Emerson: Rubio+2, Burr+4, Heck+4, Feingold+5, Portman+14

[Vosem]

538 adjusts Emerson for having a +2 pro-R house effect, which is reasonable, but the problem with just adjusting all of these 2 to the R side is that that's the average amount it's off by, not consistent. I think based on other polling, for instance, we could expect a Rubio+2 and Feingold+5 to stay where they are, while Burr and Heck are both tied with their opponents.

But ultimately these things are kind of unknowable.

[Vosem]

Even Emerson can only find Rubio up by 2? Looks like Murphy's going to pull this out after all.

They also pegged him as having a yuge lead (58-37?) among early voters. Little Marco better watch out.

Midterms aren't general elections, but polls pretty consistently had Charlie Crist with over 60% of the early vote in FL-2014. He still lost.