Alaska's at-large seat barely moves after a Data for Progress poll

One new poll landed in Alaska's at-large district. A Data for Progress survey of 605 likely voters, fielded July 28 to August 4, put the Republican at 47 percent and an independent candidate at 37 percent. The poll as reported to us lists no Democratic option, so its 10-point gap does not translate directly into the two-party margin our model estimates.

Our forecast did not move much. We now give the Democratic side a 73 in 100 chance in the district, with a mean margin of D+6.4 and a 5th-to-95th percentile range running from R+10.5 to D+23.3. The run before this one, at 10:30 a.m. ET on August 18, had the Democratic chance at 74 percent and the mean margin at D+6.5, with a nearly identical range. That is a one-point shift in probability and a tenth of a point in margin: confirmation of what was already in the model rather than a change to it.

The width of that range is the more useful number here. Our model's 90 percent interval spans more than 30 points, which is another way of saying a single poll in a single statewide district does not pin this race down 77 days out.

Nationally, our average of 26 generic-ballot polls over the past 45 days stands at D+6.7. Our model gives Democrats a 93 percent chance of reaching the 218 seats needed for House control, with a mean of 242.4 seats. One caveat on that figure is ours to own: our simulation correlates error nationally, by census division, by state and within each district, but we have no data to add a term linking demographically similar districts across the map. Without it, the House control number is marginally firmer than it should be.