If Joe Sestak and Bill Halter win their primaries on Tuesday night there's going to be a lot of media temptation to declare it a huge night for the party's left...but is that really what's going on in those races?
Unfortunately the Pennsylvania and Arkansas primary polls have been short on crosstabs by ideological identification. But there have been some signs that's not really what's driving Sestak and Halter's momentum.
Let's look at Quinnipiac's polls for example. In early April when Arlen Specter led Sestak 53-32, 35% of the voters they surveyed said they saw Specter as the more 'consistently liberal' candidate. On Quinnipiac's most recent poll the Specter lead was all the way down to 44-42. But that 35% of voters identifying Specter as more 'consistently liberal' was unchanged. There was a 19 point reduction in Specter's lead over that period of time, but no change in voter perception of his ideology.
That makes me question an interpretation of a potential Sestak victory as the Democratic left pushing Specter out of office. It seems more like an anti-incumbent/desire for a new face thing than an ideological one.
I haven't seen any polling data by ideology in the Arkansas race, but when we last polled the state in February Blanche Lincoln's numbers within her party were actually their best with liberals. They gave her a 57/28 approval spread to 50/34 with moderates and 46/44 with conservatives. Those numbers suggest that if Halter wins or sends the race to a runoff it'll probably have just as much do with him pulling conservative and moderate Democrats away from Lincoln as liberal ones.
I hope some of the final polls in these races will release results by ideology, because I'm just not sure based on the data I've seen that the closeness in these races is due to backlash from liberal voters.