Last week John reported an updated estimate from our model giving the Republicans a 77 percent chance of taking control of the Senate. Beneath the surface of the overall estimate are some interesting changes — and continuity — across the electoral landscape. In short, there appear to be notably fewer potentially competitive elections, leaving the battle for party control of the Senate focused on eight states.
With the addition of first-quarter fundraising data, the number of elections that appear competitive has declined significantly. Our previous estimates (produced March 31) predicted 20 of the 36 elections favoring one of the parties, with estimated chances of winning greater than 90 percent. The updated estimates put 28 elections in this category. In 17, the estimated Republican chances are greater than 90 percent, and Democratic chances are greater than 90 percent in 11.
Compared with 16 in our previous set of estimates, there are just eight elections in which one party does not have an overwhelming advantage. Here we show the current estimates in those eight races and the previous estimates.
Assuming Republicans win the 17 elections in which we estimate their chances to be greater than 90 percent, they would need to win at least four of these eight elections to gain control of the Senate. Because we estimate their chances as better than 50-50 in five of the eight, with a sixth being a dead heat, our overall estimate gives them a solid chance (77 percent) of doing it.
Fundraising figures primarily explain the differences between the March 31 estimates and the current ones. For example, in Montana where the new forecast is more favorable to the GOP, Republican fundraising has been better than the Democrats’ by about two to one. By contrast, the forecast for Colorado is now more favorable to the Democrats. Although the Republicans got the candidate they wanted in Colorado, Rep. Cory Gardner, Republican fundraising has not yet made the election as competitive as it could be. The Democrats’ spending advantage has been 2-1, helping Mark Udall’s chances of reelection.
Our forecast in Iowa is also more favorable toward the Democrats, and we now estimate the election odds as a coin flip. Notably, just about everyone else sees an advantage for the Democratic candidate, Rep. Bruce Braley. As we have discussed, our estimate in Iowa is likely too pessimistic for the Democrats because the two leading candidates for the Republican nomination in Iowa (state Sen. Joni Ernst and businessman Mark Jacobs) both score lower on our measure of “candidate quality” than is typically observed among Republican open-seat candidates. If we assume Ernst wins the Republican nomination (she is the current poll leader), our estimate of Braley’s chances would increase to 59 percent. If the nomination goes to Jacobs, Braley’s chances would increase to 66 percent.
Finally, nothing in yesterday’s primaries in Georgia or Kentucky would alter the model’s prediction that the Republicans’ chances are better than 95 percent in both states. In Kentucky, as attention turns from the primary challenge to the GOP incumbent Mitch McConnell to the general election campaign with Democrat Alison Lundergan Grimes, we would expect to see McConnell develop a clear polling advantage. Likewise, although there is still uncertainty regarding who the Republican nominee will be in Georgia, our model would give them both better than a 95 percent chance of beating Michelle Nunn in November.
Election Day is still more than 150 days away, but with many Senate elections on their way toward being safe for one of the parties, the battlefield looks narrow and close.