I agree with the basic premise that the fork is far from in, but I think too much is being made of the convention bounce. Even if the current bounce lasts, unless McCain can gain even more, he’s sunk.
Can we all agree that state polls are what’s relevant here? Good, then let’s take a look at the post-convention polling in important states (as summarized by 538.com):
Now, those changes are relatively small. In fact, if you’d asked Obama if he willingly trade 1-2 points in OH, MI, VA, and PA for 2 in FL and 4 in CO, he probably would have agreed to the deal.
That analysis doesn’t take into account every poll out, but by all indications, Obama is still up in all the Kerry states plus Iowa. So he only needs any of the swings: CO, VA, OH, FL, IN, etc. He is still up in some of those, and actually gaining ground in FL and CO–despite the convention and Palin.
I don’t think he is. I wish he were. But I don’t think he is. I’ve said I ignore most polls at this point, but I can’t deny that it’s polling close, no matter who’s in the lead at any given moment. I think we’re still in the phase of the horserace where the racers are going to keep trading back and forth until the country gets a really good look at them side by side in the debates.
I can hope that they’ll come out for Obama over McCain. But I’m not psychic, so I don’t know at this point.
Thanks. It has been an interesting experience to leave for few months and come back. It’s like when you’re heavily editing a paper, if you stop looking at it for a couple of days, when you come back you notice new things and have a slightly different perspective.
National Polls Are IRRELEVANT and have never, ever been a good predictor of the outcome in Presidential elections because they’re only right 50% of the time.
Written by Dr. Alan Abramowitz, who is the Alben W. Barkley Professor of Political Science at Emory University, and the author of Voice of the People: Elections and Voting Behavior in the United States.
I’m not sure I understand this. If you’re arguing that the relative paucity of state polling means that, for any given state, we cannot be as certain about the state of the race as we are for the nation as a whole, I’m with you. But for the general thesis that the post-convention state polls that suggest that McCain’s national gains are mostly coming from very red states, the current polling seems more than adequate. In short, any given state poll might be an outlier (just as any given national poll might be), but the sum of all the state polling can back a thesis about the race just as much as the sum of the national polling, right?
Meh. The only must-wins in that group are WI (Obama +3), NJ (Obama +6), and MI (Obama +1). And MI is only a must-win if Obama can’t win FL.
I agree that the numbers aren’t great, but the point of my post is that Obama is still winning in the electoral college despite this bounce. From what I’ve read, the bounce typically represents a ceiling for a campaign. If this is McCain’s ceiling, he’s in trouble.
I just wanted to add that I really like Knorf’s word “bounch” and want to use it for something - maybe a bunch of bounces? Like McCain got a bounce both from announcing Sarah Palin and from the RNC Convention? So that would be a bounch?
I won’t say the polls are irrelevant, because they’re not. They’re a measure of who’s currently ‘winning’, but historically the polls at this point in the race aren’t very predictive, especially if they’re close.
This race is going to break on two different things:
Gaffes and scandals. He who Gaffes worst, loses. Both sides will have their share - I think there were about a half dozen this week alone, on both sides. But sometimes a gaffe takes on a life of its own because it touches a nerve or reinforces a fear. Let’s call it Gaffe Russian Roulette - each time a candidate puts his or her foot in it, they’re pulling the trigger. Any one of those could ‘go off’.
The debates. Reagan didn’t break away and build his landslide against Carter until after the last debate. It’s said that his biggest problem was that public was just unsure of him personally - they really wanted someone other than Carter, but didn’t quite trust Reagan. He killed Carter in the last debate, and sealed the deal with the electorate, who then broke for him big time.
The same could be said for this year. The public has doubts about Obama, but they also have doubts about McCain. Neither has ‘sealed the deal’. Palin throws a wildcard into the mix. So this race is probably going to be more volatile than any we’ve seen for a while, and I suspect the debates are going to be a huge factor.
And this scares me a little. Obama wins hands down on rational, nuanced policy with most people (I understand this doesn’t include you, Sam, or other true ideological conservatives). But for the time being, McCain seems to be doing just fine running on truly empty rhetoric and applause lines. I mean the really meaningless stuff, that when you try to figure out what he actually was stating, you realize there was nothing there at all, except “I’m selfless and patriotic and a maverick, and Obama isn’t.” And for the time being, the public seems to be eating it up. Nobody’s pointing out to his face that every “buck the party” position he held, he turned his back on, except for the surge, where he was able with others to persuade the government to move his way. Nobody is calling out Sarah Palin’s lies and exaggerations to her face about her supposed reforms and economies while mayor and governor.
So I’m concerned that at the debates, Obama will thoughtfully explain policies, and McCain will get out meaningless zingers. McCain is smart, even though he’s not quite in Obama’s league, and he’s a funny man who knows how to tell a story. He’ll tell charming little anecdotes about his past that ultimately will have nothing to do with the question asked (and generally will work in the fact that he was a POW). And no one is going to say “But that has nothing to do with the question I just asked, Senator.” or “But you’re not actually saying anything there, Senator.” Obama won’t do much of this, both because it’s beneath him morally and intellectually, and also because it’s not his shtick: even though he never called his campaign the “Straight Talk Express,” that’s really what he’s tried to run. And I’m afraid that the Americans who do not necessarily ideologically buy into the conservative agenda will nonetheless believe that McCain will have “won” the debates because he will tell his little stories and get off his meaningless little zingers.
I’m too old (unless I were rich) to emigrate, and I have elderly parents anyway. I can’t leave. But if McCain wins, God, will I wish I could!
Wow, I’ve never coined a new word before. “Bounch.” Starts as a typo, becomes a word? Is that possible?
I’m not that impressed yet with McCain/Palin’s bounch. We’ll see how things are next week. I still suspect that the Palin pick is not going to bring McCain any independents.
Of course it’s possible. All we have to do is start using it. Everywhere, not just on the Dope.
There was an episode of a fairly short-lived (two-three seasons, I think) show called Veronica’s Closetwhere the PR guy made up a word “acribitz” for one of financial reports, as in “profits will continue to acribitz over the next six months.” When someone who cared indignantly brought it to the owner (Veronica)'s attention, she responded “What, do you think they’ll decribitz?” The entire B story tracked the use of this word spreading until finally it was used by the anchor on the evening use. The original complainer was clearly on the verge of suicide, while the coiner was utterly gleeful.
No, my point is that, as a group, the most recent state polls are older than the national polls. We have up-to-date polls for some states, but not for others. Swing state NH hasn’t been polled in three weeks. Potential swing state ND hasn’t been polled in nearly two months. Oregon hasn’t been polled in over a month: it seemed to be fairly safe then, but how about now? We’re flying blind. NV and NM, both swing states, haven’t been polled since just before the conventions. You get the idea.
So we’re looking at 50 state races, but in many cases we only know where they were several weeks ago. All we know is that the national polls have moved. Voter preferences in these states may have moved with the national polls, but by the time we find out, the national polls themselves will have moved on.
It’s kinda like the cos and sin functions - the latter is always a quarter-turn behind. That’s how I see national polls v. the sum of the state polls. If McCain’s post-convention bounce in the national polls peaked/peaks yesterday or today, I’d expect McCain’s post-convention bounce at electoral-vote.com to peak in another week or two.
The point of mine would be, his EV count hasn’t yet peaked. He’ll be rising for awhile yet.
Gotcha. In other words, while we have recent polls for most of the swing states, some of the more leaner states may have shifted. That’s true, as far as it goes, I suppose. Though one important note is that we do have polls in the remotely swingy states that most people think create the easiest route to Obama victory. And one of the beauties of the 538 model is that one can predict the results in unpolled states based on demographic similarity. Obviously this isn’t as predictive as actual polling in the state. But I think we have enough state polls from different regions that we can support the theory that much of McCain’s bounce has occurred in non-swing states.
Yes, and further, we’ll see how that works out now that Palin is going back to Alaska - McCain will be stumping by himself again. Let’s see who draws the crouds. I don’t think his bump is going to pull through.
But 538 sure seems to say that much of it has occured in swing states as well. VA and OH have gone from light blue to pink. NV has gone from white to pink. CO has gone from light blue to white. MT has gone from light pink to bright red. ND has gone from light pink to dark pink. NH is a very, very light blue. That’s seven swing states right there.
States with 265 EVs are varying shades of red, and states with 264 EVs are varying shades of blue, on Nate’s map. Nate’s counters say the popular vote’s split at 49, the EVs are 275.1-262.9 for Obama, and Obama has a 52.1% chance of winning.
I bet those numbers look worse after tonight’s update. Takers?
Actually, it should be interesting to see when Nate’s model peaks for McCain, as compared with the RCP average and the daily trackers on one hand, and Electoral-Vote.com on the other.
First, the evidence catching up with the reality. The state of the race in various places is not being perfectly reflected by the polling because there isn’t enough polling. On this dynamic, it seems reasonable to expect that there won’t be much more change* in those states that have been polled* (perhaps a few points here or there as we get more robust averages). If these states turn out to be predictive of others as they so far have been, then we actually know somewhat well where the race is at this moment.
Second, the momentum of McCain’s gain itself. The notion here is that since polls are a snapshot, the next snapshot is likely to continue along a similar regression line. This is a matter of pure speculation, really, except for the empirical analysis of post-convention bumps. What limited evidence we do have (e.g., history), suggests that a few days after the convention is the apex.
Do you think both dynamics will continue to trend McCain, or just one of them?
