![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
maxomaiRasmussen's been acting funny of late, but a lot of people still trust this lean-R poll. For my money, lean-D PPP is a bit better once you account for the bias. But Rasmussen seems to hold more sway with conservative audiences. And that is why their latest poll of Kentucky should scare the shit out of Mitch McConnell.
500 Kentucky LVs polled January 29-30 2014
McConnell (R) 42
Grimes (D) 42
The rest of the poll is behind a paywall, but Rasmussen claims that they show McConnell's primary opponent, Matt Bevin, beating Grimes by six points.
Rasmussen has a Republican bias in their likely voter model — meaning, they count more Republicans as likely voters than are actually likely to show up. This cuts two ways here. On the one hand, if exaggerated GOP turnout leads to a McConnell tie, then Grimes is probably a few points ahead of him. On the other hand, an exaggerated Republican electorate might prefer a more conservative candidate like Bevin over McConnell, but that means that Bevin will have an easier time beating Grimes if Rasmussen's LV model is correct. And, it bears mentioning, if Bevin doesn't make some stupid, inane remark that makes him the next Todd Akin. I would bet money that one of those two conditions will fail.
Nonetheless, this poll makes the case that McConnell is a weaker candidate than Bevin, and does so in a way that's compelling to McConnell's base. Between this and probably trailing Grimes, this poll has to sting at McConnell HQ this morning.
500 Kentucky LVs polled January 29-30 2014
McConnell (R) 42
Grimes (D) 42
The rest of the poll is behind a paywall, but Rasmussen claims that they show McConnell's primary opponent, Matt Bevin, beating Grimes by six points.
Rasmussen has a Republican bias in their likely voter model — meaning, they count more Republicans as likely voters than are actually likely to show up. This cuts two ways here. On the one hand, if exaggerated GOP turnout leads to a McConnell tie, then Grimes is probably a few points ahead of him. On the other hand, an exaggerated Republican electorate might prefer a more conservative candidate like Bevin over McConnell, but that means that Bevin will have an easier time beating Grimes if Rasmussen's LV model is correct. And, it bears mentioning, if Bevin doesn't make some stupid, inane remark that makes him the next Todd Akin. I would bet money that one of those two conditions will fail.
Nonetheless, this poll makes the case that McConnell is a weaker candidate than Bevin, and does so in a way that's compelling to McConnell's base. Between this and probably trailing Grimes, this poll has to sting at McConnell HQ this morning.
