John says we will be making election predictions on our VIP Live program tomorrow night (8:00 p.m. Eastern Time), and he runs the show. Therefore, I figured I should bone up on the Senate races, in case John goes there.
FiveThirtyEight failed to predict Donald Trump’s victory in 2016, but has a decent record of forecasting Senate races. Its analysis of the polls finds that the election will probably put Democrats in control of 51 to 52 seats. It finds that the Democrats have a 77 percent chance of controlling the Senate next year.
Fifty-two seats would represent a net gain of five. Since the Democrats are likely to lose the seat they hold in Alabama (FiveThirtyEight puts this probability at 80 percent), it would mean that the Dems picked up six seats they don’t hold now.
Which six? The most likely, according to FiveThirtyEight, are in Colorado (85 percent), Arizona (80 percent), North Carolina (64 percent), Maine (60 percent), and Iowa (55 percent). A sixth seat might come from one of the races in which FiveThirtyEight forecasts a GOP win, but gives the Democratic candidate a fairly good chance — perhaps in Georgia where Republicans are defending two seats in races that appear to be close.
Are Senate bulls Mitch McConnell and Lindsey Graham in trouble? McConnell doesn’t seem to be. FiveThirtyEight gives the Majority Leader a 95 percent chance of winning. Graham’s chances also seem good at 77 percent.
Can we reasonably hope for a GOP pickup anywhere other than in Alabama? Not according to FiveThirtyEight.
Michigan represents the best hope. But FiveThirtyEight gives challenger John James, who is now pretty far behind in most polls, only a 19 percent chance of winning. It gives Jason Lewis in Minnesota a 7 percent chance.
Of course, all of this goes out the window if the Republicans have a big night. And I believe the 23 percent chance FiveThirtyEight gives the GOP of holding the Senate is a little better than the chance it gave President Trump four years ago.