What does state general election polling tell us?

Survey USA released a bunch more general election polling, commented on by the Hedgehog Report and DaveG at Race42008.  Rudy Giuliani, Fred Thompson, and Mitt Romney were matched up against Hillary in Virginia, Kentucky, and Alabama. I have rearranged the data to show both margins and the %-undecided. SEveral things are worth noting.

First, the GOP is in bad shape in both Virginia and Kentucky due to reasons that have nothing to do with candidates. Kentucky has an election this year where a seemingly-corrupt Republican incumbent governor, Ernie Fletcher, is polling down 2-1. This election is probably framing people’s party identification strongly, and that election is in the papers on a daily basis. This poll illustrates the difficulties that Republicans have in that state. I suspect that once the catharsis of firing Fletcher happens, these numbers will adjust somewhat. Similarly, in Virginia, the state GOP has lost two governor’s races in a row with lousy candidates, and the state party has been rolled legislatively on taxes and transportation issues, the bread and butter. Virginia is now, again, a swing-state with popular Dem leaders.

In other words, Kentucky’s numbers are probably more result of the local environment, while Virginia’s actually represents something bad going on.

Second, these numbers show some important differences. Even in the South, Giuliani is performing in a tier above the other first-tier candidates, while Romney is polling a tier below the other first tier candidates. Romneybots will argue that this is due to name ID, but Gallup polling consistently indicates that Thompson’s name ID is lower than Romney’s but his performance in polling is (often substantially) higher. This is yet more evidence that Romney’s electability problem is real.

The inescapable conclusion is that people know things about Mitt Romney and don’t like him for it.

Third, these numbers are likely to move. People are going to learn things about Rudy Giuliani. (divorces, married first cousin, things about his record, etc.) that are going to move his numbers down. They are also going to learn things about Fred Thompson (thin record, blah blah) and Mitt Romney (flip-flopped on every issue in sight), but they both have the opportunity to frame that first impression.  In other words, Giuliani’s numbers will fall — they are a ceiling — while Romney and Thompson’s can still go up. Some.  The evidence suggests, however, that Thompson’s ceiling is higher than Romney’s.

On the other hand, there is probably nothing to learn about Hillary Clinton. After all, 3 books were written about her recently that were supposed to be interesting. They weren’t, and no one noticed.

In other words, these polls confirm our sense that things in some of these states are weird and that these will be tough elections. They don’t mean that much yet because so few people are paying attention. But we do know where some of the dragons aren’t.