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 campaign
- Daniel 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 Center
- Margie Omero: managing director at Purple Insights
- Patrick Ruffini: co-founder and partner at Echelon Insights and chairman and founder of Engage
- David 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
| DELEGATES | |||
|---|---|---|---|
| DATE | CONTEST | AVAILABLE | PROJECTED FOR TRUMP |
| 3/22 | Arizona | 58 | 58 |
| Utah | 40 | 4 | |
| American Samoa | 9 | 2 | |
| 4/5 | Wisconsin | 42 | 25 |
| 4/9 | Colorado | 34 | 7 |
| 4/16 | Wyoming | 14 | 1 |
| 4/19 | New York | 95 | 71 |
| 4/26 | Maryland | 38 | 31 |
| Connecticut | 28 | 19 | |
| Rhode Island | 19 | 10 | |
| Pennsylvania | 17 | 16 | |
| Delaware | 16 | 15 | |
| 5/3 | Indiana | 57 | 37 |
| 5/10 | Nebraska | 36 | 1 |
| West Virginia | 34 | 33 | |
| 5/17 | Oregon | 28 | 12 |
| 5/24 | Washington | 44 | 17 |
| 6/7 | California | 172 | 93 |
| New Jersey | 51 | 51 | |
| South Dakota | 29 | 0 | |
| Montana | 27 | 0 | |
| New Mexico | 24 | 10 | |
| New Trump delegates | 513 | ||
| Delegates to date | 695 | ||
| Total | 1,208 | ||
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 politics podcast.