Mapping Democracy: The U.S. House of Representatives

There’s a lot more to our democracy than just the President. On Wednesday, I showed you a map of current U.S. Senators, and explained how the clustering of similar states fools your eyes and brain into seeing our urban-rural divide as a divide between the coasts and the middle of the country.

Today, I map the U.S. House of Representatives. The map is below. As before, each hexagon shows one elected representative. I show all the representatives for a single state together, enclosed by a thicker yellow line; the number of representatives varies from one each (Alaska, Wyoming, Montana, North Dakota, South Dakota, Vermont) to fifty-three (California).

I also tried to keep the locations of districts within states roughly geographically accurate. For example, New York’s 26th District covers the city of Buffalo in Western New York, and so in the map it is second from the left in New York.

Red shows Republican representatives, blue shows Democratic representatives, and white shows a few vacant seats whose incumbents died or resigned in the second year of their terms. The single splotch of yellow is Michigan District 3, represented by Justin Amash, who switched to the Libertarian Party in April 2020 and became the first third-party representative in seventy years. Unfortunately for third parties everywhere, he is not running for re-election. Each district is labeled with the name of the person who represents it (and bite me for having the same last name and first initial, Adam Smith [D-WA9] and Adrian Smith [R-NE3]).

The map below is so compressed that the names are extremely hard to read, so if you’d like to see who represents where, click for a larger version.

Map of U.S. House representatives. Red hexagons show Republican senators, blue hexagons show Democratic senators, yellow shows Libertarian, and white shows vacant seats.

Click for a larger version.

What can we learn from this map?

House districts are designed to give equal representation to all Americans, so by definition and by design the House shows a more accurate picture of the state of American democracy. Each district is home to about 700,000 people (the 308.7 million people in the 2010 U.S. Census divided by the 435 available seats). The specific boundaries of the districts are set by state legislatures, sometimes with hilarious consequences, but that’s a post for another day.

The map makes it much clearer that the blue/red divide (the colors are still arbitrary) is driven not by state boundaries, nor by each state’s positions within the nation. Rather, the partisan divide is actually an urban/rural divide. It’s easy to see that in this map, where I have graffiti-circled and labeled the metro areas of Denver, Chicago, Houston, and Atlanta. Those areas tend to elect Democratic representatives, while the rural areas of their states tend to elect Republican representatives.

The same map as above, but with four metro areas circled.

Now that I’ve got these maps of the composition of the U.S. Senate and House today, I can use them to predict what the composition might be after next month’s elections. That’s coming next, along with an updated prediction for the 2020 Presidential Election.

Mapping Democracy: The U.S. Senate

It’s been fun using maps to track current polling for the 2020 U.S. Presidential election. But there’s a lot more to our democracy than just the President. Because each state’s electoral votes are equal to the number of senators plus representatives, it’s easy to go to from elected legislators to Presidential electoral votes. Or the other way around. And so it’s easy to turn a map of electoral votes into a map of Senators.

And so. The map below is organized roughly-geographically, and shows the Senators from each state by name. Roses Republican Senators are red, violets Democratic Senators are blue, and light bluie shows the two independent Senators, Bernie Sanders (I-VT) and Angus King (I-ME), both of whom caucus with the Democratic Senators.

It’s hard to see the names in the regular-size map below, so click on it to see a larger version.

Map of U.S. Senators, two per state. Red hexagons show Republican senators, dark blue hexagons show Democratic senators, and light blue hexagons show independent senators who caucus with Democratic senators. Each hexagon is labeled with one of the Senators who represents that state. Click for a larger version.

Looking at Senators displayed on a map like this, with equal importance given to all states and thematic geography, shows how the rules of the Senate exacerbate the urban/rural differences that keep us divided.

Every state has both urban and rural areas, but states with similar urban/rural percentages tend to cluster together: more urban on the east coast, west coast, and the western side of the Great Lakes, and more rural everywhere else. That, combined with the traditional yet 1000% arbitrary red-and-blue color scheme results in a map with large stretches of red and large stretches of blue. Look how lonely Doug Jones (D-AL) and Susan Collins (R-ME) are as the sole members of their parties for miles around.

I’ll be using this map for many purposes in the future, starting with making predictions for the 2020 Senate election… soon. I need to look at the polls in a lot more data. That was a typo, I meant “look at the polls in a lot more detail,” but honestly “look at the polls in a lot more data” is way better. But first…

You can probably guess what’s coming on Friday, and why it’s 4.35 times more work.

Presidential election prediction 3: COVID-19 campaign edition

A microscope image of SARS-COV-2, the virus that causes COVID-19
One year ago, this virus only infected bats and pangolins.
Today, the President of the United States.

Well that was unexpected.

We got our October Surprise (TM) early this time. Just 25 hours into the month of October, at 1 AM ET on Friday, October 2nd, President Donald Trump announced on Twitter that he had tested positive for COVID-19.

Later the same day, he was reported to have started showing symptoms, although it is unclear when his symptoms began. By that night, he had been moved to Walter Reed Medical Center and placed on supplemental oxygen.

Today at 6:30 PM ET, Trump left Walter Reed Medical Center. I don’t think he’ll be back – not because he won’t need to go back, but rather because his medical staff is busy installing whatever is required to turn the White House into a hospital. If Trump’s condition worsens, the hospital will come to him.

And unfortunately, the typical trajectory of COVID-19 is for patients to improve for a few days and then get worse again. The up-and-down cycle continues until either the patient is healthy enough that they no longer need acute care, or they die. Obviously I hope for a quick, steady, and painless recovery for him and for everyone he may have infected – but I fear that is unlikely at best.

How this will affect the presidential election depends on how quickly Trump recovers, and how effectively his doctors can hide any relapses. We are seeing data only from the very first polls since Trump announced his diagnosis, and they seem to be heavily critical of Trump. For example, in a new Ipsos poll, 67 percent of registered voters agree with the statement, “If President Trump had taken coronavirus/COVID-19 more seriously, he probably would not have been infected with the coronavirus/COVID-19.” Considering how incredibly difficult it is to get 67 percent of people to agree on anything in today’s partisan environment, that’s a strong signal.

Similarly, polling data is starting to turn more strongly in favor of Biden, and that is reflected in the updated prediction map below. I realized that linking directly to the 270towin geographic map means that I can’t link to a larger version of the prediction map. So below is the prediction map, and if you’d like to try it for yourself, see the link below it.

As always, I try to report the data as clearly as I can. I care more about the truth than I care about what I think.

My predicted election results as of today, October 5, 2020

Make your own predictions using mine as a template at 270towin.com!

The final score prediction is the same: Biden 320, Trump 218. But some of the predicted confidence levels have changed:

  • Probably most importantly for determining the outcome: with Biden opening up a six-point aggregate polling lead there, I have switched Pennsylvania from “tilt Biden” to “lean Biden”
  • And similarly, I have changed both New Hampshire and Nebraska’s second congressional district to lean Biden
  • Perhaps most surprisingly, polling in Georgia is now basically a dead heat, so I have moved Georgia from “lean Trump” to “tilt Trump.”

Don’t be surprised if some of these states tilt back toward Trump now that the President is out of the hospital. But still, time is running out for Trump’s re-election chances – many people have already voted.

Including me – I dropped my ballot in the city lockbox on Saturday afternoon, and I got confirmation today that it has been received and counted. My voice has counted, and I hope yours will too. Vote!

I need a better way (very brief daily COVID-19 data update CLXXXII)

A hamster running on a wheel very very fast
My Mac’s processor trying to keep up with the calculations required by my COVID-19 spreadsheet

Unfortunately, COVID-19 has affected so many people in so many places that my spreadsheet has gotten too big to work with. Going from the raw data provided by the Corona Data Scraper citizen science project to my graphs now literally takes about 4 hours of constant attention to Excel. I’ll have to rethink my approach.

Most likely I’ll do the preprocessing in Python, a simple but powerful programming language used by scientists all over the Universe. I’ll try to provide step-by-step instructions on how to how to run Python, and to document my programs extensively, so you can try them yourself. Even though Python can make graphs, I plan to continue making the graphs in Excel, because it’s a simple tool that so many of you already know how to use.

It might be a few days until my next COVID-19 update. In the meantime, I’ll keep posting other things on my usual Monday-Wednesday-Friday schedule, including something coming later today.

Repeal and Replace or Circumvent?: Comparative Uses of the Presidential

Approximately a photo of Mac: dude with a mustache and bushy sideburns
Not actually Mac

Guest Post!
One of the joys of being alive is having smart, curious friends to talk with – or to write guest posts for your blog. I’d love to see more of these, especially from friends with perspectives and opinions different from my own – email me your ideas!

Today it’s awesome friend Mac writing with a cool little study on how the last four U.S. Presidents have used their power to write executive orders. Enjoy!

Trump is widely known for his criticisms of Obama’s use of executive orders to circumvent congress and the political process.

Why is @BarackObama currently issuing executive orders that are major power grabs at authority? This is the latest

However, since Trump became president he has frequently used executive orders as a way of getting things done. For an explanation of and historical primer on executive orders, check out this JSTOR daily article. True policy and history nerds, read on. In the first year of his presidency, he was on pace to use double the number of executive orders that President Obama had used. In fact, Obama used the executive order less than any president since Cleveland. So why has Trump, once a critic of the executive order, suddenly begun using it to pass legislation?

One explanation for this could be that he is simply using executive orders to erase the bad policy of his much maligned (on the political right) predecessor, “Mr. Trump…worked to deregulate industries and dismantle Obama-era programs through executive order.” And certainly, examples of this abound, highlighted by news media and Trump’s own promises of undoing Obama’s work. But how common is this? How often did past presidents use executive orders to revoke legislation of their predecessors? And taking this into account, if Trump is really only using executive orders to repeal his predecessors’ legislation, how often is he using it to circumvent the regular legislative process?

Luckily for us, there’s some data to dig into. Enter the Federal Register. An office of the National Archives and Records Administration, the Federal Register helps citizens and policy wonks alike understand current proposed and passed legislation covering everything from marine safety to administrative practice and procedure to government procurement and to our friend, the executive order.

Listing documents back to 1994, the Federal Register lets anyone download all executive orders from almost all of Bill Clinton’s presidency to the current administration. As part of the dataset, they include disposition notes that contain helpful information on whether the executive order revokes, amends, or supersedes any other executive orders with relevant dates. From here, we can further analyze which executive orders were rolling back previous legislation that was viewed as outdated or bad policy, and which were used to fully circumvent the legislative process. Below is that breakdown.

Presidential Use of Executive Orders and Revocations

President # of EOs Revocations of previous EOs Revocations as % of EOs Months in Office Average EOs – revocations per month
Trump 86 18 21% 21 3.24
Obama 276 60 22% 96 2.25
Bush 291 79 27% 96 2.21
Clinton* 274 66 24% 84 2.47
*Clinton’s executive orders are circa 1994, as far back as the Federal Register tracks them.

We can see here that not only does Trump have the highest number of executive orders per month of being president compared to his three predecessors, but he actually has the lowest number of revocations as a percent of total executive orders issued. So, not only is Trump issuing a lot of executive orders, but he, more than his three predecessors, is doing so mostly to circumvent the legislative process, not to revoke Obama’s, or any other presidents’, own executive orders.