Today’s primary elections will tell us a lot more about whether Donald Trump is likely to win a majority of delegates to the Republican convention. In particular, the Florida and Ohio contests are key. Both states are “winner-take-all” and could therefore supply a treasure trove of delegates.
In our open election forecasting tournament with Good Judgment, we have been asking about both contests. In Florida, the forecast leans strongly to Trump, who is given a 71 percent chance of winning. This is even higher — 90 percent — among the elite group of “superforecasters,” whose forecasts are not published but were shared with me by Good Judgment.
In Ohio, however, Kasich is narrowly favored in the tournament, with a 50 percent chance of winning compared to Trump’s 44 percent. The superforecasters give Trump a smaller (33 percent) chance of winning Ohio.
The superforecasters have also made predictions in Illinois, Missouri and North Carolina, and they give Trump a solid chance of winning in all three states (75 percent, 60 percent and 70 percent, respectively).
If these forecasts prove true, Trump would clearly expand his delegate lead, even with a potential loss in Ohio. Nevertheless, the tournament forecasters do not consider his nomination a certainty: Currently, he has a 70 percent chance. This is about equal to what the prediction markets say. These markets have become a bit more bearish on Trump, although only a bit, in the past two weeks.
One reason why Trump is not a lock is that forecasters still see at least a small chance of a “brokered” convention. When asked whether a Republican candidate will win a majority on the first convention ballot, the tournament forecasters say there is a 26 percent chance that this will not happen. The superforecasters say there is a 16 percent chance this will not happen.
Interestingly, the probability of a brokered convention has been increasing among the tournament forecasters. The 26 percent figure is a median among the most recent 40 percent of forecasts. If you look day by day, the trend is clear: When the question debuted in January, the initial estimate was about 16 percent, and now it is about 32 percent.
What has not changed, however, is the forecast for the November election. Both the tournament forecasters and the superforecasters continue to give the Democrats better than 50-50 odds: 66 percent and 69 percent, respectively. This has been stable over time.
And, as I noted before, asking forecasters about the November election if Trump were the nominee leads them to give the Democrats even better odds: 75 percent.
In short, the forecasts today suggest Trump will proceed toward the nomination but then only hurt the Republican Party’s chances in the fall.