On Sunday, The Washington Post and our polling partners at ABC News released our last pre-election poll.
This was close to the margin seen in polls from other established pollsters. It seemed to be at odds with the confidence expressed by Republican politicians and leaders, certainly, but it was what it was: a relatively narrow gap for the GOP.
And a narrow margin favoring the Republicans is what we saw Tuesday.
This isn’t meant to be gloating, really, though obviously it is a credit to our pollsters. It’s particularly not meant to be gloating given that 1) not all of the votes have been counted and 2) it is never wise to assume that polls will be perfectly accurate. When we consider polls we should recognize that they include uncertainty, both measured in the margin of error and by the nature of asking questions about an election in which things might change. But, particularly given questions about polling in recent election cycles, it’s worth pointing out that the predictions of that last Post-ABC poll reflect what happened.
Cook Political Report tracks the results of House contests as they come in. On Friday morning, those numbers reflected a national total of 51 million votes for Republican candidates and 45 million for Democrats — which, if you include third-party votes, give the GOP 52 percent of the total. But there are still as many as 4.6 million votes to be counted in California, a state where the Democratic Senate candidate is getting 60 percent of the vote (both Election Day and mail-in). So that difference will narrow.
More interestingly, the Cook Political data allows us to compare where races are with how those places voted in 2020. Using DailyKos’s analysis of the presidential vote in each of the districts of the 118th Congress (and excluding House races where candidates ran unopposed), we see that the average shift across House seats was about 5.1 points. Joe Biden won the 2020 election nationally by 4.5 points — in line with a 1-point Republican advantage in 2022.
On Wednesday, we noted that the likely Republican majority in the House would be narrow enough that it could be credited to the redistricting process — both in newly drawn maps (as in Florida) and in maps that were blocked by the courts (as in New York). Curious whether there were obvious patterns in redistricting and the election results, I made this excessively complicated graph showing all 435 districts. The short orange line shows how the 2020 margin of the district before redistricting shifted after redistricting. The purple lines show how the margin (as of Friday morning) compares to the new 2020 margin.
Put more simply: Purple lines pointing right shifted to the GOP relative to the 2020 vote. Purple lines pointing left shifted to the Democrats.
In most cases, you’ll notice, the redistricting didn’t change the lean of the district very much, and the 2022 results generally reflect that lean. There are a lot of districts with big shifts, though you’ll notice those often go all the way to the edge of the chart: candidates getting 100 percent of the vote in 2022 because they ran unopposed.
We don’t yet know what the final margin in House voting will be. We don’t even know whether Republicans will have taken the majority! But we do know that, as with Senate polling, polls showing that the contest for the House was close were, in fact, accurate.
A noteworthy discovery, to be sure.