There are some good polling results for Barack Obama today -- but not really in the places where he needs them. Take for instance Rasmussen's poll in Alabama, where Obama now trails John McCain by 15 points after having been 28 points behind before. Or the new SurveyUSA poll in Massachusetts; SurveyUSA finally gives Obama a 13-point lead in the Bay State after he had failed to break single digits in any of the six polls they had released in Massachusetts earlier this year. But Massachusetts is too blue, and Alabama too red, to matter in this election.
In Florida, on the other hand, Rasmussen has John McCain holding steady with a 7-point lead. Rasmussen had surveyed Florida barely a week ago, then showing McCain ahead by 8 points. The main problem that Rasmussen seems to be detecting for Obama in Florida is that the Democratic defection rate remains relatively high -- he's losing 20 percent of Democrats to McCain, and 4 percent to "some other candidate". The puma is native to Florida, I would point out. Nevertheless, since Quinnipiac and ARG seem to feel differently about the state, it might be time to get a Mason-Dixon or a SurveyUSA on the case. Hell, I'll even take an Insider Advantage poll.
Rasmussen also sees no movement toward Obama in Georgia, where he trails in their survey by 10 points, just as he did a month ago. Unlike Insider Advantage, which found Bob Barr polling in the mid-single digits, Rasmussen gives him just 1 percent support. I have been a little inconsistent in my treatment of Georgia. It is a state that Obama could win if a number of things go right, but then again, that's true of probably 40 of the 50 states on the map this year.
Lastly, SurveyUSA has Obama ahead by 2 points in Virginia. This brings SurveyUSA into line with Rasmussen and PPP, each of which show nearly identical numbers, but is a step backward for Obama from SurveyUSA's May poll, which had Obama up 7 points in Virginia. As I argued yesterday, if you see a result that looked aberrant before -- and a 7-point lead for Obama pre-unity bounce definitely looked a little weird -- you sometimes have to ignore the trendlines and take the result on its own merits. Virginia is perhaps the closest state on the map at the moment -- the state whose county names we might all learn by heart staying up late on Election Night -- and all the polling is reflecting that.
Monday, June 30, 2008
Today's Polls, 6/30
-- Nate Silver at 6:27 PM
Labels: alabama, florida, georgia, massachusetts, today's polls, virginia
50 comments
That's some bad news for Obama if we're staying up late to watch Virginia. That means he's been beaten easily earlier in the night in Ohio and/or Michigan or some other states that should have him at about 290-300 electoral votes without Virginia.
Puma jokes going meta in 3...2...1...
Yeah, I like Obama should write off FL and focus on other 2004-red states like CO, NM, VA, OH, IA, AK, and MO. FL has too many old Democrats that won't vote for him (like my grandma).
well, Virginia is EST, so we'll proably know abot it before the result is final in OH or MI
PPP will release a Florida poll, most likely tomorrow, with a result somewhere in between the most recent Rasmussen and Quinnipiac surveys of the state.
Tom Jensen
PPP
What I'd be interested in is, as Rasmussen weights their surveys by party ID while SUSA does not, is the weights Rasmussen uses. Do we know them? Are they being liberal or conservative (not in a political but a temperamental sense)?
Does Quinnipiac weight by party ID?
If we ever have a situation where the pollsters who don't weight are regularly showing a better picture for Obama than those who do, one could conjecture that the party ID model used by those pollsters is out of date and doesn't sufficiently take into account the Dems' growing advantages, the voter registration drives, or whatever else. But I've no clue if we have anything approaching that kind of situation right now.
I wonder why Obama is investing any time or effort in Georgia and I think that will quickly fade as the campaign goes along.
Maybe they have internals that are more positive than the public polling is showing, but I just never bought the Barr Georgia phenomena. Barr just doesn't have any money and without money you can't run a campaign at all.
Even if he ran flat out all the time in Georgia, I doubt his final vote would be enough to matter. I think Obama's campaign is spinning here, trying to psyche out McCain and it's not working.
Yes there are lots of African Americans, but even if you add 30-40% to the African American vote in Georgia, he probably still loses.
I'd really like to see more emphasis on Indiana. I think there's something to the old Confederacy that just isn't going to vote for a black man for President. Every single Union State in the civil war is now Obama territory but in every single rebel state he's trailing or at best tied (Virginia).
Forget Florida and concentrate on states he has a chance to win. What's wrong with Missouri? Why isn't he making more of an effort there? I would think that's at least as likely as North Carolina and a hell of a lot more likely than Georgia.
If Obama wins however, that could totally change depending on how he does. If successful, then all doors are open. If not, it's a setback.
Since wrapping up the nomination, Obama has improved greatly in the south and the border states. We've seen improvements in Alabama, Kentucky, and Tennessee now. Obama was getting slaughtered in the polls there, but now he's not running any worse than any other non-Southern Democrat. I suppose that's good news.
Spike, I hate to burst your bubble, but both OH and MI (except for the UP) are EST as well.
Ohio and all but the 4 counties bordering WI in MI are EST too- so they should be known before VA- since VA seems to be closer- especially since the 4 counties in MI have very few residents
The puma was native to all of the Lower 48 prior to European colonization!
Yes, OH and MI are Eastern, but their polls close at 8:30 and 8:00 respectively, and VA's close at 7:00. So we'll likely know VA first (IN closes even earlier, IIRC).
Can someone post when polls close on election day.
I'm interested in seeing the PPP NC poll.
I am curious about this issue of how much money to spend in various places. I can see both sides of it. GA is the perfect example.
I think the biggest lesson of the primary was its pretty easy to run up the score when the opposition isn't playing defense.
If Obama is in GA and McCain isn't, can he win it?
I believe he can. I believe he can win a couple of these states if McCain leaves them uncontested.
I also believe that he would benefit more in OH, MI, etc if McCain has to spread the field to play defense than if it is a direct battle there (even if Obama has significantly more resources there).
If he can keep GA in striking range, he should totally attack there. He can easily win without it, but I'm not sure McCain can. Its hard for me to picture any scenario where Obama wins GA and loses. Same goes with FL and couple other longshot states.
You do have to be realistic though. If a state isn't within 6 by labor day, pull the plug.
A lot of "Democrats" in Florida haven't voted for a Democratic candidate for President since the 60s. In the last election we saw counties with 75% Democratic registration go 80-20 for Bush. It's not unusual in some of the more rural areas.
I agree. I don't think Florida will be a swing state. Obama should look elsewhere. There are a few ways to 270. For the purposes of this I define "base" as the Kerry states minus NH plus Iowa:
1. Base + Ohio = Win
2. Base + Virginia = Win
3. Base + NH + Colorado + Nevada = Win
4. Base + NH + Colorado + N Mexico = Win
I wonder if the media will mention nearer the election (if it remains close) that a tie is certainly possible. Obama wins the Kerry states + Nevada + NMexico + Iowa = 269. Certainly very possible
I like the new senate scorecard, but it would be even better if we could see the distribution of actual possibilities, like for the presidential election. It's easier to gauge the mean, median, and standard deviation that way.
Cugel,
I see the point you're making about union vs. confederacy states, but I have to nit pick. First of all I assume you're not counting border states (Missouri and Kentucky) that remained in the union but permitted slavery. Second I assume you're not counting free states that joined the union after succession (Nevada and West Virginia). But even so, Kansas was admitted to the union as a free state nine days before the Confederacy was declared and months before the war actually started with the attack on Fort Sumter.
Sorry for the brief (and random) history lesson.
If I were in charge of ad buys for the Obama campaign my priorities would be VA, MO, ND-MT-AK, IN and NC. That's 28 electoral votes from southern states and 31 from non-southern.
Obama may have 2-3 times as much money as McCain. I hope he does invest some money/time in - well not really a 50-state strategy - but in every region of the country.
This is a historic opportunity to build a democratic party that can compete on a national-level once again. If he doesn't invest in that possibility the moment will pass.