ECU poll puts Cooper up eight, and our model nudges him to 86 percent

A survey from the East Carolina University Center for Survey Research, fielded August 31 to September 3 among 675 likely voters, put Democrat Roy Cooper at 51.7 percent and Republican Michael Whatley at 43.6 percent, with 0.9 percent for other candidates. That is a Cooper lead of about eight points.

Our model now gives Cooper an 86 in 100 chance of winning the seat, with a mean margin of D+6.7 and a 5th-to-95th percentile range running from R+3.4 to D+16.8. The run before this one, at 12:32 PM ET today, had Cooper at 81 percent with a mean margin of D+5.3. The ECU numbers sit close enough to that estimate to confirm the shape of the race rather than change it; the effect was to firm up an existing edge by five points and widen the Democratic end of the plausible range.

The seat is open, with Republican Thom Tillis the incumbent and neither candidate holding it. Our forecast still allows for a Republican win: the low end of the range is a Whatley victory by more than three points, and the model puts his chance at 14 percent.

Nationally, our average of 24 generic-ballot polls over the past 45 days stands at D+6.3. In the Senate, our model gives Democrats a 40 percent chance of winning control and Republicans 59 percent, with a mean of 49.7 Democratic seats. Democrats need 51 seats for control and Republicans need 50, because the vice president is a Republican and a 50-50 tie goes to the Republicans. So a Cooper win in North Carolina is, on our numbers, more likely than not without being sufficient: the Senate map still leans Republican in our simulations even with this race priced as a probable Democratic gain.