ME-2 forecast steady at 89 in 100 as DCCC poll shows a one-point gap

One new poll is in the ME-2 file. The DCCC Targeting Team surveyed 621 likely voters August 11-13 and found Paul LePage at 49 percent and Matthew Dunlap at 48 percent. That is a one-point gap in a survey of that size, which is to say the race is close in that poll rather than that either candidate is ahead in it. No margin of error was reported with the numbers we have, and the sponsor is a party committee.

Our forecast did not move on it. The model gives Dunlap an 89 in 100 chance in Maine's 2nd District, the same figure as our 10:32 a.m. ET run on September 22. The mean margin ticked from D+10.8 to D+11.1, and the 5th-to-95th percentile range shifted from R+4.0 to D+25.5 out to R+3.8 to D+25.9. Those are rounding-scale changes, not a revision.

The distance between the poll and the model is the thing worth stating plainly. A survey with the two candidates a point apart sits well inside our range, but it is toward its Republican end; our 5th percentile outcome is R+3.8. The model is also weighing a national environment our average puts at D+7.8 on the generic ballot, across 28 polls fielded over the past 45 days.

Nationally, our model gives Democrats a 98 percent chance of winning House control, with a mean of 244.5 seats against the 218 needed. One caveat is ours to own: our simulation correlates error nationally, by census division, by state and by district, but we have no cluster data to correlate demographically similar districts across state lines. Without that term, the House control number is marginally firmer than it should be.