about getting more votes. Democratic candidates for the House got more votes among women than Republicans did in 2014. Their problem was that Republicans did really well with men. So Republicans can benefit from a gender gap, too. Right now, though, Trump is losing that tradeoff. He’s down big time in the polls, as I pointed out early Tuesday.

natesilver: But for all that said, I’m wondering if we’re going about looking at Trump’s general election numbers sort of backward? We know they’re really terrible, enough that he’d lose in a landslide today against Clinton, even though she’s also pretty unpopular. We also know that numbers can change a lot and that March general-election numbers have historically not been all that predictive.

Listen to the latest episode of the FiveThirtyEight elections podcast.

http://c.espnradio.com/s:5L8r1/audio/2714714/fivethirtyeightelections_2016-03-28-180814.64k.mp3?ad_params=zones%3DPreroll%2CPreroll2%2CMidroll%2CMidroll2%2CMidroll3%2CMidroll4%2CMidroll5%2CMidroll6%2CPostroll%2CPostroll2%7Cstation_id%3D4278Subscribe: iTunes |Download |RSS |Video

micah: So that’s my question: How much can Trump’s popularity with women change between now and November?

natesilver: Or to put it just slightly differently: Do the gender splits in Trump’s numbers suggest he’ll have a harder time improving his numbers than if he were equally unpopular with both men and women?

clare.malone: A great mystery … or perhaps not a mystery at all. To Nate’s point above, about the various theories of why Trump says sexist things, I think he kind of can’t help it, and while his sexist comments are being written off by the GOP electorate who’s voting for him, it’s going to be increasingly hard for him to pivot away from that image in the general election. I think he’ll try to do it, though, by deploying his oh-so-poised and career-oriented daughter, Ivanka, and I think he’ll also do it by just being less vituperative in his speech — he’ll pour a little whole milk on the hot-sauce rhetoric and hope for the best.

micah: But, Clare, hasn’t Trump just said too many sexist things? His record on this was long and bad when the campaign started! And it has only gotten worse.

clare.malone: I mean, to us sitting here, sure, but maybe not for the electorate, right? Who knows, people, women included, might think that his “outsiderism” is more important to the country and outweighs his sexism. I think that’s unlikely — he’s gonna catch a lot of flak — but that’s a thing that could happen.

natesilver: You’re probably going to see a lot of ads like this from Democrats:

micah: But that is why, to Nate’s question, I think this will be harder for Trump to change than if he were disliked equally by both genders …

clare.malone: Why? Because it’s more entrenched and women will reinforce the dislike more and more, as in an echo chamber of Trumpian disgust?

micah: Yeah. That is, there’s a very clear real-world explanation for Trump’s unpopularity with women — it’s not due to the vagaries of the news cycle or a well-run campaign against him. He’s sexist, and women don’t like him.

natesilver: One other big theme is that Trump had an element of surprise in the Republican race, which is part of why his opponents never really developed an effective strategy against him. That won’t be true for him against Clinton, who has had months to think about a strategy and put together opposition research that the GOP campaigns skimmed over.

clare.malone: Right. The Clinton people are just biding their time; they’re not directly engaging with Trump yet, but it’s gonna happen and it’s gonna be a full-scale assault.

harry: Again, I call on the Simpsons for discussing Trump’s strategy:

micah: OK, let me argue against myself now … the press has almost been begging Trump to start acting “presidential,” which suggests to me that if he does mellow/moderate, the media won’t do a very good job of holding him to his record, and instead marvel at how much of a “statesman” Trump has become. So in a few months, the raging sexist we’re all familiar with might seem a distant memory, no?

clare.malone: No. There will be his years of comments to contend with, and if people start treating him like a real candidate and asking him about that historical record, the sexism issue won’t die.

micah: IDK, you seem to have more faith in the media than I do.

clare.malone: You’re a member of it! Have faith in yourself.

natesilver: The media’s a big, complicated thing. The “Morning Joe” panel will almost definitely be talking about the amazing “pivot” that Trump has made.

micah: :crying_cat_face:

clare.malone: This Cruz’s wife dust-up is the amuse bouche of sexism talk for the general election; even if Trump tones down his rhetoric, I think I have faith in people — or maybe I’m encouraging people — to apply some rigor to examining his record, writing about his record, thinking about his record.

harry: Yeah, Trump has faced more of an onslaught from fellow candidates in the past month than he had previously. That will only continue, especially from the Clinton camp.

clare.malone: Sexism is alive and well, and exists not just in the flagrant foul form that Trump has exhibited, but also in other insidious ways. I think even if Trump “pivots,” people will continue to call him on these subtler forms of sexism, and that’s a darn good political strategy. It’s also good for American society, because we’re all pretty sexist, frankly. Including yours truly — social conditioning is hard to shake!

natesilver: I mean, this thing could go a lot of different ways. It’s possible that “the fundamentals” will take over, and that we’ll have the close election we might have expected before the GOP’s dumpster fire of a nomination process. In that event, I think you’ll see the media covering Trump’s sexism through the same tired “both sides” framework evident in headlines like this:

Over the years, Donald Trump has said things about women that his critics have called offensive https://t.co/udo4DodsmF via @arappeport

— NYT Politics (@nytpolitics) March 25, 2016

It’s also possible that the bottom will fall out from Trump. Everyone will abandon him, including a lot of Republicans who would ordinarily be defending him, because they want to (i) save their seats in Congress and (ii) eviscerate Trump so that they will get their party back in 2020. As John Sides, Lynn Vavreck and other folks have pointed out, a lot of the reason the “fundamentals” work in general elections is because both sides are pulling on the tug-of-war rope with equal force. A really divisive nomination process could change that.

harry: Of course, Trump hasn’t won any nomination yet.

natesilver: No, he hasn’t. And stuff like this won’t make it easier for Trump if he comes up short of the 1,237 delegates needed to clinch the nomination.

clare.malone: Whatever the outcome, the men’s rights movement is certainly going to be typing frantically away on their subreddit channel.

Elections Podcast: Momentum And Misogyny

Our elections podcast crew discusses whether “momentum” is real and whether Bernie Sanders has it coming off of his wins in Hawaii, Alaska and Washington. Plus, how much will Donald Trump’s record of sexism on and off the campaign trail hurt him with voters if he reaches the general election in November?

http://c.espnradio.com/s:5L8r1/audio/2714714/fivethirtyeightelections_2016-03-28-180814.64k.mp3?ad_params=zones%3DPreroll%2CPreroll2%2CMidroll%2CMidroll2%2CMidroll3%2CMidroll4%2CMidroll5%2CMidroll6%2CPostroll%2CPostroll2%7Cstation_id%3D4278Subscribe: iTunes |Download |RSS |Video

You can stream or download the full episode above. You can also find us by searching “fivethirtyeight” in your favorite podcast app, or subscribe using the RSS feed. Check out all our other shows.

If you’re a fan of the elections podcast, leave us a rating and review on iTunes, which helps other people discover the show. Have a comment, want to suggest something for “good polling vs. bad polling” or want to ask a question? Get in touch by email, on Twitter or in the comments.

Trump Will Have A Hard Time Turning Blue States Red In November

Donald Trump’s performance in primaries and caucuses has created some unusual geographic patterns. What sort of candidate dominates in both Alabama (where Trump won 43 percent of the vote on Super Tuesday) and Massachusetts (where he got 49 percent)? Why was Trump so strong in Hawaii — and so weak in Kansas? The answers to these questions are complicated. Trump attracts voters for multiple reasons: Economic anxiety and racial resentment are important factors in his success, but hardly the only ones.

Let me propose One Simple Trick that makes the geographic patterns at least a bit clearer: Trump’s strengths and weaknesses are easier to understand if you consider how many Democrats and Republicans turned out in each primary. Republican voters are a small minority of the overall electorate in states like Massachusetts and Hawaii, so their support for Trump isn’t a good indication of how those states might behave in November.

Take Massachusetts, for example. Trump’s 49 percent — his highest fraction in any state to date — was on light turnout: Only about 630,000 voters participated in the Republican primary, compared with 1.2 million for the Democratic one. Thus, Trump won only about 17 percent of the overall vote among Bay Staters who turned out that day.

By contrast, while Trump’s performance in Ohio might seem poor at first glance — he got 36 percent of the vote and lost to John Kasich — it’s better once you consider that Republicans turned out substantially more voters than Democrats. As a share of the combined Democratic and Republican primary turnout, Trump got 22 percent of the vote in Ohio — a fair bit better than he did in Massachusetts.

Here are those figures for all states where both Democrats and Republicans have voted so far.1

TRUMP’S SHARE OF PRIMARY/CAUCUS VOTESTATEREPUBLICANDEMOCRATIC+REPUBLICANMississippi47.3%30.6%Alabama43.429.6Tennessee38.927.1Arizona47.126.6Florida45.726.5Missouri40.924.5Georgia38.824.4Ohio35.622.4South Carolina33.021.9Nevada45.921.7Arkansas32.821.3Louisiana41.420.4North Carolina40.220.2Virginia34.719.7Michigan36.519.2New Hampshire35.218.7Texas26.717.8Massachusetts49.317.0Oklahoma28.316.4Illinois38.816.1Kansas23.315.2Iowa24.313.9Vermont32.710.2Utah14.09.8Maine32.69.4Minnesota21.37.6United States37.121.3

Source: The Green Papers, United States Elections Project

This calculation makes it clearer that Trump’s strengths are mostly in the South. Of Trump’s top seven states so far by his share of the combined primary or caucus vote, five or six are in the South, depending on how you classify Missouri.

New England looks like a poor region for Trump, by contrast. His share of the combined primary or caucus vote was slightly below his national average in Massachusetts and New Hampshire and well behind it in Vermont and Maine.

The industrial Midwest has been about average for Trump. The Ohio result, as I mentioned, is better for Trump once you consider the very high Republican turnout there (and that he lost to the state’s governor). But his performances in Michigan and Illinois rate as middling by this metric even though Trump won both states. The Great Plains states have been a poor region for Trump, while the West has been a mixed bag. Trump’s big win in Nevada is less impressive once you consider that Democratic turnout outpaced GOP turnout. But his results from Arizona hold up well.

This method leaves a lot of things to be desired. If you were being more exacting, you’d want to adjust for whether each party held an open or closed primary in each state, how many opponents Trump faced at various stages of the race, and other factors.

But as the calendar turns toward bluer states, be wary of making extrapolations from Trump’s performance in the primaries to how he might perform in the general election. Overall, Trump is deeply unpopular with general election voters and will have a lot of work to do to repair his image should he become the Republican nominee. The race can and will change, and Hillary Clinton shouldn’t take a lot for granted. But Trump is more likely to “transform the electoral map” by turning red states blue, rather than the other way around.

Listen to the latest episode of the FiveThirtyEight elections podcast.

http://c.espnradio.com/s:5L8r1/audio/2709334/fivethirtyeightelections_2016-03-21-173834.64k.mp3?ad_params=zones%3DPreroll%2CPreroll2%2CMidroll%2CMidroll2%2CMidroll3%2CMidroll4%2CMidroll5%2CMidroll6%2CPostroll%2CPostroll2%7Cstation_id%3D4278Subscribe: iTunes |Download |RSS |Video

Past Terrorist Attacks Helped Trump Capitalize On Anti-Muslim Sentiment

Before President Obama made a statement on the terrorist attacks in Brussels today, presidential contender Donald Trump had already weighed in: first on Twitter and then on Fox News and NBC, which interviewed him by phone.1

Trump would “close up our borders” in response to Brussels, he said in the Fox News interview, adding that the U.S. would have to be “very, very vigilant as to who we allow into this country.” The comments echoed proposals issued by Trump after previous acts of terror. Following the series of terrorist attacks in Paris on Nov. 13, Trump appeared to propose (and then partly pulled back from) a national database to register Muslims. Then, after the attacks in San Bernardino, California, on Dec. 2, Trump called for a “total and complete shutdown of Muslims entering the United States.”

We might expect more rhetoric along these lines from Trump because the Paris and San Bernardino attacks appeared to boost his standing in national opinion polls, as well as the amount of media attention he received. And Trump’s rhetoric on terrorism will likely continue to focus heavily on immigration — according to exit polls, Republican voters aren’t particularly keen on Trump’s crisis-management skills, but most favor his ban on Muslim immigrants, and immigration overall is one of the top draws for Trump voters.

Below, you’ll find two versions of the FiveThirtyEight national polling average as it ran from June 16 last year, the day that Trump entered the presidential race, through Nov. 12, the day before the Paris attacks. One version (the gray line) is essentially the same as our regular national polling average,2 which is deliberately designed to be conservative and slow-moving. The other (the red line) is a “high sensitivity” version that places a much greater premium on poll recency; it’s noisier but is probably better able to capture the public response to developing news events.

As you can see, Trump’s national polls had stagnated in the mid-to-high 20s in the two months before Paris; Ben Carson was approaching him in national surveys and pulling slightly ahead of Trump in Iowa. Meanwhile, the news cycle had become much less Trump-obsessed than usual. According to a forthcoming study we’ll be publishing of headlines at Memeorandum.com, the political news aggregator, Trump was the lead political story on only three of the 50 days before the Paris attacks. Google search interest in Trump was also considerably down from its peaks in August and September.

But the Paris and San Bernardino attacks were associated with an uptick in Trump’s numbers. According to our high-sensitivity polling average, Trump improved from 28 percent of the vote just before the Paris attacks to 32 percent on Dec. 1, the day before the San Bernardino attacks. His numbers then rose further, to about 35 percent by mid-December.

It’s possible the timing of Trump’s polling bounce was coincidental, but that seems unlikely given his focus on both Islamic terrorism and immigration since the start of his campaign. Furthermore, this period was associated with a sharp rise in news coverage for Trump, which has tended to both reflect and reinforce his gains in the polls. Following the relatively limited coverage of Trump in October and November, he was the lead news story for six straight days on Memeorandum after he first floated the Muslim ban on Dec. 7. Google searches for Trump more than doubled after he proposed the ban.

Exit polls in states that have voted so far suggest that immigration and anti-Muslim sentiment may be bigger factors in Trump’s gains than his ability to handle terrorism itself. On average in states where exit polls have been conducted, 38 percent of voters who said terrorism was their top issue voted for Trump — the same as his overall level of support in these states. But 56 percent of voters who listed immigration as their top issue went for Trump.

TRUMP SUPPORT AMONG VOTERS WHO LISTED THIS AS THEIR TOP ISSUESTATETRUMP’S SHARE OF VOTETERRORISMIMMIGRATIONAlabama43%44%61%Arkansas333253Florida464760Georgia394157Illinois394069Iowa242144Massachusetts495074Michigan373962Mississippi474356Missouri413963Nevada463662New Hampshire352953North Carolina403660Ohio363468Oklahoma283429South Carolina333151Tennessee394450Texas273435Vermont334370Virginia353442Trump supporters care more about immigration than terrorism

Results for immigration voters in Oklahoma and Vermont are inferred based on answers in other categories

Source: National Election Pool Exit Polls

In some states, the exit polls have also asked two other relevant questions: First, whether voters favor “temporarily banning Muslims who are not U.S. citizens from entering the U.S.,” and second, which candidate they think would best “handle an international crisis.”3

Trump has performed well on the Muslim ban question, winning close to half of the roughly two-thirds of Republicans who agree with the ban, while underperforming with those who don’t. But Republicans may have concerns about whether Trump passes the commander-in-chief test: In all eight states where the exit polls asked the question, fewer Republicans said they preferred Trump in an international crisis than voted for him overall.

STATETRUMP’S SHARE OF VOTESHARE OF VOTERS WHO SAY TRUMP WOULD BE BEST IN AN INTERNATIONAL CRISISTRUMP’S SUPPORT AMONG VOTERS WHO FAVOR BAN ON MUSLIMS ENTERING U.S.Alabama43%40%52Arkansas333038Georgia393645New Hampshire353045South Carolina332741Tennessee393545Texas272637Virginia353144GOP voters aren’t into Trump for his crisis-handling skills

Source: National Election Pool Exit Polls

So if the past is any guide, Trump will escalate his attacks on Muslims and immigrants in the coming days. His opponents, especially Hillary Clinton, will press him on his steadiness in office and his fitness to handle a crisis. After a couple of days of coverage from Brussels, the news coverage will revert to being all about Trump.

Listen to the latest episode of the FiveThirtyEight elections podcast.

http://c.espnradio.com/s:5L8r1/audio/2709334/fivethirtyeightelections_2016-03-21-173834.64k.mp3?ad_params=zones%3DPreroll%2CPreroll2%2CMidroll%2CMidroll2%2CMidroll3%2CMidroll4%2CMidroll5%2CMidroll6%2CPostroll%2CPostroll2%7Cstation_id%3D4278Subscribe: iTunes |Download |RSS |Video

The Most Important States On Trump’s Path To 1,237 Delegates

On Monday, we published the results of a survey of fellow presidential election delegate obsessives that we conducted in an effort to map out Donald Trump’s route to the 1,237 delegates he needs to clinch the Republican nomination. So for this week’s politics Slack chat, we also gathered the available members of that panel to talk over/geek out on the results. The transcript below has been lightly edited.

micah (Micah Cohen, politics editor): All right, everyone, here’s the average estimate of Donald Trump’s delegate haul for each state from our expert delegate panel: [Check here for some notes on the delegate numbers in this table.]

DELEGATESDATECONTESTAVAILABLEPROJECTED FOR TRUMP3/22Arizona5858Utah404American Samoa924/5Wisconsin42254/9Colorado3474/16Wyoming1414/19New York95714/26Maryland3831Connecticut2819Rhode Island1910Pennsylvania1716Delaware16155/3Indiana57375/10Nebraska361West Virginia34335/17Oregon28125/24Washington44176/7California17293New Jersey5151South Dakota290Montana270New Mexico2410New Trump delegates513Delegates to date695Total1,208How many more delegates will Donald Trump win?

Trump projections based on an Olympic average of estimates from Nate Silver, Harry Enten, Adam Geller (National Research Inc.), Daniel Nichanian (Daily Kos), Henry Olsen (Ethics and Public Policy Center), Margie Omero (Purple Insights), Patrick Ruffini (Echelon Insights) and David Wasserman (Cook Political Report).

We’ll get to some of the most interesting and critical states in a moment, but what were your biggest takeaways from the panel’s estimates?

harry (Harry Enten, senior political writer): My biggest takeaway is how on the knife’s edge this is going to be. We probably won’t know whether Trump is going to clear the 1,237 threshold until at least June. It’s March 21 — we have a long way to go.

dave (David Wasserman, House editor at the Cook Political Report and FiveThirtyEight contributor): First off, while we don’t know whether Trump will hit 1,237 or not, we should all be able to agree on one takeaway: For the first time in a very long time, every state will matter — and yeah, this thing’s going all the way to June. I don’t see any way for Trump to attain 1,237 until June 7, and I don’t see any realistic way for him to be mathematically eliminated from 1,237 before June 7.

natesilver (Nate Silver, editor in chief): When I was filling out my projections, I was a little bit surprised that I had Trump coming up short of 1,237 since I thought I was being fairly optimistic for him in individual states. However, once you account for the facts that (1) there are quite a few uncommitted delegates and (2) Trump isn’t likely to do so well in less populous states west of the Mississippi, which is a fair bit of what’s out there — well then, Trump has to do really well everywhere else.

daniel (Daniel Nichanian, contributing editor to Daily Kos Elections and PhD candidate in political science at the University of Chicago): I agree that it’s striking that Trump is hovering right around the number of delegates he needs (1,237) with estimates that are generally optimistic for him. But they’re also not his best-case scenario. It’s easy to see how slightly underperforming these benchmarks in just a few of these states would leave him far from 1,237, but also how slightly outperforming them (say, in a state like New York) gets him higher still.

harry: I think what Daniel (or Taniel) is saying points to the fact that there’s been a lot of overconfident predictions both that Trump will and won’t make it to 1,237. Caution is warranted.

natesilver: Can I get geeky here? I think this is a case where the mean and the median estimate of Trump’s delegates might be pretty different.

micah: Geek out!

natesilver: I wound up being toward the lower end of the group on Trump because I was looking at things probabilistically. For instance, I have him as likely to win Arizona but not certain. Meanwhile, in a few other states where other folks regarded Trump as certain to lose, I gave him a chance. So often my estimate was one of the ones screened out by the “Russian judge”/Olympic-style scoring method.

micah: (Where the highest and lowest estimate for each state get tossed before we calculate the average.)

natesilver: In general, though, there are a lot of states that look pretty good for Trump but could cost him a lot of delegates if he slips, but fewer where you have it the other way around. So if you were taking some sort of weighted mean from a probability distribution, I think it would wind up with fewer delegates for Trump than a median or modal projection.

dave: Another thing we should make clear here to those who haven’t spent many nights geeking out on The Green Papers or Ballotpedia: A big myth going around is that states are either purely winner-take-all or they’re proportional. The fact is, there are a ton of delegates remaining that are winner-take-all by congressional district, like in California, Wisconsin and Maryland. The calendar from here on out will really require Ted Cruz and John Kasich to know where to “cherry-pick” for delegates.

daniel: That’s where the dynamics of a three-way race will be fascinating to watch: Will Kasich and Cruz boost the anti-Trumpers’ hopes by each picking up delegates in congressional districts where they have some strength (as happened in Illinois at the district level), or will they neutralize each other at the statewide level, as happened in Illinois as well?

natesilver: If Kasich and Cruz were implicitly (or explicitly) collaborating, their strengths and weaknesses might line up pretty advantageously to stop Trump from getting 1,237. However, it’s not clear that Kasich is playing along.

dave: This is where the anti-Trump forces seem to fail. They’re convinced they can get voters to vote strategically, but there’s little evidence of that working so far.

micah: How big of a punch bowl turd is Kasich for the #NeverTrumps?

natesilver: A pretty big one, I think, to the point where it’s fair to wonder whether Kasich might be interested in a VP slot on a Trump ticket.

micah: Message-wise, that would be an odd fit.

natesilver: Our polls-only forecast has Cruz just barely getting over 50 percent in Utah, for instance, which means he’d get all the delegates there in a potentially big (albeit not terribly surprising) blow to Trump. But Cruz would be more assured of surpassing 50 percent if Kasich were out campaigning in Wisconsin or New York instead of in Utah.

daniel: Many Republicans are trying to marginalize Kasich. And there are great arguments to be made that he’s hurting anti-Trump forces by allowing Trump to continue winning with pluralities — not to mention endangering the Cruz campaign’s efforts to shut Trump out of Utah, as Nate said. But upsides of having Kasich in are that: (1) The GOP needs to prevent Trump from hitting 50 percent triggers in New York and Connecticut, and (2) Kasich could play better in some congressional districts (as he did in the Chicago suburbs) where it’s harder to see Cruz overtaking Trump.

dave: It seems to me like Kasich’s genuinely convinced there are still some places only he, and not Cruz, can beat Trump. And he may have a small point — so far he’s won more votes in a lot of leftie hangouts like Ann Arbor, Michigan; Burlington, Vermont; and “Harry Entenville,” Hanover, New Hampshire. It’s just that there may be more delegates at stake in places where Kasich is robbing Cruz of outright wins.

harry: See, that’s the thing about Kasich. If he were willing to “play along,” he could help the anti-Trump folks in the remaining New England states and the northern Mid-Atlantic states. The problem is he is also trying to play in the West and Midwest, where Cruz can probably do fine all by himself.

micah: So will it all come down to how organized/targeted the anti-Trump candidates/voters/forces will be? That seems like a recipe for a Trump nomination, based on their record to date.

natesilver: I’d distinguish the voters from the candidates. The voters have sometimes been willing to play along. See Ohio, for instance, where Marco Rubio’s voters dutifully seemed to go to Kasich. The candidates have been another matter.

daniel: To Nate’s point: We may have also seen that after Super Tuesday, when Rubio’s support collapsed in a matter of days. It’s possible some of that was due to strategic consideration, and to some voters coalescing around the strongest anti-Trump alternative. (That’s when Rubio fell from around 20 percent in Louisiana’s early voting to around 10 percent of primary day voters.)

micah: All right, let me turn the convo briefly: What factors did you use to estimate Trump’s delegate haul in each state besides that state’s delegate allocation rules? Where is Trump strongest and weakest, and what’s that based on?

harry: Well, for me, these estimates were based off three factors. (1) The results so far. Trump has generally done better in the South and Northeast, while being weaker in the Midwest and West. (2) Polling of upcoming contests, where available, to confirm that regional pattern. It generally has. (3) How much coalescing of the anti-Trump vote there will be in a given state.

natesilver: I threw sheep entrails around on a large map of the United States and made predictions based on where they landed.

micah: Sheep are notoriously anti-Trump; they’re always making baaaah-d deals!

harry:

dave: Trump seems to be benefiting a great deal from the fact that there are many places where neither Cruz nor Kasich has much natural appeal. For example, Trump has won a lot of delegates in African-American-majority congressional districts in places like St. Louis and Chicago where there aren’t many evangelicals, wealthy whites — or for that matter, Republicans period. So I see Trump winning a lot of delegates in minority-majority districts in places like California and New York, and that’s a big problem for #NeverTrump right now.

natesilver: I guess that’s one difference — I’m a little bit less certain how those super-blue congressional districts will go. They actually seemed to be pretty good for Rubio earlier in the campaign. Will that support now transfer to Kasich?

daniel: For me, an uncertain part in making these projections was the dynamics of the three-way race: (1) The GOP field is finally consolidated enough that there is some uncertainty as to whether Trump can go high enough to win states that he would have been clearly favored to win with a plurality had they voted earlier. (2) But it’s still unclear how Kasich plays in some states where his degree of viability will be a major factor in the allocation of statewide and district delegates.

harry: In Illinois, the majority-minority districts were clearly pro-Trump, but in Missouri they were far less so. And we have very limited information in a lot of states. Can Kasich keep it up? I have no clue.

natesilver: Yeah, Kasich is the biggest “known unknown” in a lot of ways. I think we have a pretty good idea of where Cruz’s strengths and weaknesses are. But how high do Kasich’s numbers go, and where does he take his support from?

dave: Personally, I just said screw it and gave Cruz every district bordering Canada.

daniel: It’s late March, 60 percent of delegates have been allocated, and the rest is dependent on the strength of a candidate who for now has only won his home state!

harry: This year is bizarre.

dave: And Daniel, I’m sure we’ll talk about this in a moment, but it really does matter how much of a home-state bounce Trump gets in New York. He’s the only one left whose home state hasn’t voted.

daniel: Yes, Kasich could take in the votes of Rubio supporters, and he has shown that he can attract significant support. But he has also disappointed in states he has focused on other than Ohio, like Michigan, where he came in third. It really remains to be seen how much support Kasich can attract outside of his home state.

micah: All right, so let’s start really getting into specific states: What contests emerged from this exercise as the most crucial in determining whether Trump reaches 1,237?

daniel: There is no path for Trump to 1,237 if he doesn’t do well in California.

natesilver: California was the state with the largest standard deviation among the panelists. It has a lot of delegates, and we don’t have a great idea of what’s going to happen there.

dave: There are really three “clusters” left, right? (1) Acela Corridor (2) Great Plains (3) West Coast. And Cruz is lucky the Republican National Committee’s delegate allocation seem to be based on sheep rather than people.

harry: I think the first big test is Wisconsin in early April. That’s a state where Rick Santorum did reasonably well in 2012 against a far stronger candidate, Mitt Romney. Polls have shown Trump stuck at around 30 percent and with very high negative ratings, including from the Republican stronghold of Waukesha.

Can the anti-Trump vote consolidate there?

natesilver: The thing about Wisconsin is that it feels like a state that could easily wind up Trump 38 percent/Cruz 31 percent/Kasich 31 percent. Not too far from where Illinois was.

Wisconsin, like Illinois, issues a lot of its delegates by congressional district. But a top-line vote like that wound up playing pretty well for Trump at the district level in Illinois.

dave: To Nate’s point about Kasich spoiling his way to Trump’s VP, I actually think that role could be played by Scott Walker not going full Romney on Trump in Wisconsin.

Listen to the latest episode of the FiveThirtyEight elections podcast.

http://c.espnradio.com/s:5L8r1/audio/2709334/fivethirtyeightelections_2016-03-21-173834.64k.mp3?ad_params=zones%3DPreroll%2CPreroll2%2CMidroll%2CMidroll2%2CMidroll3%2CMidroll4%2CMidroll5%2CMidroll6%2CPostroll%2CPostroll2%7Cstation_id%3D4278Subscribe: iTunes |Download |RSS |Video

daniel: Other very important states to watch are New York and Connecticut: Can Trump reach 50 percent in those states to activate winner-take-all triggers? If he does, it could compensate for a disappointing performance in a state like Wisconsin. Also important, more obviously so: The candidates have little room for error in the winner-take-all states. Even a small Trump loss in Arizona or New Jersey, or a small Cruz loss in Nebraska or South Dakota, would have magnified consequences.

dave: Exactly, Daniel. If Trump were to win 51 percent uniformly across New York, he would claim 88 of 95 delegates, by my count. That would be enormous heading into the other “Acela” states the following week.

natesilver: It’s hard for me to see Trump losing New Jersey. Arizona, though? Early voting and Trump’s immigration stance will probably get him across the finish line, but I wish we had a more recent poll there.

harry: I want to talk about New York for just a second here. Trump should be very strong in the western part of the state (i.e., around Erie County), but most of the delegates won’t be awarded in that part of the state. Only 14 delegates are awarded statewide. The majority of the delegates will be awarded downstate and in New York City. We’ll really need to know the congressional delegate vote percentages to know how Trump did in New York.

natesilver: There are, what, 11 or 12 congressional districts that are mostly based in New York City proper?

dave: That’s about right, maybe a little higher.

natesilver: In 2012, some of those districts had fewer than 1,000 voters in the Republican primary! Granted, the race wasn’t competitive by the time New York voted four years ago. But if you’re living in the South Bronx — or even in Park Slope — your vote is REALLY high-leverage.

daniel: And the 50 percent winner-take-all trigger rules apply at the district level as well: If a winning candidate crosses 50 percent, he’ll get all three of a district’s delegates. If not, just two.

harry: I mean, who the heck are the Republicans voting in José Serrano’s district in the South Bronx?

dave: I’m not sure, but I’d hate to see their mailboxes in early April.

daniel: The same dynamic exists in Maryland, Washington and California — states where the bulk of delegates will be allocated by district, which makes these projections especially challenging!

natesilver: There’s one state that I think we’ve badly neglected: Indiana.

dave: Another state we haven’t really mentioned, but I think is a really big deal: Indiana. Jinx!

natesilver: It’s got quite a few delegates — 57, more than Wisconsin’s 42. Of those, 30 are statewide winner-take-all and 27 are winner-take-all by district.

daniel: And I’d add West Virginia: The state uses a loophole primary, and we saw in Illinois that loophole primaries can produce some unexpected results if voters don’t want to vote for a given delegate, as Dave

Elections Podcast: What Is Kasich Doing?

Our elections podcast crew discusses whether John Kasich is helping or hurting Donald Trump’s chances of winning the Republican nomination, as well as how the political tussle over Merrick Garland’s nomination to the Supreme Court will affect the election. Plus, FiveThirtyEight politics intern Anne Li describes the GOP’s shift in focus from “gay marriage” to “religious liberty.”

http://c.espnradio.com/s:5L8r1/audio/2709334/fivethirtyeightelections_2016-03-21-173834.64k.mp3?ad_params=zones%3DPreroll%2CPreroll2%2CMidroll%2CMidroll2%2CMidroll3%2CMidroll4%2CMidroll5%2CMidroll6%2CPostroll%2CPostroll2%7Cstation_id%3D4278Subscribe: iTunes |Download |RSS |Video

You can stream or download the full episode above. You can also find us by searching “fivethirtyeight” in your favorite podcast app, or subscribe using the RSS feed. Check out all our other shows.

If you’re a fan of the elections podcast, leave us a rating and review on iTunes, which helps other people discover the show. Have a comment, want to suggest something for “good polling vs. bad polling” or want to ask a question? Get in touch by email, on Twitter or in the comments.

Will Trump Clinch The GOP Nomination Before The Convention?

The Republican race for the presidential nomination is down to just one man and one number: Donald Trump and 1,237 — the number of delegates required to clinch the nomination. Can Trump win 1,237 delegates by the end of the primary season on June 7? Will he be forced to plunder among the more than 100 unbound or currently uncommitted delegates who will make the trip to Cleveland in order to win on a first ballot at the Republican National Convention? Or are we all but assured of a multi-ballot convention?

Any pundit giving confident answers to these questions is full of it, so FiveThirtyEight surveyed some of the best delegate obsessives and political experts we know on how many delegates they expect Trump to win in the remaining contests. Trump has 695 delegates now, and, on average, our respondents estimate he will still be just a little bit short of 1,237 on June 7, when California wraps up the primary calendar. He might be close enough, though, that he could clinch the nomination in the six weeks between California and Cleveland.

If that’s the baseline case, however, it wouldn’t take much for Trump to deviate from it in either direction. Outperform these estimates in California, for instance, and Trump could reach or surpass 1,237 delegates on June 7. Lose a winner-take-all or winner-take-most state where these estimates have him favored, however, and Trump could be well short of a majority.

There are few truly proportional states left on the GOP’s primary calendar, so small shifts in any state can have major consequences in the delegate count. That’s why we surveyed this group. My own personal1 delegate projection took a beating Tuesday because I thought Ted Cruz would win Missouri by a small amount instead of losing it by a small amount.

In addition to Nate Silver’s and my estimates, we tried to include a diverse set of opinions from across the political spectrum. We also focused on analysts who have been following the delegate race closely and know the intricacies of the GOP’s delegate rules. Our panel included these people:

Adam Geller: founder and CEO of National Research Inc. and the lead pollster for Chris Christie’s presidential campaignDaniel Nichanian: contributing editor to Daily Kos Elections and PhD candidate in political science at the University of Chicago.Henry Olsen: senior fellow at the Ethics and Public Policy CenterMargie Omero: managing director at Purple InsightsPatrick Ruffini: co-founder and partner at Echelon Insights and chairman and founder of EngageDavid Wasserman: House editor at the Cook Political Report and FiveThirtyEight contributor

To best reflect the group consensus, we used an “olympic average” in which we discarded the highest and lowest estimates in each state and averaged the remainder. Because of this averaging process — and because we told panelists they could list probabilistic forecasts instead of deterministic ones2 — in some cases the average reflects a number of delegates that it would be mathematically impossible for Trump to achieve. For instance, our average has Trump with 15 delegates in winner-take-all Delaware, which has 16 delegates at stake. This is equivalent to saying that Trump is highly likely but not quite certain to win Delaware, according to the panel.

Overall, our average response suggests that Trump will win 513 delegates the rest of the way. When combined with the 695 he’s won so far, that means he’d fall 29 delegates short of the 1,237 needed to win on the first ballot. Here is the olympic average for each upcoming contest (we’ve left out some contests with only unbound delegates; see the footnotes for more detail):3

DELEGATESDATECONTESTAVAILABLEPROJECTED FOR TRUMP3/22Arizona5858Utah404American Samoa924/5Wisconsin42254/9Colorado3474/16Wyoming1414/19New York95714/26Maryland3831Connecticut2819Rhode Island1910Pennsylvania1716Delaware16155/3Indiana57375/10Nebraska361West Virginia34335/17Oregon28125/24Washington44176/7California17293New Jersey5151South Dakota290Montana270New Mexico2410New Trump delegates513Delegates to date695Total1,208How many more delegates will Donald Trump win?

Trump projections based on an olympic average of estimates from Nate Silver, Harry Enten, Adam Geller (National Research Inc.), Daniel Nichanian (Daily Kos), Henry Olsen (Ethics and Public Policy Center), Margie Omero (Purple Insights), Patrick Ruffini (Echelon Insights) and David Wasserman (Cook Political Report).

If Trump does, in fact, get 1,208 delegates, he still might win on a first ballot. He would need only a fraction of the delegates that are currently unbound (or will be unbound) to reach 1,237.

Who exactly are these unbound or uncommitted delegates? Some, like the six from the Virgin Islands, were elected by voters to be “uncommitted,” but they may commit to a candidate closer to the convention. Others, like the 54 Pennsylvania district delegates, are automatically unbound and have been elected as unbound for decades (see: when Gerald Ford beat Ronald Reagan in the 1976 primary). These delegates are free to choose whichever candidate they want on all ballots.4 In addition, some delegates from candidates who have withdrawn from the race may become available to Trump, depending on the state’s rules. Although it’s hard to know Trump’s exact chance of getting 29 delegates from this group, Trump probably would have a decent shot at reaching 1,237.

All of the respondents agree that Trump is not likely to get close to 1,237 delegates before June 7, when California and four other states vote. The closest Trump came was 1,088 delegates. And even the most optimistic Trump projection has him hitting 1,244 after all the states have voted. That leaves Trump with very little room for error to reach a majority of delegates without at least some of the currently unpledged or uncommitted delegates coming to his aid.

Part of the reason we’ll have to wait so long is how the rest of the calendar breaks down. The month of April, which includes a lot of primaries in the Northeast, should be good for Trump. May has far fewer contests, and Trump is expected to do poorly in Nebraska, Oregon and Washington.

Not surprisingly, our respondents’ estimates differed greatly in a number of states. If you’re looking for the states that could be make-or-break for Trump, then look to Wisconsin, New York, Indiana and California. In all four, Trump’s expected number of delegates won differed by at least 36 among the respondents.

Wisconsin (April 5): Forty-two delegates are at stake, and it’s winner-take-all on the congressional district and state level. Trump led in the most recent poll, but with only 30 percent, and he had a very high unfavorable rating.New York (April 19): All of our respondents had Trump winning a majority of the state’s 95 delegates, but some believe the other candidates can cut into Trump’s edge by keeping him under 50 percent in a number of congressional districts or statewide. If Trump wins more than 50 percent in a district or statewide, he wins all delegates in that district or statewide.Indiana (May 3): It’s hard to say to whom Indiana’s 57 delegates will go because there hasn’t really been any polling there, and the Hoosier State doesn’t line up well demographically with any other Midwestern state.California (June 7): The biggest prize of all, California will award 172 delegates — 159 by congressional district and 13 to the winner statewide. No one knows how the very Democratic districts (and hence those with very few Republican voters) around Los Angeles or San Francisco will vote. The average statewide poll shows Trump ahead, but again with only 30 percent.

Indeed, there was a somewhat bimodal distribution in the total number of delegates our respondents expected Trump to reach. Three of us have Trump earning from 1,136 to 1,156, and three have him winning from 1,237 to 1,244. (The other two respondents have him in the 1200s but short of 1,237.) That may be why you read some pieces that seem to indicate that Trump is well on his way to winning on the first ballot, but other people seem to think there’s very little chance of it. Smart people disagree. And a single upset in a winner-take-all state could change the map significantly. Our panel has Trump winning all 58 delegates in winner-take-all Arizona, for instance, which votes Tuesday. If he lost there, it could make it very hard to get to 1,237.

Perhaps what becomes clearest is that there is still a lot we don’t know. We’ll have to watch and see if there are any clues in the voting over the next month. For now, hang on for a wild ride.

Listen to the latest episode of the FiveThirtyEight elections podcast.

http://c.espnradio.com/s:5L8r1/audio/2709334/fivethirtyeightelections_2016-03-21-173834.64k.mp3?ad_params=zones%3DPreroll%2CPreroll2%2CMidroll%2CMidroll2%2CMidroll3%2CMidroll4%2CMidroll5%2CMidroll6%2CPostroll%2CPostroll2%7Cstation_id%3D4278Subscribe: iTunes |Download |RSS |Video

Michigan State And The Biggest NCAA Tournament Upsets Ever

Having grown up a few minutes from the Michigan State campus, I’m a little bit upset about today’s upset. After two decades of Michigan State coach Tom Izzo’s amazing record in post-season play, we Spartan fans had grown to assume this was the sort of thing that only happens to Duke.

But No. 2-seeded Michigan State lost to No. 15 Middle Tennessee State in St. Louis in one of the most memorable upsets ever in the NCAA Tournament. Unlike several of the purported “upsets” earlier in the first round, where the worse-seeded team was actually the favorite, this one was a real shock. In fact, our March Madness predictions had given Michigan State a 95 percent chance to win, while Vegas had the Spartans as 16.5-point favorites.

So here’s a question I’m almost too embarrassed to ask: Was this the biggest NCAA tourney upset of all-time? The answer is that it isn’t quite, although Michigan State is possibly the best team to have lost its opening game.

According to our

No. 16 Seeds Are Due*

It’s the sort of statistic that seems ripped from pages of the Washington Generals media guide. Since the men’s NCAA Tournament went to a 64-team format in 1985, No. 16 seeds are winless: an imperfect 0-124 record.

No. 16 seeds can be pretty bad basketball teams, of course. Often, they’re teams from small conferences that won automatic bids by winning their conference tournament in a series of upsets after having barely cleared .500 during the regular season. (Small-conference teams that win both the regular season and their conference tournaments will usually wind up with No. 13, 14 or 15 seeds instead.) Furthermore, No. 16s have the misfortune of being matched up against No. 1 seeds, which are theoretically the four best teams in the country.

But being bad is one thing; going 0 for 124 is another. My hunch is that No. 16 seeds have been unlucky not to have pulled off at least one upset.

Consider that in the 1998 women’s NCAA Tournament, No. 16 seed Harvard (those plucky upstarts)