This is a daily feature of this blog. For further detail, please see this post.
| State | RCP | Pollster | 538 | 538 Win % |
| NV | +4.8 | +3.8 | +2.7 | 77% |
| CO | +7.0 | +6.5 | +6.6 | 95% |
| IN | +0.3 | +0.9 | -0.3 | 48% |
| MO | +0.6 | +1.8 | +0.7 | 59% |
| OH | +6.0 | +4.8 | +3.8 | 84% |
| VA | +7.3 | +8.0 | +7.2 | 96% |
| NC | +1.5 | +1.9 | +1.2 | 64% |
| FL | +2.7 | +2.2 | +3.0 | 79% |
The one thing that leaps out at you on today's table is that 538 now gives John McCain a slight edge in Indiana. It probably isn't anything new, however. This state appears to be a real toss-up and should give some real suspense on election day.
Otherwise, Barack Obama's position in Florida moves up a bit, after a Rasmussen poll gives him a 51-47 lead there. I, for one, hope that Florida is not close. I don't care in which direction it's not close, I just don't want a mess.
Also, most of us are considering Ohio as more or less a toss-up state. For instance, Marc Ambinder has it there in his recent post. But look at Obama's numbers there! Obama's actually been dominating the state recently. Ignoring polls that are marked with a (D) or an (R) on pollster.com, and ignoring Zogby's internet polls (which are crap), here are Obama's margins in the last twenty polls of Ohio, as per pollster.com (ordered most recent to least):
+4, +4, +5, +10, +12, +3, +4, +14, -2, +9, -1, (even), +5, +2, +5, +4, -2, +3, +3, -1
I picked twenty arbitrarily and not because the good news stops there for Obama. In fact, even if you go to thirty the news looks good. You have to consider Ohio as a lean-Obama state at this point. If anyone still considers it a toss-up, I'd like to hear why.
2 comments:
Wow...those swings look like the stock market..from plus 9 to down 2, to plus 14 down to plus 4, to up 12 to now, up 4...either the electorate is as fickle as McCain is on his message or the polls are BS...or both. I kinda wish that Colin Powell would have held off on his endorsement until this Sunday...perhaps one last, big up tick to really set a mandate.
Anonymous (2:22), I think you misunderstand the nature of statistical sampling and how they relate to polls. I'll leave it to the math guys on the blog to fill in the details but swings like this aren't in of themselves indicating that the polls are "BS." There could be major events intervening between them or they could result from radically different likely voter filters or some other thing.
Post a Comment