To determine which races and candidate to support, we follow a three-part process to prioritize state chambers, forecast outcomes by district, and select a slate of high-quality candidates.
WE CELEBRATE EVERY SEAT THAT WE CAN FLIP, BECAUSE IT BRINGS US THAT MUCH CLOSER TO PROMOTING A PROGRESSIVE AGENDA IN OUR STATES. BUT ULTIMATELY, WE’RE INTERESTED IN FLIPPING ENTIRE CHAMBERS.
Real change happens when progressives are able to gain a majority (or, in some cases, break a supermajority) in a state senate or state house.
We have developed a “decision tree” algorithm to help us prioritize our state chambers. Relevant inputs include:
The number of votes it will take to flip the chamber.
This is informed by both the number of seats Democrats need to gain a majority and the estimated number of votes it will take to flip the seat.
The balance of power in the state.
We look at whether we have an opportunity to break a Republican trifecta or make a Democratic one, as well as the relative power of the governor and each state chamber (e.g., tie-breaking votes, veto power, and supermajority status).
Impact on gerrymandering.
Here, we look at how redistricting decisions are made (and maps are approved), as well as the “efficiency gap,” or degree of gerrymandering, present in the state.
Impact on voting rights.
We look at the state’s record on voter ID, voter registration, early voting, and ballot access policy and legislation.
Electoral college votes.
This both demonstrates the potential impact of improved electoral laws on national elections and serves as a proxy for the population of the state, and therefore the number of people affected by progressive shifts in state government.
IN ANY GIVEN STATE, WE PREDICT THE COMPETITIVENESS OF STATE LEGISLATIVE RACES BY DISTRICT, RANKING THEM BY LIKELY MARGIN.
We’re building a regression model that uses previous electoral history to understand the historical relationship between state legislative, statewide (e.g., gubernatorial), and national (e.g., Senate) elections and project likely 2017-2018 margins. So far, we’ve focused on previous electoral outcomes and candidate incumbency status because they’re most predictive of electoral results.
In 2018, we plan to layer in additional variables: Census demographic data on race, educational attainment, and rural/urban status; dynamic variables including Presidential and Congressional approval ratings; and campaign finance and other candidate-level data.
A few notes:
We’ve used regression-based modeling rather than polling data because of the small sample sizes in state legislative races.
In recent years, political analysts and data scientists have relied on polling for forecasting purposes, particularly for executive and federal races. While polling is the best tool available for national elections, which are infrequent and cover larger districts, regression analyses are more appropriate for state legislative races (as polling data from state legislative races is sparse and inconsistent).
We have trained our model on historical data, testing its ability to predict previous elections.
At each step, we check the historical predictiveness of our model, applying it to previous electoral years and adjusting as fit. As we gather more data, this will become a more automated process, resulting in fewer human assumptions applied to the model; but for now, we are relying mostly on electoral fundamentals commonly used among political scientists and analysts we follow. In upcoming versions of our model, we will quarantine a subset of historical elections, separating the districts we use to build the model from those we use to test it.
WE SUPPORT CANDIDATES WHO HAVE WON THE DEMOCRATIC PRIMARY IN THEIR DISTRICT.
We interview candidates and dig into their platforms to ensure that they are committed to gender and racial equity, economic opportunity, and environmental sustainability. Finally, we work with candidates to ensure that they have a viable plan to raise funds and to steward donors’ resources effectively.
Throughout the process of district and candidate selection, we are committed to providing transparency in our process and outcomes.