Election Day edition: A final look at the under-the-radar polls that still appear brutal for Trump

“I’m not in a whole lot of suspense about the outcome of the presidential race."

Good morning, and happy Election Day! Welcome to Margin of Error, a newsletter from me about the polls and the way they are covered.

If you’re new here, thanks for signing up! If you find it interesting, I’d be grateful if you would encourage others to sign up. If you missed last night’s edition, you can find it here.

If someone sent this your way or you found this post through Twitter or other channels, make sure to subscribe below.

Here are the final poll averages:

• National: Biden +8.4 (via 538)

• Florida: Biden +2.5

• Pennsylvania: Biden +4.7

• Michigan: Biden +7.9

• Wisconsin: Biden +8.4

• Arizona: Biden +2.6

• Georgia: Biden +1.2

• North Carolina: Biden +1.8

• Texas: Trump +1.1

• Ohio: Trump +0.8

Now, a final look at the district polls.

---

A few weeks ago, I focused an edition of Margin of Error on congressional district polls. They have been painting a consistently clear and brutal picture for President Donald Trump as we head into Election Day. Even as there has been a lot of focus on national and state polls, these district polls are, in many ways, most telling about the state of the race.

One of the biggest misnomers about the 2016 election was that the polls missed Trump’s late surge. In fact, the data was clear: Hillary Clinton was collapsing and more voters were swinging to Trump. 

The state polls found this, and in some states—Michigan, Wisconsin, and Pennsylvania—they couldn’t catch it fast enough. But some pollsters, like Tom Jensen, the director of the firm Public Policy Polling, saw it happening on a smaller but more telling scale: polls of congressional and legislative districts. 

Jensen is a Democrat, and this was the big factor giving him heartburn ahead of Election Day in 2016. His firm was seeing the numbers shift rapidly to a point where Clinton was significantly trailing Barack Obama’s 2012 numbers in key parts of places like Pennsylvania and Michigan.

For this final 2020 election edition of Margin of Error, I wanted to do a check-in of sorts on these polls. 

With help from FiveThirtyEight over the past few weeks, I have gathered together more than 80 US House polls, all of which were released in September, October, or the early days of November. I used The Daily Kos’ comparison of presidential results for 2012 and 2016, which is organized according to congressional district lines that are in place for 2020.

Together, these polls continue to paint a clear picture. Biden is performing better than Clinton by an average of 9.5 points. (This spreadsheet has a fuller breakdown.) Only in two of the districts is Biden performing worse than Clinton did in 2016. 

A few recent entries stood out to me:

• NY-11: This is the district that includes Staten Island, the only Republican-leaning portion of New York City. Obama won it in 2012, then it swung wildly for Trump, who won it by almost 10 points. It’s back to a swing race, with Biden trailing by just three in a recent Marist College poll. Max Rose, the incumbent Democrat, led by one point in one of the most closely watched congressional races of the cycle.

• OH-12: Biden’s campaign made one of its final stops in Ohio, and this district could be a key signal if he’s going to swing it back to the Democratic column. Trump won this district by more than 11 points in 2016, and now he only leads by one.

• TX-03: This is a poll conducted by the Democratic congressional campaign arm, so caveats required. But it shows this district, which includes the Dallas suburbs, swinging an astounding 25 points toward Biden (from Trump +11 to Biden +14). The early vote numbers out of Texas have been stunning. If it does end up swinging, districts like this one will be why.

On Sunday night, I emailed Jensen to get a final temperature check on the state of the race. He told me that PPP has conducted about 400 polls in October. In only one of them—a state Senate district in upstate New York—did PPP find Trump performing better than he did in 2016. And that was by a point. On average over the past week, Jensen has seen about a two-point shift toward Biden.

“I’m not in a whole lot of suspense about the outcome of the presidential race,” he told me.

More from Jensen:

“I’m not in a lot of suspense about the US House, either. We’ve found Democrats leading in 22 different Republican held districts over the last couple months. Of course they won’t win all of them—in a lot of cases the leads were 1 or 2 points. And there are a handful of seats Democrats could lose too. But I’m pretty confident in a double digit gain in House seats for Democrats.

“So we will he most closely watching the Senate and the state legislature battles in Arizona, Georgia, Iowa, Michigan, Minnesota, North Carolina, Pennsylvania, and Texas on Tuesday night. There were a ton of legislative districts in early October across the country where Biden was up by 6 and the incumbent Republican legislator was up by 2 points, with the vast majority of the undecideds being Biden voters. If a lot of those 2 point deficits turned into 2 point leads over the final month … a whole lot of big state legislative chambers will flip Tuesday, too.”

The key story of the race has been its relative lack of suspense and stability, at least in terms of the data. If there’s one thing to take away from my still-nascent exercise in this newsletter, it’s that 2020 is not 2016.

Enjoy Election Day.

Thanks for reading Margin of Error. If you have any tips, comments, or insights about polling, email me at bplogiurato@gmail.com, or find me on Twitter @BrettLoGiurato.

If you liked what you read, share it with a friend and make sure to subscribe.