House forecast: Democrats will win 229 seats (and the House majority) while Republicans will win just 206 seats. A Democratic win of 205 seats and 262 seats is within the margin of error.
Senate forecast: Republicans will hold 52 seats (and maintain control of the Senate) next Congress while Democrats will hold just 48. Anything between Republicans holding 47 seats and 57 seats is within the margin of error.
You’ll notice the wide margins of error with our Senate estimates. This includes our overall estimates with the Republicans forecasted to win 52 seats, but them winning as few as 47 seats and as many as 57 seats. That may seem nuts, though it really isn’t when you look at how many close races there are.
Right now, there are an amazing nine races that are within 6 percentage points. These include Arizona, Florida, Indiana, Missouri, Montana, Nevada, North Dakota, Tennessee and Texas. Democrats are currently favored in five of these races: Arizona, Florida, Indiana, Missouri and Montana. Republicans are currently favored in four: Nevada, North Dakota, Tennessee and Texas.
There are an additional five Senate races where one side is heavily favored, but where a win by the other side is within the margin of error: Minnesota’s special election, Mississippi’s special election, New Jersey, West Virginia and Wisconsin. These races are within the margin of error, though where one of the candidates is favored by at least 9 percentage points. All but Mississippi are favored to go to the Democrats.
Let’s walk through these races and see what happens in three broad scenarios.
Everything happens the way we think it will: let’s say each side won all the races they were favored in, then Republicans would have 51 seats to Democrats’ 49 seats.
You’ll notice this 51-to-49 split is one less less favorable to Republicans than the overall estimate put out by the model. The main reason for that is because Democratic leads in Arizona, Florida, Indiana and Missouri are all at 3 points or less. Only one of the Republican advantages is that small (Nevada). Based upon past elections, the model thinks that the Democrats are more likely than not going to lose one of those seats.
Keep in mind: it is likely that the next Senate won’t be split 51 Republicans to 49 Democrats or 52 Republicans to 48 Democrats. These are just two of the more likely scenarios that will occur.
A Senate in which Republicans control anywhere from 50 to 53 seats is not only within the margin of error, but well within it.
The Democrats have a good night: Let’s say Democrats won all the races within 6 points. This means they take all the seats they are favored in plus Nevada, North Dakota, Tennessee and Texas. That would give them 53 seats to the Republicans 47 seats. That’s pretty much their best case scenario.
There is a very low probability (not within the 95% confidence interval) that they win all of these and the Mississippi special election. That election involves a strange jungle primary in which Republican Sen. Cindy Hyde-Smith and Republican Chris McDaniel are facing off against Democrat Mike Espy. If McDaniel and Espy go to a runoff, Democrats have a real chance. Right now, it is highly likely that Hyde-Smith and Espy advance to a runoff in late November. Hyde-Smith is favored by 13 points in that runoff.
More likely than any of these scenarios, but still good for Democrats, is if they win all the seats they are favored in plus Nevada and either the Mississippi special election, North Dakota, Tennessee or Texas. Such a result is within the margin of error, though the chance of it occurring is shrinking by the day.
It would be interesting though if Democrats do win all the seats they lead in, Nevada, and McDaniel and Espy advance to a runoff. That would mean the Senate majority would be decided in late November runoff.
The Republicans have a good night: One might argue that having a good night for Republicans is them maintaining their majority. I’ll push it one step further and say they have a significant seat gain.
That would happen if Republicans win all the races within 6 points. This means winning all the races they are favored in plus Arizona, Florida, Indiana, Missouri and Montana. That gets them to 56 seats overall. That’s not a particularly wild scenario. Arizona, Florida, Indiana and Missouri are all within 3 points. They could easily go the other way. Remember: President Donald Trump won all of these states in 2016. We’ve also had a lack of public polling data in Montana. There is some internal Republican polling to suggest the race there is closer there than we currently believe.
Winning even three of these five races fall into the Republicans hands would put them at 54 seats, which would give them a lot more breathing room in the next Senate.
Now getting Republicans to 57 seats is a bit of a tougher task, though remember that there are four seats in which the Democrats are favored by more than 6 points but are within the margin of error. If the polls are underestimating Republicans across the board right now, one of these seats may fall. I’d keep a particularly close eye out on West Virginia, which Trump captured by over 40 points in 2016.
Of course if Republicans are really gaining six seats in the Senate, then the big news out of election night probably isn’t this massive gain.
It’s probably that Republicans held onto the House as well. It is extremely unlikely that the national environment that allowed Republicans to control 57 seats in the next Congress doesn’t also send back a Republican majority in the House as well.