In the race to see who loses to Republican John Boozman, I thought incumbent Arkansas senator Blanche Lincoln had the advantage over Lt. Gov. Bill Halter. The polling during the runoff says otherwise, but I’d feel better if we had one poll that wasn’t Daily Kos/Research 2000.
For example, in the most recent Daily Kos/Research 2000 poll, 48 percent say they voted for Halter in the first round, 46 percent say they voted for Lincoln, and 3 percent say they voted for conservative candidate D. C. Morrison. But in the actual results, Lincoln had 44.5 percent, Halter had 42.5 percent, and the conservative, Morrison, finished with 13 percent. Does a conservative who voted for the Ron Paul–backing Morrison switch to netroot favorite Halter? Or switch to Lincoln? Or do they stay home?
Doesn’t that suggest that perhaps this sample leans a bit too heavily to Halter? And shouldn’t more people remember voting for Morrison? Or do voters change their mind when the guy they voted for finished third?
The 57–43 split women-to-men seemed a little high to me, but apparently it was 60–40 in the 2008 primary, so I guess that’s in line with expectations . . .