Daniel Donner

Daniel Donner

@donnermaps.bsky.social

Contributor @the-downballot.com. Focusing on data, maps, and demographics. Find our data (and the CD hexmap template!) at https://the-db.co/data

4.6K followers818 following589 posts

Pinned

Replying to @donnermaps.bsky.social

Several blocks in downtown Chicago are not part of *any* congressional district map (new or old). How on earth is this possible? The answer involves billions of dollars, brothels, Abraham Lincoln, and a bit of geomorphology. 1/ New IL map: t.co/QzQX9spFUO

Daniel Donner reposted

Manhattan peaked in 1910 at 2.3 million people. One Lower East Side census tract held 44,000 people on 50 acres; those blocks now hold about one person for every four. Data: IPUMS NHGIS · Bromley atlases, NYPL #dataviz #maps #NYC #history

Daniel Donner reposted

Replying to @stephenwolf.bsky.social

There will be a major tendency among pundits to downplay GOP gerrymandering when Democrats flip the House and the GOP fails to flip some of its mid-decade remap targets, but the pre-2026 baseline was still very gerrymandered. Even in a wave, the national map could cost Dems two dozen seats or more

Stephen Wolf@stephenwolf.bsky.social

GOP gerrymandering shifted the median congressional seat from Trump+3.1 in 2024 to +4.5 in 2026. He won the 2026 median by 3 points more than his national margin. That gap could grow by several points in 2028 if only the GOP redraws more states, but new Dem maps may be able to fully counteract them

Daniel Donner reposted

New: Low-info voters, who put Trump in the White House, now look even more anti-Trump than high-info voters, and give Dems bigger margins on the generic ballot. 1 out of 3 Trump voters who can't name the party in charge of Congress now back Dems for 2026. www.gelliottmorris.com/p/poll-low-i...

gelliottmorris.comPoll: Low-information voters have turned hardest against Trump and the GOPTrump voters who can't answer two basic questions about politics are three times as likely as other Trump voters to back Democrats for the House. This is a big change since 2024.

Daniel Donner reposted

The election is coming... time to get maps ready! We have both normal maps and our hexmap cartogram available for you as image files, shapefiles, and svg files. The cartogram can also be found on Datawrapper under USA>>Congressional Districts Elections 2026 (hexmap) docs.google.com/spreadsheets...

A nice twist on the House non-contiguous cartogram; each state appears to be its own voronoi diagram, which makes for straight borders between districts.

Karim Douieb@karimdouieb.bsky.social

So every seat gets the same space here. Seat by seat (@cookpolitical.com): 214 favor Democrats, 200 favor Republicans, 21 toss-ups decide the majority.

Daniel Donner reposted

‘New data shows that this year, the average cost of a family health plan crossed $28,000, rising by 5%. The share coming out of employees’ paychecks shot up even faster—to $7,471, an increase of 9% from 2025.’ www.wsj.com/health/healt...

The average of all four indicators, which has an excellent track record (especially when all four indicators are as close to each other as they are now), is now 11.3

The NCB and GCB have been consistently converging with the values calculated from special elections and primaries this summer and fall. Less than two weeks to go!

Daniel Donner reposted

Yet another reason why you should instead tune into The Downballot/Strength In Numbers election night livestream! If you're a real Sicko, you can even bookmark it now: open.substack.com/live-stream/...

Oliver Darcy@oliverdarcy.bsky.social

Scoop from @brettlogiurato.bsky.social: CBS has sold its local election night coverage to DraftKings, stirring more controversy inside the David Ellison-owned company. Details in @status.news: www.status.news/p/cbs-draftk...

A timely reminder. But note that while the slope of every graph is positive, the x-intercept is sometimes positive (2014; Rs outperformed on average in light blue states)... and sometimes negative (2012; Ds outperformed on average even in moderately red states).

Owen Winter@owenwntr.bsky.social

In *every* election for *both* House and Senate since 1998, polls have overestimated Democrats by more in red states than blue states (and vice versa) www.economist.com/graphic-deta...

Daniel Donner reposted

unreal piece by @federicacocco.bsky.social @teoarmus.bsky.social. look at those job numbers falling off a cliff! www.washingtonpost.com/politics/202...

Washington Post Data Team@wpdatateam.bsky.social

David Reyna voted for Trump. He didn't expect it to hurt his business. “We didn’t think this administration was going to crack down on undocumented workers that hard,” Reyna said. "It is completely devastating the local construction industry. It’s grinding it down to a halt.”

Daniel Donner reposted

ChatGPT is adding the signatures of real New Yorker cartoonists to its AI-generated images. So far, I've found the signatures of more than 15 contributors. "It’s slop, and unlike the other slop that I’ve encountered, this is slop that’s pretending to be me." www.niemanlab.org/2026/10/chat...

Democrats don't want to stick around either, but didn't reach the record of 1992. 9 of these seats are open because of redistricting. 7 because of a primary loss.

Daniel Donner reposted

With both Lake Powell and Lake Mead at record-lows, here’s a look at how their water levels have varied since Powell was filled in 1980. Neither has recovered from a severe drought in the early 2020s, while Colorado River flows continue to drop. #dataviz #climate #globalwarming

Daniel Donner reposted

An all around weak -- but not tragic -- payrolls report for September. +29k jobs created, about 60k below expectations. Revisions made both July and August look weaker, too. Over the past 3 months, we've an average of a mere +17k per month.

Daniel Donner reposted

Replying to @kjhealy.co

I’ve written this up here. In addition to there not being a Covid-scale drop in tourism recently, you might also be wondering why visits from Mexico seem to double around 2004-05. The answer is not a sudden shift in Mexican preferences but rather our old friend, There Was A Change In The Measure.

kieranhealy.orgTrump Tourism EffectsIn the past day or two on social media I kept seeing this graph from a Semafor story on the declining volume of international visitors to the United States. Semafor Graph. It seems to show, and was be...

Daniel Donner reposted

Lake Erie turned toward autumn in a single week. Every hour of its surface currents, 16 to 23 September 2026, colored by water temperature: the surface lost about 6°F in seven days. Data: NOAA Great Lakes Operational Forecast System (model) #LakeErie #GreatLakes #dataviz

There's still more to show in the relationship between primary votes and the House popular vote! You can plot the data in graph below as a scatterplot. But a) it's obvious from the original graph this is a bad idea and b) it's not too helpful if you want to predict things. But...

The Downballot@the-downballot.com

So what if we compare the margin in the national primary to the House popular vote? In recent years, the two have matched with almost scary synchronicity. The purple (primary) and yellow (House) dots from 2006 onward are almost identical. But things break down before then. So is this real?

View full thread

These mismatches then show up as outlying data points on the regression of *change* in House popular vote versus *change* in national primary margin. But all the other years fall nicely along one tight line.

1938 is the end of a dramatic realignment period. The years 1966, 1982, and 2006 were previously discussed in this thread:

Daniel Donner@donnermaps.bsky.social

We were pretty surprised to find such a clear relationship between primary vote margins and the House popular vote stretching across decades with such different political environments. But there's some years where one zigs and the other zags. What happened there? We'll start with 1966.

There's still more to show in the relationship between primary votes and the House popular vote! You can plot the data in graph below as a scatterplot. But a) it's obvious from the original graph this is a bad idea and b) it's not too helpful if you want to predict things. But...

The Downballot@the-downballot.com

So what if we compare the margin in the national primary to the House popular vote? In recent years, the two have matched with almost scary synchronicity. The purple (primary) and yellow (House) dots from 2006 onward are almost identical. But things break down before then. So is this real?

View full thread

But because four of the epochs have very similar slopes, we can transform the chart by looking at the *change* in both variables. And when we do that, the epoch divisions transform into mismatches between the two time series.

There's still more to show in the relationship between primary votes and the House popular vote! You can plot the data in graph below as a scatterplot. But a) it's obvious from the original graph this is a bad idea and b) it's not too helpful if you want to predict things. But...

The Downballot@the-downballot.com

So what if we compare the margin in the national primary to the House popular vote? In recent years, the two have matched with almost scary synchronicity. The purple (primary) and yellow (House) dots from 2006 onward are almost identical. But things break down before then. So is this real?

... this plot hides within it four epochs with nearly the same slope (and one epoch that is a mess, 1918-1938). It's kind of like a Simpson's paradox (not sure if it counts since it doesn't reverse the sign of the slope).

Daniel Donner reposted

Deaths now outnumber births in reliably Republican counties for the first time since at least 1982. We explain how this may already be affecting voting patterns in America bit.ly/471apDL

This is why diesel prices - which selectively affect Trump voters and rural areas in particular - matter for the election. www.kff.org/public-opini...

Daniel Donner@donnermaps.bsky.social

If you wanted to design a policy to hurt Trump voters in particular, a diesel fuel price increase would be very effective. Three states - WY, ND, AK - even use more diesel than gasoline. Details by state; not all use is paid for by residents of course: www.eia.gov/dnav/pet/pet...

If we could go back in time to use the national primary margin to predict the midterm House popular vote with the data available at that time, how well would it have done? Pretty well overall, with some significant misses like 2006. (Note we see no warning signs of a miss like 2006 for this year.)

The Downballot@the-downballot.com

In primaries this year, Democratic turnout utterly dwarfed Republican turnout. But does it mean anything for November? It turns out the answer is very much yes. A new analysis from @donnermaps.bsky.social suggests that Dems are on track for a 13-point blowout in the House vote. Here's why.

View full thread

We can see both measures do a very good job since 2010; in the 20th century, the generic ballot polls were a little worse, especially mid-century. However, this uses the *final* generic ballot polls, which aren't available until election eve. What about earlier?

If we instead look at the data available in July of election year, both for primaries and generic ballot polls, the predictions based on primaries are much better than the generic ballot polls.

If we could go back in time to use the national primary margin to predict the midterm House popular vote with the data available at that time, how well would it have done? Pretty well overall, with some significant misses like 2006. (Note we see no warning signs of a miss like 2006 for this year.)

The Downballot@the-downballot.com

In primaries this year, Democratic turnout utterly dwarfed Republican turnout. But does it mean anything for November? It turns out the answer is very much yes. A new analysis from @donnermaps.bsky.social suggests that Dems are on track for a 13-point blowout in the House vote. Here's why.

The biggest misses can be reasonably explained by this thread from yesterday:

Daniel Donner@donnermaps.bsky.social

We were pretty surprised to find such a clear relationship between primary vote margins and the House popular vote stretching across decades with such different political environments. But there's some years where one zigs and the other zags. What happened there? We'll start with 1966.

Daniel Donner reposted

WSJ: The AI build-out is on track to become the biggest economic bet in US history, dwarfing the investments made to fund other huge US infrastructure projects — such as the railroads, the highway system and the plumbing for the internet.” (via Jefferies)