February 25, 2008
Has Obamamania Peaked?
Late last week, I noticed some trends in the Gallup tracking poll. Specifically, Barack Obama was no longer gaining on Hillary Clinton in tracking polls, and indeed seemed to be losing ground. I noted that the downturn in his poll numbers seemed to have coincided with the beginnings of Clinton’s attacks on him, and the increased media scrutiny he was drawing with his status as a frontrunner.
At the time, I cautioned that there simply wasn’t enough data to draw any conclusions one way or another. Some additional data points seem to indicate that, indeed, Obamamania reached a peak shortly before the WI primary. Whether it reached the peak is a question I can’t answer, but for now, it has become somewhat more likely that Clinton will eke out a narrow win in TX and a more substantial win in OH.
This is what the Gallup tracking data look like to date:
2/3: Clinton +4
2/4: Clinton +5
[Super Tuesday] 2/5: Clinton +13
2/6: Clinton +11
2/7: Clinton +7
2/8: Clinton +5
2/9: Clinton +5
2/10: Clinton +2
2/11: Clinton +1
[Potomac Primary] 2/12: TIE
2/13: Obama +1
2/14: Obama +2
2/15: Obama +7
2/16: Obama +5
2/17: Obama +7
2/18: Obama +1
[Wisconsin]: Obama +5
2/20: Clinton +1
2/21: [No Polling]
2/22: Obama +2
2/23: Obama +1
Having traded leads in the single-digit range now for four of the last five polling observations, it appears that Clinton has blunted the momentum Obama was gaining shortly before the WI primary, when he steadily rose in the polls. As I noted, this is consistent with the exit poll data from WI which showed that the bulk of Obama’s support came from people who had made up their mind 2-4 weeks before the voting began. Gallup has included a nice graph of these data, which I have stolen appropriated borrowed here:
Rasmussen is showing a similar phenomenon. Now understand, Rasmussen and Gallup sometimes seem like they are polling different races. On the Republican side, Rasmussen currently shows McCain with a 23-point lead over Huckabee. Gallup shows a 45-point lead. What matters isn’t so much the absolute numbers as the direction of the change, if any. Do they move in tandem? It is less clear from the Rasmussen numbers, but it still appears that Rasmussen has shown that Obama is no longer gaining on Hillary, and indeed may be slipping a bit. Over the same time period, the numbers look like this:
2/3: Clinton +11
2/4: Clinton +6
[Super Tuesday] 2/5: Clinton +7
2/6: Clinton +4
2/7: TIE
2/8: Clinton +4
2/9: Clinton +6
2/10: Clinton +8
2/11: Clinton +5
[Potomac Primary] 2/12: Clinton +2
2/13: Obama +5
2/14: Obama +12
2/15: Obama +8
2/16: Obama +4
2/17: Obama +3
2/18: Obama +4
[Wisconsin]: Obama +4
2/20: Obama +7
2/21: Obama +5
2/22: Obama +3
2/23: Obama +1
Significantly, both pollsters show movement at the same time: Clinton generates consistent leads pre-Super Tuesday, which turns to momentum for Obama shortly thereafter. Both show Obama taking the lead the day after the Potomac primary. And both show Obama hitting a high point of support of around 49% on around 2/14 (2/15 in Gallup), before slowly declining into the mid-40s.
Again, this roughly correlates with Clinton’s decision to go negative, the breaking of the so-called plagiarism story (which I personally find to be ridiculous — you don’t footnote a political speech, unless you are pulling a Biden and wholesale stealing a speech without permission), and continuing through a fairly rough week last week.
Which all goes to this point: State polls tend to lag federal polls in how quickly they come out. In OH, two polls: SUSA and ABC/WaPo, were conducted shortly after 2/14, when Obama was only beginning his decline. They show Obama at 43. Two polls have been conducted largely after the Obama decline began nationally, and they show him at 40. The most recent poll — the Q poll — was taken from 2/18-2/23, and I am very interested to see what the day-by-day breakdown was, although with 130 people taken each day, I’m not sure you could draw any helpful conclusions.
With TX, the situation is even more difficult to read. All the polls except for Rasmussen were taken in the pre-WI peak period.
At any rate, OH shows HRC with a roughly 10-point lead, while TX shows a much narrower lead. There are some indications, however, the HRC may have staunched the bleeding, and that her leads may hold. Obviously there are other questions: Will TX and OH — where the candidates are most active — follow the national trend? And this one: Is it too late? Has Obama gained an aura of inevitability that is causing the punditocracy to write Hillary off prematurely? Today on RCP I see articles from the Left and from the Right declaring Clinton’s campaign over. Conventional wisdom is like a battleship: It is difficult to alter its course, but once it has changed, it then difficult to change it back. Will a narrow HRC win in TX and a solid win in OH be enough to change it back? This may be the most important question of the campaign right now.
fn — Personally, I think that the importance of the pledged delegate count is overstated. If HRC win TX, OH, and PA, I really have a hard time seeing her not winning the nomination.
http://race42008.com/2008/02/25/has-obamamania-peaked/trackback/
February 25th, 2008 at 11:56 am
HUCKABOOOOOOM!!!!
(just kidding)
February 25th, 2008 at 12:12 pm
Hello, Sean-
A well-researched analysis. However, I must disagree with your point at the end:
“If HRC win TX, OH, and PA, I really have a hard time seeing her not winning the nomination.”
Assuming, for the sake of argument, that HRC does win all three:
Obama is likely to win all the others, many by large margins, such as North Carolina and Mississippi. He is the only one with money left to spend and perhaps more importantly, the prospects of being able to raise more. He has more donors in absolute terms and moreover, among his larger donor base, a lower percentage of maxed-out donors than HRC.
Even if HRC manages to win in Texas statewide, her delegate haul will not be very high, since under the Texas rules, the state Senate districts that have voted the most heavily Democratic - the cities - get a disproportionate number of delegates.
She has to win OH and TX and PA just to stay in the game….
February 25th, 2008 at 12:28 pm
I should have explained this more. First, a lot of this is about momentum. If HRC wins OH and TX — and either one by large margins — the story will largely be that she has blunted his momentum. This will invite even more scrutiny of him, and whether he can take a punch. The fact that Obama has started hitting back for the first time in the campaign in the past couple days makes me think their internal tracking numbers are showing similar results.
The one poll out shows her leading RI and getting beaten in VT. This would keep the same knock on Obama alive: He does well with Limo Liberals and blacks, not so much with the white working class.
After that come the WY caucuses and MS. He’ll win both, but that doesn’t help his case. He’s supposed to win both.
After that is the PA primary. It is almost a month later. A win there for Hill would mean that she has won seven of the ten largest states, including all the swing states in the category, and 12 of the largest 20 (with Obama winning only two swing states — WI and VA).
On May 6, there will be primaries in IN and NC. Assuming momentum hasn’t shifted one way or the other, HRC likely wins blue collar IN, while BHO wins NC. NC will be interesting — there are still a lot of whites in the Dem party (unlike AL or GA) and there won’t be as many affluent whites as in VA (though there will be some).
On May 13, we are in HRC territory in WV. May 20 is KY and OR; KY is likely HRC territory, while OR could actually go either way — Portland is solid liberal Dem territory, but outside of that is pretty blue collar (which is why it is always a narrowly-won purple state). That leaves us with primaries in MT and SD, and a caucus in PR.
In other words, if she can hold BHO in TX and OH, I don’t see where he gets momentum back. He’s pretty much out of definite big-win states, and she’s set to do pretty well after that.
There’s a number of variables — she could have overtaken him in the popular vote if you include FL and MI (which strikes me as more relevant than the delegate count with FL and MI). Moreover, I think if he can’t put her away in the condition she’s in now, people will question if he can put McCain away.
We shall see. Interesting stuff.
February 25th, 2008 at 12:34 pm
And yes, I actually do think that this is somewhat akin to the Huckaboom. Except it is occurring in February.
February 25th, 2008 at 1:20 pm
Sean,
I too think Obama’s bubble will burst between now and Nov., and that many voters are “dating” him right now. When the McCain campaign starts hitting him on his radical leftist voting record and his paucity of experience, the tide will change.
February 25th, 2008 at 1:32 pm
Note Rasmussen’s daily General Election matchup polling….McCain has been ahead of Obama for the last 3 days, after a long
period of being behind Clinton and ahead of Obama; he is now ahead of both of them by the same margin…..
Does Obama have a glass jaw?
February 25th, 2008 at 2:12 pm
I agree with Sean that if Clinton wins CA, TX, NY, FL, PA, OH, MI, and NJ it will be tough to deny her the nomination. She will have won 8 of the 9 largest states electorally and if you add the electoral votes of the states they’ve won she will have gotten past 270 while he does not.
For the GOP the best thing is fir this to go on as long as possible to wound the dems in the fall.
Does everyone think she has to win both OH and TX to continue? Would she stay in for PA if she split the 2?
February 25th, 2008 at 2:35 pm
I think she’d probably stay in if she narrowly loses TX and has a solid win in OH.
February 25th, 2008 at 3:05 pm
Has Obamamania peaked???
I CERTAINLY HOPE SO.
February 25th, 2008 at 3:08 pm
One thing you have to realize, is that if Ohio and texas keep Hillary alive and then she wins the nominaiton, she will be indebted to MIDDLE - CLASS - WHITES. Not the limo liberals and black special interest groups.
February 25th, 2008 at 3:09 pm
THAT would be “one” OK thing about an HRC presidency.
February 25th, 2008 at 3:18 pm
I dn’t think Hillary could beat McCain if she wins.
The dems need 85%+ of the black vote to win the WH. They have since 1964.
If Hillary takes down Baraack at the convention with super delegates, and that’s her only chance at this point, I think blacks would be rightly peeved and black turnout in November would plummet.
If McCain puts JC Watts or Colin Powell or Condi Rice on the ticket, he’d beat her by double digits easily.
February 25th, 2008 at 4:32 pm
wow colin powell…. never thought of that BUT IT WONT HAPPEN. Colin powell has already declared that he may support the dem.
February 25th, 2008 at 4:58 pm
Sean,
How do these relate to your hypothesis?
http://blogs.cqpolitics.com/polltracker/
February 25th, 2008 at 5:54 pm
Tano,
It’s hard to tell without better apples-to-apples comparisons, especially for OH. But I tend to think these polls are consistent with the idea that Hillary has blunted his momentum somewhat. It appears, however, that it might be too little, too late to save her in TX:
OH: You have the Q poll, which I reference above showing an 11-point Clinton lead (unfortunately, the previous iteration of the Q-poll, showing a 20+ point lead, was conducted around Super Tuesday, so we can’t test recent momentum). The UCincy/OH poll (no previous iteration), taken from 2/21-2/24 shows an 8-point Clinton lead. These are consistent with the results from the 2/17-2/21 polling from SUSA , ABC/WAPO, and Rasmussen, and would be consistent with the thesis that Obama’s momentum has stopped.
There is also the Public Policy Polling piece showing a 4-point Clinton lead. This poll was taken from Feb. 23/24. Simply put, it is inconsistent with the idea of Obama’s momentum being blunted in the last few days, though it would be helpful to see how they saw the race a week ago. We’ll see if this is an outlier or a harbinger of a trend. It’s a North-Carolina-based Democratic polling company. The “Dem” part shouldn’t be a problem, since it is a Dem polling company, but the fact that it is an NC-specialty company polling a state I’ve never seen it poll before raises something of a red flag in my mind. We’ll see.
As for the TX polls, ARG has shown Obama leading Clinton since Valentine’s day, when everyone else had her up by high-single digits. Simply put, they see Obama performing better with Hispanics (see the lengthy post I linked to in my subsequent post discussing TX polling). But regardless, the fact that Obama has only extended his lead by two points in the past ten days — well within any error margin — is at the least not inconsistent with the theory that she has succeeded in slowing his momentum. Similarly, Rasmussen shows Hillary down a point and Obama up a point over the past week, compared to her spectacular 13-point collapse in the previous week.
The one poll they don’t mention is the new CNN poll showing him up four and her down two in TX. This does show some movement on his part, and while it could just be noise, it could also be signs that he’s still on the move.
Regardless, it is important to remember my caveat: It is possible that they could have reached stalemate as a whole nationally, and still have some movement at the state level. Which could be what is going on in TX.
BTW, nice catch on their error margin mistake. Pretty embarrassing for CQ to make that error.
February 26th, 2008 at 10:11 am
Let’s hope Obama has peaked and Hillary too. FIGHT ON DEMS, FIGHT ON.
February 26th, 2008 at 12:31 pm
Sean… With all due respect, I think the tie goes to Obama right now. He has the pledged delegate lead, the overall delegate lead, the popular vote lead, and the most states won. He’s currently only about 60 peldged delegates behind her (with less than 400 left to pledge). It’s now his to lose and Hillary would have to do more than stop his momentum and keep it close the rest of the way.
Even in your analysis, at worst Obama will win 3 of the next 6 states (possibly 4 if he pulls off Texas) and it is becoming increasing unlikely that Hillary will win by enough in Texas or Ohio to dent much of his lead.
After that… She will likely win Pennsylvania by about the same margin she wins in Ohio. I am not as confident that Indiana (right next to Illinois) is an easy Clinton win, nor do I expect that she will necessarily do well in Montana, South Dakota, or Oregon. Obama will win North Carolina which offers over 100 delegates.
The reality is once March 4th is over… the race is effectively 80% over. There just isn’t enough states or delegates left for her to overcome the deficit. I expect over the course of the next couple of weeks that more Superdelegates will commit both ways (again, an even distribution helps Obama). Unless Obama Super delegates start leaving him which I don’t see happening it will be increasingly difficult for Hillary to mathematically win enough delegates to win the nomination.
February 26th, 2008 at 12:34 pm
opps - should say less than 60 Superdelegates with less than 400 to commit.