Monday, October 18, 2010

Missouri Senate close

The Missouri Senate race is getting closer, with Robin Carnahan pulling within 5 points of Roy Blunt in a new PPP poll conducted for her campaign. Blunt's lead is 46-41, in contrast with the 45-38 advantage he had when we last took a look at the race in August.

There are two things driving the increased competitiveness of the race:

-As is happening across the country Democratic voters are getting more interested in this year's election as the big day comes closer. In August only 33% of those describing themselves as likely to vote in November were Democrats, while 38% were Republicans. Now the likely voter pool is composed of 36% Democrats and 35% Republicans. While Republicans have been extremely excited about voting all year, many Democrats are just now starting to tune into the election. This is causing many races across the country, including this one, to tighten down the stretch.

-Roy Blunt's support from Democrats has decreased since August, while Robin Carnahan's support from Republicans has seen a slight increase. When PPP last looked at this race there was a large gap in the two candidates' crossover support, with Blunt getting 11% of Democrats while Carnahan had only 4% Republican support. Now Blunt is getting just 8% of Democrats with Carnahan up to 6% of Republicans, essentially erasing the gap in party unity that is causing many Democratic candidates across the country trouble. Blunt continues to lead overall thanks to a 46-31 advantage with independents.

Some pundits have written off this race as an opportunity for Democrats to pick up a seat but Carnahan is within the margin of error and picking up support and if her party's base continues to awaken in the final 15 days before the election this race could provide a surprise.

Full results here

26 comments:

Mark said...

Stunning results here. I thought Secy. Carnahan was down for the count, I really did. I still expect she'll come up short when the chips are laid down, but it's going to force the NRSC to keep spending here.

Chuck T said...

Get ready for the inevitable "Every other poll has Blunt up by double digits" crap.

Anonymous said...

"Interest of disclose: our Missouri poll was done on behalf of the Carnahan campaign"
-PPP tweet

Keep humping that Robot Chicken.

Unknown said...

I respect you guys due to your transparency, but your polling results are perplexing. Only Democratic campaign internals are skewing more Democratic.

In August your own poll had the election 38% Republican/33% Democratic. Now you have it 36% Democratic/35% Republican. You're now showing a more Democratic electorate than 2004 or 2006. I don't see Republicans becoming less interested in the election.

What's even more weird is that in August you had a 51% McCain/44% Obama electorate. Now it's 53% McCain/45% Obama. So while the electorate has similar numbers of McCain and Obama voters, and McCain voters are even bigger, the McCain voters now contain more Democrats?

Someone else/don't remember has dropped from 6% to 2%. Yet that's now reversed and, I guess, accounts for Carnahan's improvement.

Polls like this one and the Arizona-3 poll you did for Kos are exciting Democrats and might help them at the polls. That seems to be more your mission, rather than accurate polls.

Chuck T said...

DBL said...

Polls like this one and the Arizona-3 poll you did for Kos are exciting Democrats and might help them at the polls. That seems to be more your mission, rather than accurate polls.

yeah, but you accept out of hand Rasmussen and Fox polls (which are done in association with Ras)? If you say that PPP is trying to help spur on Dems to the polls, don't you think that Ras and his gang has been trying to write the election scenerio all year? PPP has had a very good track record this year and we will see in about two weeks if Dems in deed have woke up or if they do stay home.

NRH said...

Arizona-3 has a particularly bad Republican candidate who had an extremely hard-fought and bitter primary. The demographics in that poll had a less Republican sample than 2008, yes - but there's not a presidential candidate from Arizona on the ballot this time, and this is entirely in line with polling showing Illinois and Alaska reverting to the mean without a presidential race, too. If you want to accuse PPP of partisan polling, you might want to provide some evidence - something that can stand up to PPP's excellent track record this year.

Anonymous said...

This is a good news for Carnahan's campaign. People begin to realized the difference between Carnahan and Blunt.. very simple, a candidate that will defend the hard working middle class or a candidate that has a long record in congress as a Wall Street butler... Carnahan or Blunt.. your choice, your future

JGibson said...

Let's not forget that Beck (CP) and Dine (L) are snatching away some of Blunt's votes, thereby making the race closer.

Jo Ann Brown said...

Robin needs to make it to St Louis and Kansas City. We don't see her.

Anonymous said...

Polling on behalf of the Carnahan campaign:

Blunt +5

When your best just isn't good enough...

Shadrach said...

Attacking PPP's polling is really getting old, they don't skew results and they have been the most accurate. Just because you don't like the result, doesn't mean that it's wrong, things don't work like that.

Anonymous said...

Ha! PPP polling.

Anonymous said...

I believe PPP polls, not internal polls, same is why I also don't believe the PPP poll for Grayson that has him up by 12 pts in FL-8.

AG said...

You want to talk about setting a narrative, every single issue poll released by Rasmussen has been a Republican Party talking point. Then they have the audacity to only ask it of "likely voters" as if Americans who don't, or can't, vote are incapable of having an opinion.

Maybe next week they'll ask "Why are Republicans so freaking awesome?"

Unknown said...

Faced with such incredible evidence, how will I hold up? I don't trust Rasmussen, because they aren't forthcoming with their in tabs. There's no transparency.

Rasmussen doesn't skew Republican. I know you read that on the web, but it's the second most Democratic skewing pollster of the five that do enough polls that we can measure it. PPP is the most Democratic skewing. That is evidence that intentional or not, their polls are consistently 1.5 points to the left of Rasmussen, 2 from CNN/Time, and more from Quinnipiac and SurveyUSA.

SurveyUSA and Quinnipiac have been far more Republican than Rasmussen. Really far. SurveyUSA's in tabs have looked so crazy that I'm skeptical of every poll they come out with. Their CA-11 poll recently is a case in point.

PPP's party breakdown for Arizona-3 is 45% Republican/30% Democratic/25% independent. So, contrary to what you're saying, the sample is very Republican. And the 2008 Republican candidate IS on the ballot. If you look at the poll, you'll see that he leads Rodney Glassman 53-39 in this district. The poll results say that Hulburd is not only getting 2008 McCain voters. They're getting 2010 McCain voters and 2010 Brewer voters. Are people going to check McCain/Brewer/Hulburd? Really?

What is PPP's excellent track record? Can you please provide me with an analysis of PPP's election results compared to the other pollsters to back up your statement? How are they the most accurate? How do you know they don't skew their results? This poll is done for a campaign. Nate Silver of 538, a good lefty, has studied campaign polls and finds them 6 points away from independent polls.

Just because you like the result, doesn't mean that it's right. Things don't work like that. I know all you Democrats believe anything PPP puts out because you want to believe it.

I'm skeptical of everybody's polls. I don't take anything at face value without first analyzing it.

Charlie Cook has 115 Democratic seats in play. RCP has 133. Republican campaigns have been flooding the zone with polls showing uncompetitive races competitive. I think both of these experts are falling for this and reporting the races as competitive. As a result, Democrats are spending money in districts they shouldn't. They just made a big buy in AZ-7, a district that isn't flipping. They've taken the Republican bait and fallen for it.

Unknown said...

Americans who don't vote have opinions, but since they don't vote, asking them who they're going to vote for is moot.

RenaRF said...

Ground game! One thing that very people are actually talking about is the get-out-the-vote infrastructure that exists in many of the states with competitive races. I don't believe a great GOTV effort can overcome a major Republican lead in any given race. But where Democrats can close and/or keep it tight, a GOTV infrastructure will make all the difference. I believe I even heard on CNN that NRSC and NRCC "officials" acknowledge that their GOTV machine is lacking.

Too bad polls can't reflect that.

NRH said...

McCain is on the ballot, yes, but he's not a presidential candidate this year. He had a 'favorite son' effect boosting turnout in 2008 in Arizona, the same way Illinois had an extra-Democratic year and Alaska had an extra-Republican year. This year he's competing only within Arizona, not against someone from another state, and does not produce that same boost. It's frankly silly to pretend that a state that significantly outperformed its usual partisan margin due to the presence of a candidate from that state on the ballot would sustain the outperformance in an off-year.

It's entirely unsurpising that Quayle, after his bitter primary and awful campaign, is unable to hold voters who go for Brewster and McCain. Not everyone votes a straight ticket, particularly when there's an awful candidate who offended some of his own party members in his primary.

As for Rasmussen - ahem. Rand Paul by 25. Angle's post-primary 'bounce.' Paladino's. Whitman. Rasmussen has a long history of 'narrative-setting' early in the race and then adjusting his numbers closer to reality late in the game so he can make a claim to accuracy. He makes no pretense of being non-partisan.

As for PPP's track record, look at their primary polling this year. They got PA-12 when nobody else did. Their 538 rating is in the top ten.

Unknown said...

It actually is entirely surprising. In most polls and most of my interviews with Republicans and independents leaning right this year it's not that much about the candidate as it is about rejecting the Democrats. People in Nevada talk about dumping Harry Reid, and not much about Sharron Angle. If this district is 45% Republican/30% Democratic, Quayle wins. It's the reverse of Delaware and the reason why O'Donnell stands no chance there.

You decide to refute none of my Rasmussen analysis, but instead cite post-primary bounces. You missed Marshall's big bounce in North Carolina, a poll that went against the Republicans. Rasmussen conducts their polls in one day without call backs. A poll conducted the day after a primary is likely to produce unusual results. Outside of that one poll, Rasmussen was fairly consistent with others on Kentucky.

Rasmussen has a long history of narrative setting if you read Swingstate. If you do an analysis of Rasmussen's likely voter polls compared to other likely voter polls you find that the two closest pollsters are Rasmussen and PPP.

Can you please provide detailed statistical evidence that Rasmussen changes their numbers late in the game to show reality? I mean, besides someone writing it on dailykos? Nate Silver at 538 has shown Rasmussen has a very good track record. Until I see your analysis I'll go with Nate.PPP got one poll right. That's statistically significant to show... nothing.

Unknown said...

Neither the NRCC nor the NRSC are responsible for GOTV. GOTV is done by state and local parties. I spent the weekend with two GOP campaigns and they were very aggressive in their GOTV. I never heard the words "NRCC" or "NRSC."

Republicans don't need a strong GOTV to do well this year. Every door or phone call where someone was supporting a Republican was very motivated. My favorite was the woman who told me her daughter, son-in-law, and grandson would all vote Republican this year. "You have four votes in one household!"

AG said...

This whole idea of setting a narrative is silly. I don't let the media tell me what to think, and I'm not going to let a pollster tell me whether its worth it to vote. Next I suppose people will stop going to football games because AccuScore says their team has no chance of losing.

Anonymous said...

Sure, guys - that's why the DSCC just pulled out of MO. 'Cos their internal polls show only a 5 pt race, and an electorate that's even more Democratic than the general partisan breakdowns in MO... in a Republican year. Whatever.

keetz4 said...

@ DBL......I've got news for you. I am doing GOTV for a Dem candidate in a swing district. Today, I was talking to independents in the district and some were very enthusiastic and couldn't wait to vote for my candidate. I know what the media meme has been which is all the enthusiasm is on the R side of the aisle but that's not what my experience in the field is showing me.

However, I didn't get a household with a daughter, son in law, grandson and mother all voting the same but then Dems don't have the same CLOSE family values like you all. ;-)

Dustin Ingalls said...

"What is PPP's excellent track record?"

Our 2008 record is proven. As far as this year:

Only pollster to call Christine O'Donnell upset, and only one of two to predict Rick Scott win. Closest to calling Kendrick Meek's margin of victory. First to show Scott Brown winning, and final poll had him winning by exact margin by which he won. Nailed Patty Murray's margin over Dino Rossi in WA primary exactly.

Chuck T said...

For those of you doubting this poll because it was a campaign poll I'd like to point out to you that today Fox/Pulse (an off shoot of Rasmussen) came out with a MO senate poll which is identical to this--except I believe they have Carnahan down by 6 but an improvement from the last poll. It seems that often PPP is ahead of the curve and then other pollsters confirm it.

Dustin Ingalls said...

"It seems that often PPP is ahead of the curve and then other pollsters confirm it."

Tends to be the case, yeah. I can tell you also that we're not going to be the only ones in the coming days showing Sestak gaining serious ground.

 
Web Statistics