Which Democratic presidential candidates are viable in your district?

If you’re in between candidates, following your neighbors can help make your vote count. Here’s a tool to help you decide.

The Democratic presidential primary system is weird. While 3 in 4 Democrats support electing the president by national popular vote, the party selects its nominee with a delegate system even more complicated than the Electoral College.

Here’s how it generally works: Each state and district (state or congressional) is assigned a number of delegates in proportion to the number of Democrats in that jurisdiction. Delegates are then assigned within those jurisdictions based on the number of votes each candidate receives.

An important feature in this process is the 15 percent threshold. In order to get any votes in a state or congressional district, a candidate must receive at least 15 percent of the vote. Delegates are assigned proportionally among candidates meeting that threshold (among the viable candidates).

This means a vote is more likely to count if it’s for someone more likely to get over 15 percent. But how’s a voter to know who’s viable in their state or district?

Enter FiveThirtyEight’s primary forecast model. The FiveThirtyEight forecast considers completed races, polls, trends, endorsements, and a host of other factors to predict the likelihood of each candidate’s delegate total, given how previous elections turned out. As of writing this, the model predicted a 63 percent chance of no candidate getting a majority of pledged delegates, a 21 percent chance that Joe Biden gets a majority, and an 18 percent chance that Bernie Sanders gets a majority.

Image for post
Image for post

Underlying this forecast are mini-forecasts for each jurisdiction, based on state polls, demographics, and other factors. For example, here’s how FiveThirtyEight forecasts the San Francisco House district:

Image for post
Image for post

San Francisco has 7 delegates, of which Sanders is likely to get about half, followed by Biden and Warren. But if someone wanted to vote for Bloomberg, should they? His average forecasted delegates is 0.4, indicating there’s a good chance he’ll get zero.

To give a more precise answer to that question — and avoid the challenge of mousing over your district — I built a tool showing FiveThirtyEight’s predicted probability of each candidate reaching the viability threshold in a given district. Here’s how it works:

When you open the tool, you may be prompted to log into your Google account (it collects no information about you), and then you’ll see something like this:

This is Google’s Colaboratory product for building data science notebooks in the cloud. The dialogs on the right let you enter the state and district, which are pre-populated for San Francisco. Leaving that as-is, you can run it using the Runtime...Run all command:

Image for post
Image for post

It’ll prompt a warning, which you can accept by clicking RUN ANYWAY (again, I promise it doesn’t collect any information or do anything malicious). Then it’ll run the code, and after about 10 seconds you can scroll past all the code to the chart at the bottom:

Image for post
Image for post

Each candidate is plotted with the predicted range of vote shares, based on the latest data from FiveThirtyEight. Below it is a higher-quality version:

Image for post
Image for post

Taking Sanders as an example, FiveThirtyEight forecasts that he’ll get 35 percent of the vote or more in at least half the simulations (the blue dot is the median). The edges are 80 percent confidence bands, meaning that Sanders gets at least 45 percent of the vote in 10 percent of simulations, and at least 28 percent of the vote in 90 percent of simulations. That is, Sanders is extremely likely to pass the 15 percent viability threshold, and he’s marked accordingly as “>90% viable.”

Biden is also highly likely to be viable in San Francisco. FiveThirtyEight’s data only supports broad categories, so he’s marked as “50–90% viable,” but the position of the confidence band indicates that he’s viable in about 90 percent of simulations. More of Warren’s band is below 15 percent, so she has a strong but lower chance of viability, and as expected, Bloomberg’s chances of falling below the threshold and receiving zero delegates are over 50 percent. A San Francisco voter deciding between Bloomberg and Biden might make their vote count more by going with Biden, though these circumstances will vary by district.

To try your own district, give the tool a whirl. Feel free to drop me a comment if you have any questions or difficulties, and happy voting!

Written by

Economist. Founder and president of the UBI Center. Studied at MIT and UC Berkeley. YIMBY. Former Google data scientist.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store