For much of the last year there were two things the vast majority of political prognosticators—including those here at Slate—could agree on: 1) Donald Trump would not be the Republican nominee, and 2) there would not be a contested convention in Cleveland. With half of the Republicans delegates now awarded, it now seems one of the two has to be wrong.
So what would a contested convention actually look like? How would it work? How did we get here? You have questions; we’ve got answers.
Let’s start at the very beginning. What exactly is a contested convention?
Definitions vary but, generally speaking, a convention is considered contested when it begins with the ultimate nominee still in doubt. Typically, that means no single candidate has locked up a majority of delegates—the threshold needed to win the nomination—by the time the political festivities get underway. In a broader sense, though, a contested convention can happen even if someone does have the necessary majority of delegates as long as one of his or her rivals is unwilling to give up before the convention begins—as was the case in 1980 when Ted Kennedy mounted an unsuccessful last-ditch effort to snatch the Democratic nomination away from then-President Jimmy Carter via a rule change that would have freed pledged delegates to vote any way they wanted on the floor.
Is a contested convention the same thing as a brokered convention?
The terms contested and brokered are often used interchangeably when talking about conventions, but in reality the latter is now an anachronism since state party leaders no longer wield the power they’d need to do any large-scale brokering.
But they used to?
Yes. Back in the days before the modern primary system (which began in the early 1970s, roughly speaking) the bulk of delegates from a state were selected directly by the local party apparatus. Many had formal arrangements with their state leaders to vote as instructed, while the rest normally did so out of either party loyalty or their own self-interest. That kept things relatively simple for candidates looking for votes on the convention floor—promise the New York party chairman he’ll be your ambassador to France, and gain the support of the New York delegation. But now that most delegates are selected by either the candidates or via elections at local conventions, they have much less incentive to act in concert as a single state delegation.
OK, back to more recent history. Did you just say Ted Kennedy tried to change the rules at the convention? Can they do that?!
Yep! Aren’t politics great? When you hear that “anything could happen” at a contested convention, this is a big reason why—anything really is possible since the convention rulebook can be changed before the delegates have the chance to select a nominee. Remember: Conventions are party-run affairs, not government-run elections. We’ll get into just how crazy things could theoretically get, but for now it’ll be easier if we focus on how things would most likely play out assuming there are no major rule changes—which would still be plenty crazy.
So what happens if Trump or someone else hasn’t locked up a majority of the delegates before the convention starts?
In a word: chaos. In a few more: Roughly 95 percent of the 2,472 GOP delegates will arrive at the convention “bound” to a candidate—that is, they will be required by their state party (or, in a few cases, state law) to vote for the candidate they were assigned to as a result of their state’s primary or caucus. (The other 5 percent arrive “unbound” and can vote however they like; we’ll get to them later.) That means the first vote tally should look more or less like the delegate trackers kept by the Associated Press and other news outlets. But if no candidate wins a majority of delegates during the first round—which they wouldn’t in this scenario—many of those delegates would then be unbound and allowed to vote however they want during the second round. Even more would become free agents in the rounds that followed, until eventually we’re looking at a free-for-all that would make the Republican debates look like ordered affairs.
Is there time between rounds for negotiations? Or do the votes just come one after another in rapid fire?
Conventions run on a slightly modified version of Robert’s Rules of Order, which means that decisions like the timing of the votes will likely be decided by the convention chairman and by the delegates themselves. The most likely scenario, though, is that there will be plenty of downtime between each vote in order to allow candidates, party officials, and delegates to cobble together a big enough coalition to select a nominee.
How many rounds of voting could we be looking at?
As many as it takes to come to a consensus. In 1924, it took Democrats 103—yes, 103!—ballots to finally settle on John W. Davis as a compromise nominee following a protracted fight between frontrunners William McAdoo and Al Smith. Unsurprisingly after that mess, Davis did not win the general.
Is a winner of a contested convention always doomed in the general election?
Not always, though history hasn’t been kind to most of them. By the Pew Research Center’s count, in the past 150 years, of the 10 nominees who needed multiple ballots and then went on to face a candidate who didn’t in the general election only three went on to be elected president. (On four occasions, both party’s nominees had to survive a contested convention.) The last winner was Franklin Roosevelt in 1932. Since then there have been three multiballot candidates who lost the general: Wendell Willkie in 1940, Thomas Dewey in 1948, and Adlai Stevenson in 1952. Candidates who had to fight for their nomination at the convention—either because they arrived a few delegates short of a majority or because their challenger was plotting a rules-committee Hail Mary—but who nonetheless went on to win on the first ballot have likewise struggled in recent decades: Gerald Ford found the extra delegates he needed to hold off Ronald Reagan at the convention before losing to Jimmy Carter in the 1976 general; Carter survived Kennedy’s rule-rewriting gambit before losing to Reagan in 1980, and Walter Mondale rounded up the delegates he needed to withstand a brief challenge from Gary Hart before losing to the Gipper in 1984.
Has anyone other than Davis arrived at the convention without the most delegates to their name but still ended up as the eventual nominee?
Yes, such come-from-behind victories weren’t unheard of before the modern nominating systems were put in place—back when far more delegates arrived still up for grabs. Willkie, for example, placed third on the first ballot at the Republican convention in 1940 before eventually winning the nomination on the sixth ballot; Stevenson was in second place after the first ballot at the Democratic convention in 1952 before winning the nomination on the third after then-President Harry Truman (who had decided not to run for another term) offered his endorsement. Like Davis, though, neither man would go on to win the general election.
So a comeback victory at the convention always ends in November defeat?
No—but you have to go even further back in the history books to find a success story. Since 1900, only two men who were trailing after the first ballot at their convention eventually won the nomination and then the White House: Woodrow Wilson, who needed 46 ballots to win the Democratic nomination in 1912, and Warren Harding, who clinched the Republican nomination on the 10th ballot in 1920.
History, so great. Now, tell me more about how bound delegates become free agents.
Exactly which ones will be freed and when depends on a rather complicated set of state party rules—which, like the national ones, are not written in stone—but by the New York Times’ count, the number of unbound delegates would grow from 5 percent during the first round of voting to 57 percent in the second, and then to 81 percent in the third. Even those are rough numbers, though, since states can unbind their delegates if the candidate they were assigned to vote for withdraws from the race or fails to meet certain vote thresholds on the convention floor.
Wait, but why would a Trump delegate defect to another camp? Trump fans don’t seem like the type of people who change their minds about Trump.
True. But while the bulk of delegates arrive in Cleveland bound to a particular candidate, that doesn’t mean they necessarily personally favor that candidate. A Trump delegate could very well loathe Trump; a Ted Cruz delegate might prefer John Kasich. And so on. Some states allow candidates to select the delegates that will represent them—making them more likely to remain loyal even once unbound—but those delegates only account for about 14 percent of the total. The vast majority of the rest are selected during state or district conventions that are held well after the actual state primaries or caucuses. Those slots tend to be filled with rank-and-file Republicans who are involved with the state and local party, making them theoretically more open to the GOP establishment’s anybody-but-Trump entreaties than primary voters have been.
So once a delegate becomes unbound, who can he or she vote for?
Anyone else whose name has been formally placed into nomination—be it someone who is already an official candidate this year, like Cruz, Rubio, or Kasich, or someone who isn’t, be it Mitt Romney, Paul Ryan, or whoever else the GOP poobahs try to cast as a white knight.
How does someone get his or her name placed into nomination?
We’ll get to that later.
OK. What about those delegates you mentioned before who arrive unbound? Are they the same as superdelegates?
Republicans don’t use the term superdelegate, but that’s effectively what these unbound-from-the-outset delegates are. Most of them are state party officials—the Republican National Committee allots each state party three slots to fill with its own members, most of whom are free to vote as they wish—while others are regular delegates from states and territories, like Colorado and Guam, that opt not to hold presidential preference votes during primary season, which allows them to send their pre-allotted number of delegates to the convention unbound. These lucky delegates don’t have to wait for a round-one deadlock to start wheeling and dealing.
So how the heck will Republicans ultimately settle on a nominee?
We’ll find out! Decades ago, back before the modern primary system went into place, state party bosses could cut deals with a candidate or interest group to deliver their state delegations en masse to a specific candidate. That’s not to say there won’t be favor-trading and arm-twisting this time around. An unbound delegate can be pressured, convinced, or enticed to back a particular candidate who might not be their his or her choice, but ultimately they get to make the final decision for themselves—which means a candidate looking to cobble together a majority will have to hand out a lot of favors, as opposed to in the past when they only needed to dole out a few big ones. When it comes time to vote, meanwhile, a county treasurer from rural Nebraska will have just as much say as the head of the Ohio Republican Party.
What about a unity ticket? Could Ted Cruz and Marco Rubio team up on a ticket and pool their delegates to defeat Trump?
Yes and no. Rubio could “release” his delegates and urge them to vote for Cruz (say in exchange for the VP slot on the ticket)—but Rubio’s request would just be a recommendation. Once a bound delegate is released, he or she becomes a free agent like the rest of the unbound voters. Likewise, national leaders—like RNC Chairman Reince Priebus, House Speaker Paul Ryan, and Senate Majority Leader Mitch McConnell—could put their weight behind that type of unity ticket, but they couldn’t compel it to happen on their own. They’d also need to be careful if they want to avoid looking like they personally have subverted the will of all those GOP primary voters who cast ballots for Trump.
So Trump would definitely lose a contested convention?
Not necessarily. If he arrives at the convention only a few delegates shy of the 1,237-delegate majority needed to win the nomination, he could still pick up the remainder from those unbound delegates in the first round of voting, as Ford did in ’76 and Mondale did in ’84. While many of the delegates who arrive unbound will be state party officials eager to block Trump, it stands to reason that there will be at least a few Trump confederates in their ranks. Trump could also entice any of those who are on the fence to deliver him the win in exchange for any number of things—from the promise of a plum job in his administration to a ride on his personal helicopter. As the Donald doesn’t hesitate to remind anyone, he loves to deal. A contested convention would give him the chance to prove it. But the more rounds of voting it takes for someone to win a majority—and the more delegates that become unbound—the more likely that the eventual nominee is anybody but Trump, since for all the competing interests that will emerge, the big battle would be between Trump and everyone else. The GOP front-runner knows this, which is why he wants to lock up enough delegates ahead of time to make this whole discussion moot.
But if “everyone else” can’t decide on a single alternative, wouldn’t that work in Trump’s favor? The non-Trump candidates would split the vote at the convention just like they’ve been doing in the primaries.
They could, but only temporarily since a split-vote status quo would simply prolong the existing stalemate. Trump’s rivals could continue to split the vote indefinitely but that wouldn’t hand the nomination to Trump since he’d also be short of a majority. The delegates would just keep voting until eventually, exhausted but not quite defeated, they consolidate around a single candidate. Unlike the primaries, Trump won’t have the luxury of running out the clock while his rivals fight amongst themselves for the non-Trump vote. There is no clock.
OK, thanks for laying out what will happen if Trump doesn’t win 1,237 delegates before the convention. Now, back to these rules that can apparently be rewritten willy-nilly. Are you really saying that even if he Trump does win 1,237 delegates before the primary season is over, the GOP could still deny him the nomination at the convention?
Conceivably, yes. In a normal year, the RNC rules committee makes a few relatively minor convention tweaks that the delegates then sign off on, but the major nomination-deciding rules stay the same. This, though, is not a normal year. (The front-runner pulled out steaks at a victory speech—not normal.) In theory, the rules committee could craft language that would do what Ted Kennedy tried and failed to do at the Democratic convention four decades ago: Unbind the national delegates from the candidates they were sent to the convention to vote for. The delegates would still need to vote to approve that change, but—if they did—it could deny Trump the nomination, since his 1,237-plus delegates would no longer be under any obligation to vote for him. (Delegates from some states—such as Arizona—might still be bound by state laws that require them to vote for their assigned candidate on the first ballot, though some legal experts doubt such laws would ever be enforced. Regardless, in a tight race, even a few dozen defections from Trump’s camp could be enough to ensure a second ballot.) If such a rule were put in place, the voting would then begin with all of the delegates free to back whomever they want during the first round and each one that followed until someone won a majority of votes.
But wouldn’t that be a PR disaster for the Republican Party?
Oh, god, yes. Which is why as desperate as the GOP establishment is to stop Trump, it’s unlikely they would try to deny him the nomination if he arrives in Cleveland with the necessary majority of delegates. The more realistic scenario, though, is the one we’ve already discussed: Trump arrives with a plurality of delegates but not the majority needed to win the nomination outright. At that point, the rules committee would go to work on smaller changes designed to undermine Trump’s chances of surviving the floor flight that would ensue. That would still risk a backlash from Trump’s legion of fans—and potentially other Republican voters who would be turned off by parliamentary gamesmanship—but the GOP establishment would find it much easier to justify using small rule tweaks to their advantage if Trump doesn’t have a majority of delegates than they would rewriting them completely to get their way if he does.
Give me an example of how they might tweak the rules.
The rule that has probably gotten the most attention is Rule 40, which requires a candidate to have the support of the majority of at least eight state delegations in order to have his or her name placed into nomination. That rule was written in 2012 to deny Ron Paul supporters the chance to stage a protest in prime time; previously the nomination threshold only required that a candidate have the backing of a plurality of five state delegations, a significantly easier task. The eight-state rule, though, will almost certainly be rewritten at a contested convention. Otherwise, it is possible that Trump—and Trump alone—would be the only name on the first ballot, in effect handing him the nomination. The rules committee will want to tweak that rule to include however many of Trump’s rivals would create the greatest chance at a deadlock in the first round or two. It could lower the threshold considerably—say, by requiring that a candidate only needs to have the backing of a plurality of a single state or territory, which would allow Cruz and Rubio to both qualify, as well as Kasich if he wins his home state of Ohio—or they could replace it with some other qualifying standard that was reverse-engineered to ensure their chosen candidates could meet it, while others could not. (Alternatively, they could scrap it all together.)
What else could the party do to undermine Trump?
The rules committee would be the best venue for Trump’s rivals to gain an advantage, but they could also try to win smaller skirmishes in a different committee, the one that formally credentials delegates. When you see delegate estimates that say Donald Trump won seven delegates in Iowa or 36 in Alabama, those are only provisional figures based on the often-convoluted formulas some states use to award delegates at their state conventions that come later. But the credentialing committee gets the final say and could decide those numbers should be a delegate or two higher—or, much more likely in the case of Trump, lower. On its own, that wouldn’t be enough to derail him, but if you string enough of them together it might make a difference in a tight race.
These options do not seem good.
Correct. Recent polling suggests that roughly half of GOP voters don’t want to see a contested convention—and it stands to reason that number would climb once the nation watches the anybody-but-Trump sausage made in prime time. For as much as the GOP establishment fears that a Trump nomination would tear its party apart, denying a man who won the most states, delegates, and votes risks the same outcome.
What really are the chances of this happening, though? I feel like I read a “there might be a brokered convention” piece every four years.
Not quite, but you’re not far off. Of the 11 open primaries that have occurred between the two parties since 1984, there has been at least some relatively serious talk of a contested convention in more than half of them: in 1984, 1988, 1992, 2004, 2008 (in both races), and 2012. Such speculation, though, never came to fruition.
So why should I take this year’s chatter seriously?
The biggest reason to take it seriously is that so many Republicans are now working to make it happen. Mitt Romney and other GOP elder statesman have publicly endorsed the strategy as the last, best hope to stop Trump from becoming the nominee. To varying degrees, so too have all three of Trump’s remaining rivals. It’s far from a guarantee, but the primary calendar and the delegate estimates suggest it’s certainly possible.
You got any math to back that up?
As of Tuesday night, Trump had won about 44 percent of the delegates that have been awarded (while winning about 35 percent of the vote). Those pluralities give him a commanding lead on his rivals: He’s up roughly 100 delegates on Cruz, 200 on Rubio, and 400 on Kasich. While that lead may soon be insurmountable—as in, his rivals can’t pass him—it doesn’t guarantee he’ll reach the magic number of 1,237.
The two biggest tests on the map will be Florida and Ohio, both of which hold delegate-rich, winner-take-all primaries next week. But even if Trump wins both of those states, he’d still need to win roughly half of the remaining delegates available in the remaining contests. That’s certainly doable—particularly given Rubio and Kasich’s support would likely dry up if they lose their home states—but it’s hardly assured.
If Trump loses one or both of those states, though, the math gets considerably more difficult for him. According to NBC News’ estimate, he’d need to win nearly 60 percent of the remaining delegates if he wins Florida but loses Ohio, roughly two-thirds of the remaining delegates if he loses Florida but wins Ohio, and nearly 70 percent of the remaining delegates if he loses both. Those would be heavy lifts—even with a few winner-take-all contests on the map.
Does anyone else have a viable path to 1,237?
Not really. Rubio’s campaign is collapsing. Kasich’s has lasted longer than anyone expected, though even if he wins his home state of Ohio next week, it’s unclear where his next victory would come. Cruz, meanwhile, has proved to be Trump’s toughest challenger, but he’s already missed out on his best chances for big wins in the Deep South—where he’s theoretically strongest—and is unlikely to be the favorite in the winner-take-all contests that dot the primary calendar after next week. Unless something major happens in the next few weeks, their only realistic hopes would be at the convention.
JUST TELL ME WHAT IS GOING TO HAPPEN.
I don’t know.
I’ll repeat that: I don’t know.
While all three of Trump’s rivals would prefer a contested convention than one where Trump wins easily on the first ballot, the former doesn’t guarantee that one of them will end up as the nominee. Rubio was once seen as the party’s savior, but his stock has dropped so low that some of his allies reportedly want him to drop out before Florida to save face. Kasich has stepped in to fill some of the establishment vacuum that Rubio’s implosion has created, though he seems more likely to be on everyone’s VP list than the guy they want in charge. Cruz, meanwhile, could make a case that he deserves the nomination if he’s able to keep the delegate race relatively close—but he’d still have won fewer delegates than Trump, making it an awkward sales pitch both to the delegates and the public. And, if the party big wigs really do get their way at a contested convention, it’s hard to imagine they’d be eager to use the opportunity to back a man they so clearly hate when they could pick pretty much anyone else.
That leaves the two men most often named as possible white knights: Mitt Romney and Paul Ryan. Romney has been careful to leave the door open while attacking Trump. Ryan, meanwhile, maintains that he has no interest in being the nominee—but he said the same thing about being the speaker of the House before finally relenting. Either could emerge as a compromise candidate if Cruz and co. are able to derail Trump on the first roll call vote. It’s also possible that some dark horse emerges halfway through the convention and rides off with the nomination. Again: No one knows!
I’m still a little confused. How would Ryan or any other noncandidate get their name placed in nomination? What about that eight-state rule you mentioned?
That will be up to the rules committee to decide. But even if some version of the eight-state rule remains in effect—and, again, it might not—all Ryan, Romney, or anyone else would need to do would be to get a majority of delegates in the necessary number of states to pledge their support prior to a roll call vote, at which point they’d be eligible for that round of voting. Remember, most delegates will arrive at the convention bound—which should stop noncandidates from being nominated before the first vote—but once enough of them become unbound, they’d theoretically be free to declare their support for someone new. As currently written, Rule 40 doesn’t require a candidate to have won eight states during the primary season—or even to have competed in them—it only requires that a majority of delegates from those states declare their support in writing one hour before the roll call vote in question. Again, the rulebook is written in such a way that it’s open to interpretation; it’ll be the rules committee and the delegates that decide how they want to do the interpreting. If enough Republicans want a white knight to slay Trump, they’ll find a way to get him to the floor.
How would Trump handle a contested convention?
Um, not well. Trump would be unlikely to go down without a fight and could even try to challenge the convention outcome in court. The man likes to sue. Meanwhile, he’s been threatening from the outset to launch a third-party bid if the GOP treated him “unfairly,” and this would seem to hit his rather loose definition of that. While “sore loser” laws in states like Ohio and Michigan—which prevent a nominee from appearing on the general election ballot for one party if they were on the primary ballot for another—would likely doom his independent bid, his campaign would also likely doom the Republican candidate’s, which may be consolation enough for Trump. Even if he doesn’t run, it’s impossible to imagine him endorsing the eventual GOP nominee and easy to imagine him telling his legions of fans to stay home in November.
Wow, that’s a lot to take in. I don’t suppose you can point me in the direction of a particularly great quote that sums this all up?
I thought you’d never ask! Here’s how H.L. Mencken described the Democrats’ never-ending 1912 convention, which he covered for the Baltimore Evening Sun (h/t Politico’s Jack Shafer):
For there is something about a national convention that makes it as fascinating as a revival or a hanging. It is vulgar, it is ugly, it is stupid, it is tedious, it is hard upon both the higher cerebral centers and the gluteus maximus, and yet it is somehow charming. One sits through long sessions wishing heartily that all the delegates and alternates were dead and in hell—and then suddenly there comes a show so gaudy and hilarious, so melodramatic and obscene, unimaginably exhilarating and preposterous that one lives a gorgeous year in an hour.
Cleveland 2016: vulgar, ugly, stupid—but unimaginably exhilarating!