Franklin & Marshall poll in PA-7 nudges our forecast down two points

A Franklin & Marshall College survey of Pennsylvania's 7th District, fielded September 9-20 among 350 registered voters, put Democrat Bob Brooks at 45 percent and Republican incumbent Ryan Mackenzie at 40 percent. The sample is small and the payload carries no margin of error for it, so the five-point gap is best read as a close race rather than a settled one.

The number changed our forecast, but not by much. Our model now gives Brooks an 84 in 100 chance of winning the seat, down from 86 percent in the previous run earlier this morning. The mean margin moved from D+9.8 to D+8.8, and the 5th-to-95th percentile range shifted a point in the same direction, from R+4.8 to D+24.4 to R+5.8 to D+23.4. That is a poll slightly to the Republican side of where our average already sat, absorbed as such.

The spread around that mean is the more useful fact. Our model still allows for outcomes from a mid-single-digit Republican win to a Democratic margin north of 20 points, which is a wide enough band that a single 350-person survey does little to narrow it. Mackenzie's 16 percent chance is not negligible.

Nationally, our average of 30 generic-ballot polls over the past 45 days stands at D+8.3, close to the D+8.8 mean margin our model projects in this district. Our House model gives Democrats a 98 percent chance of reaching the 218 seats needed for control, with a mean of 245.8 seats. One caveat on that figure is ours to own: the simulation correlates error nationally, by census division, by state and within each district, but we have no data to correlate demographically similar districts across state lines. Without that term, our House control number is marginally firmer than the underlying uncertainty warrants.