About our data and analysis efforts

Data surrounding elections are plentiful and complicated. There are polls, surveys, results, donation records and voter files. There are many smart and thoughtful people analyzing and interpreting that data. What do we have to offer?

• Rather than predict outcomes or develop strategy for candidates, we are focused on helping frustrated progressives get involved in ways that are engaging and strategically useful in this election and elections to come. Thus far, we haven’t seen the available data put to that use.

• There is a lot of data-driven punditry and prediction and great academic work on understanding what happened in 2016. We aim to fit somewhere in the middle, using a variety of approaches to look forward and find places where time and money might be especially powerful in securing progressive outcomes now and in the future.

We understand that data have limitations which are problematic, so they need to be handled thoughtfully. As users of data, we are at risk of oversimplifying the people and situations we analyze, failing to see biases in the data, and overlooking changes and dynamics which static data fail to reflect.

• We plan to use only publicly available data. That’s true of some but not all of the analysis and prediction currently available. We also plan to make it clear how to replicate our results with explainers to accompany all data-driven posts. We will keep current versions of our data and code in our Github repository and maintain links to data which is too large to host at Github.

• Our hope is that this empowers you to keep us honest and help us get better. You can analyze the data along with us, check our work, suggest different data we might look at to improve our analyses, and point out biases in the data or our approach that we may have missed. You can reach out on Twitter (@BlueRipplePol), by email (adam@blueripplepolitics.org or frank@blueripplepolitics.org), or raise issues on Github.