House Model Consensus Trend Visualization

Benjamin Rosenblatt • October 25, 2018

Using the House Model Consensus data put together by Jason Pipkin (@jipkin), I've compiled a visualization to show how House modeling changes day-by-day.

The slider on top of the visualization changes the entire visualization based on the day listed. It currently goes as far back as October 22nd.

The map above shows each congressional district shaded based on Democratic likelihood to win. The interactive legend and House Model Consensus Ratings bar chart and pie chart, however, are shaded into a qualitative ratings system.

If you click on any of the qualitative ratings in the Interactive Legend, the corresponding districts will be the only ones to appear on the map. If you click on the Tossup rating, for example, you can see how those seats break between a higher likelihood of a Democratic win or Republican win.

The qualitative ratings system is organized accordingly:

  • 0 - 10% Dem Likelihood = Safe GOP
  • 10 - 25% Dem Likelihood = Likely GOP
  • 25 - 40% Dem Likelihood = Lean GOP
  • 40 - 60% Dem Likelihood = Tossup
  • 60 - 75% Dem Likelihood = Lean DEM
  • 75 - 90% Dem Likelihood = Likely DEM
  • 90 - 100% Dem Likelihood = Safe DEM
The House Consensus Model is a combination of the following models: CNN; DDHQ/0ptimus; Noah Rudnick's Model; The Crosstab; 538 Lite Version; 538 Classic Version; and 538 Deluxe Version.

If you have any questions or suggestions, feel free to message me with them on Twitter (@BenJ_Rosenblatt)!