The GOP Race for Delegates: An Interactive Tool
Note: An updated version of the RCP Republican Delegate Calculator can be found here.
As is always the case, national polling is drawing the lion’s share of the attention in discussions about the Republican primary race. As is also always the case, this emphasis is misplaced. This is, after all, a race for delegates. These delegates are awarded in a series of state contests taking place over the course of some four months, and they are awarded under convoluted rules that vary wildly from state to state.
Which is a lengthy way of saying: The 2016 Republican primary campaign is heavily path dependent, rendering national polling of little value. The order of the contests, the structure of the rules, and the interplay between those two factors can play a huge role in selecting the eventual winner that national polls simply cannot illustrate.
To help make this easier to visualize and understand, we’ve created the widget below that will allow you to walk through the GOP selection process yourself. The process is fairly simple.
Beginning with Iowa in February and working through South Dakota in June, and the awarding of “RNC delegates,” you simply input the share of the vote that you think each candidate will receive. You can use the next state button to advance through the elections, and the previous state button if you feel that you’ve made a mistake. There’s a “reset” button at the bottom of the page if you want to start over.
We’ve pre-loaded the RealClearPolitics polling averages for the early states, but you’re free to alter these as you see fit. You can also use the current RCP national average or, in the later states, the inputs from the previous state as your starting point. As you go through, the program will calculate the statewide delegates that will be allocated using each state’s rules (plus a few assumptions) – you cannot alter these (except through changing the candidates’ share of the votes).
You can, however, alter the congressional district delegates. Many states allocate some of their delegates to the winners of individual congressional districts. The widget uses a default setting whereby it allocates the delegates proportionally to the candidates’ statewide shares of the vote that you have already input. This is probably a good rule of thumb, but it can nevertheless be a faulty assumption. For example, part of the way that Barack Obama survived Super Tuesday in 2008 was by winning with overwhelming strength in African-American districts in the South, allowing him to rack up delegates beyond the proportionality that the statewide contest would have assumed. Our advice is to leave the defaults in place unless you have a candidate running exceedingly well in a state (in which case you might want to increase the winner’s share of the congressional district delegates).
Beyond that, the program sometimes commits rounding errors; it may tell you that you have to add or subtract a delegate even if you haven’t changed any of the default results. We believe that we have caught most of these, but two people won’t find everything that 10,000 will. (Let us know if we’ve missed any!)
The results of races are tracked at the top, with one row beneath the candidate pictures for delegates won and another for contests won (remember, a candidate needs eight wins to have his or her name placed into nomination at the convention). You can drop candidates out along the way, but the feature is disabled during multi-primary days (e.g., you can’t have someone quit in the middle of the “SEC primary”) – states are marked with thick black borders on the day that they are in play, so if multiple states have thick black borders, then it is a multi-state Election Day. If you make a mistake, you can drop candidates back in.
That’s really all there is to it. We’re sure you’ll find lots of interesting outcomes – we’ve had a blast playing around with it. We’ve included some of our more interesting findings below the widget. Note: An updated version of the calculator can be found here.
1. The importance of whether the party decides: A lot of the debate surrounding the Republican primary calendar has focused on what we might call the “Party Decides Hypothesis.” This dates back to an important political science text, which finds that party officials (broadly defined) exert outsized influence on primaries and usually manage to bring seemingly chaotic contests to an orderly close.
The debate centers on whether this time might be different, given the strength of some of the insurgent candidates, the rise of Internet fundraising, and the ability of super PACs to keep candidates going after a few losses. We saw this to some degree in 2012, when Newt Gingrich and Rick Santorum – not overwhelmingly strong candidates – managed to stretch the contest out through extended periods where they did not win any races. If they could do this, imagine what a Ted Cruz or Donald Trump might do. The counter is that “this time it’s different” has a terrible track record, so we should probably look askance at theories that depend on radical changes to our understanding of how primaries operate.
Regardless, your answer to the question “Is this time different?” -- as well as your answer to the question “If so, how different is it?” -- have substantial implications for the outcome of the race. If things aren’t different, we should probably only see perhaps three candidates emerging from South Carolina, and perhaps two coming out of the SEC primary.
But if things are different, it isn’t that difficult to see multiple candidates trying to make it to the delegate rich winner-take-all states at the end, or perhaps staying in to try to deny a candidate the eight wins needed to place his or her name in nomination.
At the very least, we do think it’s unlikely that the old maxim about three tickets out of Iowa and two tickets out of New Hampshire will hold this time. Under the current RCP averages, after New Hampshire Donald Trump will have 22 delegates, Ben Carson will have nine, Marco Rubio will have six, Ted Cruz will have three, Jeb Bush will have two, and Carly Fiorina, John Kasich, Chris Christie, Mike Huckabee and Bobby Jindal will each have one. That’s not much separation.
2. South Carolina is really important: While most of the media attention has focused on Iowa and New Hampshire, perhaps the most important early state, assuming the field remains crowded, is South Carolina. Unlike the other early states, this one is winner-take-all by congressional district, meaning that a candidate that runs well statewide could sweep the state’s delegates. Given that there are relatively few delegates that will have been awarded by that point, a candidate who won, say, 42 delegates in the Palmetto State could wind up the delegate leader for the next month or so.
3. Backdoor winner-take-all is overstated: Some have theorized that there could be a number of supposedly proportional states that evolve into so-called “backdoor winner-take-all” elections. It would work like this: Many states have minimum thresholds for allocating delegates, so a candidate would have to win, say, 20 percent of the vote to win any delegates.
In a crowded field, it is entirely possible that only one candidate would win 20 percent of the vote, taking all of the delegates. But most states have realized this, and have adjusted their laws such that the top two vote getters are guaranteed delegates regardless of the cutoff. We only find a few scenarios where “backdoor winner-take-all” occurs, and those involved fields that we kept improbably crowded.
4. Southern states winnow: One place that we can confirm the conventional wisdom is the idea that Southern states will do the winnowing. The early primaries are overwhelmingly – though not exclusively – held in Southern states. Interestingly, the exceptions tend to be Northern caucus states that have been surprisingly fertile ground for insurgent candidates; Rick Santorum ran well in many of these places 2012.
The upshot of this is that a very conservative primary electorate will be eliminating candidates from the field early on. What you think the consequences of this will be plays a huge role in how your simulation ultimately plays out.
5. Northern states decide: The corollary to this, however, is that Northern states with more moderate electorates will decide who the nominee will be. You can see a good discussion of this from Dave Wasserman at FiveThirtyEight here. This is similar to the phenomenon that Sean discussed in 2011: These late states are better suited for establishment candidates, so unless an insurgent candidate opens a huge delegate lead in the early states, the establishment would have a good chance at mounting a comeback.
Many of the Northern states are winner-take-all, meaning that the establishment candidates can close the gap quickly. For example, Florida’s delegates (and let’s be honest, Florida is an honorary Northern state these days) are awarded on a winner-take-all basis. This might work to the advantage of Jeb Bush or Marco Rubio (depending on the trajectory you expect for Donald Trump’s campaign). Ohio will award its delegates the same day, and so forth. In general, we tended to find that the surviving campaigns had only a couple hundred delegates by the time we got to mid-March, but by the time April rolled around there were usually one or two campaigns in the 500-600 range. (To win the nomination, 1,236 are needed.) At the same time, we note that the popularity of awarding delegates by congressional district means that many states that are thought to be winner-take-all, such as California, really are not.
This provides interesting incentives for campaigns. Mike Murphy alluded to this in a controversial interview with Bloomberg Politics, where he predicted that a campaign that had failed to win a single state could suddenly become the delegate leader in March. We could – and we emphasize could – therefore see a surprising number of campaigns attempting to soldier through the early states in hopes of posting big wins in mid-March.
6. The order in which candidates drop out matters. A lot. It didn’t surprise us that the order and rate of dropouts mattered. What surprised us was just how much it mattered. Consider a situation where Ted Cruz, Donald Trump, and Jeb Bush are the only candidates who emerge from the SEC primary in early March. Bush then has a huge advantage in the race that ensues, because he does well in the remaining proportional states and then cleans up in the late states while Cruz and Carson split the Tea Party vote.
However, if it is Cruz, Trump, Bush and Rubio, the dynamic is quite different. Under this scenario, it is very, very difficult to amass a huge win, and the possibility of a brokered convention increases dramatically. Of course, if one of the insurgents knocks out the other insurgents in the early races, yet two or three “establishment” candidates survive, that candidate could end up victorious in the winner-take-all states, thereby ending the race quickly.
7. RNC delegates matter: It’s often said that, unlike the Democrats, the GOP does not have “superdelegates” – that is, unpledged delegates who can vote for whomever they want. But this is not, strictly speaking, true. Each state is awarded so-called “RNC delegates,” who are party officials with automatic credentials to the convention. This presumably gives an establishment candidate a leg up; but if an insurgent won the pledged delegate county and the popular vote, the pressure to back that candidate would be immense (something similar happened with Barack Obama in 2008, although he arguably did not win the popular vote).
8. Rubio/Cruz: With that said, we did tend to find that one much-discussed outcome emerged repeatedly in our simulations: A race to the end between Ted Cruz and Marco Rubio, with Rubio holding an advantage. Perhaps this merely reflects our internalized opinions regarding how things play out. But based on current polling and the state of fundraising today, we found that Cruz, Trump and Carson were all able to rack up a fair number of delegates after the Nevada caucuses, while Marco Rubio was left as the leading establishment candidate.
Cruz then tended to perform well in the SEC primary, given the huge number of delegates awarded to the state of Texas (155). Cruz maintained his large lead going to the winner-take-all states, at which point Rubio usually mounted a comeback, followed by a barnburner of a race.
But as we noted, there are an almost infinite number of possibilities. We hope you enjoy this as much as we have, and that your understanding of the real Republican primary race – the race for delegates – is enhanced by a couple of walks through the widget.