The Primary Colors
algorithm was created primarily to identify low-value Democrats in safe districts for primary challenges — but it can show us other things as well.
When we assign each member of Congress an expected score, it’s based on their colleagues’ scores in districts with similar partisan profiles.
Obviously we’d expect Democratic politicians to vote more conservative the more conservative their district gets, and more liberal the more liberal their district gets. And by-and-large, they do.
But Republicans don’t really follow that trend, as you can see in red on the graph below:
If you only take away one thing from this graph, it should be that the expected value for Republicans is nearly a perfect horizontal line. Translated into plain English, that means Republicans vote conservative almost no matter what. It doesn’t matter what type of districts they represent.
For example, the most conservative district a Democrat currently holds is an R+16 – Jim Matheson (UT-04) – where we expect a progressive score of 44.1%. Matheson votes even a bit more conservative than that, with an actual score of 41.3%.
By contrast, look at Gary Miller in CA-31. That’s the most liberal district a Republican represents (D+5), and our algorithm expects him to vote with progressives a mere 7.6% of the time. In reality, he’s only voting progressive 3.8% of the time.
Now imagine if that were flipped around. We expect a Democrat representing an R+5 district to vote progressive just 60.5% of the time. Which sounds more moderate: a score of 60.5% or a score of 7.6%?
You can expect nearly identical voting patterns from almost all Republicans whether you live in a swing district or an uber-conservative district.
Democrats moderate in response to district demographics; Republicans don’t. Commentators have been noting the lock-step Republican Party discipline for years, especially since the Tea Party came on the scene. The graph above is the proof.
The challenge remains getting this information to swing district voters this November. While No Labels gives “problem solvers” like Larry Bucshon and Scott Perry cover as good-faith cooperators, the reality is that there’s little, if any, daylight between their voting records and fire-breathing obstructionists like Steve King or Michele Bachmann.