GOP Primary Outcome Is in Kasich's Hands
If Donald Trump becomes the Republican Party nominee, the list of players who were indispensable to his rise will be lengthy: Jeb Bush, Chris Christie, Marco Rubio, and the news media. As one of my friends put it, the anti-Trump forces have always been about three weeks behind the front-runner: It was a 16-person field when it needed to be 10, a six-person field when it needed to be four, and so forth.
But assuming he makes good on his threat to continue his campaign to the Republican convention, John Kasich will ultimately play the biggest role. That is because the outcome of the race will probably be decided by whether this is a two-person or three-person race after the next couple of weeks.
Why? It goes back to this piece that I wrote in the wake of the “SEC primary.” Trump has benefited immensely from the fractured Republican field; he routinely trails or ties opponents when exit polls ask about head-to-head matchups between candidates. The race has largely fractured along lines of education and income, with Kasich likely pulling disproportionately from Rubio, and now Cruz.
As you can see from the first chart in that March 2 piece, Trump has, generally speaking, performed best among voters with a high school education or less, with Cruz having his best showing among voters with some college or college degrees, while Rubio and Kasich have performed best among voters with graduate degrees.
When we look at the race through the lens of income, Trump performs best among voters earning the least, Cruz with middle incomes, and Kasich and Rubio with the highest earners.
In other words, if we were to rank Rubio voters’ preferences for other candidates, we might reasonably surmise they would choose Kasich. (You can also see the demographic overlap between Kasich and Rubio using my colleague David Byler’s demographic calculator). This in no way means that all of Rubio’s voters will go to Kasich, or that there are no Trump voters in Rubio’s camp. All it means is that Kasich probably grabs the lion’s share of Rubio’s voters.
Now, if this had happened in the early phases of the race, it would be inconsequential. But in the late phases, where states are either winner-take-all or winner-take-all by congressional district, it is crucial.
To see how this plays out, I ran two different scenarios in our delegate calculator. I won’t give you the specifics, but the general idea is this: I generally gave Trump 40 percent of the vote, to Cruz’s 35 percent and Kasich’s 25 percent. In New England, I gave Trump 60 percent of the vote to Kasich’s 25 percent and Cruz’s 15. In West Virginia, I gave Trump 60 percent, Cruz 25 percent and Kasich 15 percent. I also skipped Colorado, North Dakota, and American Samoa, since their delegations are unbound.
Now, this obviously isn’t exactly how things will shake out in the real world – Trump would probably win more delegates in New York than the 60 percent estimate above suggests, but I’m not sure he would really run as well in Connecticut or Utah as I’ve assumed. There will probably be some Great Plains states where Cruz will run up the score. The point is to just set some sort of baseline, so that we can get a sense of how these things work. I then re-ran the scenario without Kasich, allocating 70 percent of his vote to Cruz and 30 percent to Trump.
The outcome is fairly stark. Under the first scenario, Trump wins 1,296 delegates and clinches the nomination on the last day of primary voting.
Under the second, Kasich-less scenario, however, Trump has 1,125 delegates, while Cruz collects 899. Given that under the second scenario, Cruz rattles off a string of wins at the end, and given the fact that Rubio’s and Kasich’s 300 delegates would probably disproportionately gravitate toward Cruz, this would likely be enough deny Trump the nomination.
Again, I don’t think this is precisely how things would play out. Trump would probably still win Arizona without Kasich, and would win more delegates in New York state, but I think Cruz would perform better on the West Coast and on the Plains (Trump has not performed well west of the Mississippi).
The interesting thing about the Republican race is that it slows down now. We have Arizona and Utah next week and Wisconsin in early April. That’s basically all we have between now and mid-to-late April, when the rest of the Northeast and the mid-Atlantic votes. If Kasich doesn’t bow out before then, Trump is probably going to be the Republican nominee. If he does, it’s still anyone’s game.