Poll Analysis and Election Projection
Once again, the polls were pretty close. Mitt Romney’s ground game carried him to overperform. Gingrich underperformed. A 10 point advantage became 14, and Romney approached an absolute majority closer than I imagined he could.
Romney sweeps Florida’s at-large delegates and takes a 66-25 lead among pledged delegates.
I normally like to let these races rest a week before I start looking into the polling, but thanks to one pollster sitting on a pre-primary poll to drop the bomb after the primary, I thought I’d hit today the actual first post-primary poll of the race between Democrat Chris Coons and Republican Christine O’Donnell.
SurveyUSA polled the North Carolina Senate race for WRAL, and the result is striking. In fact, it’s entirely unlike any other poll I’ve seen of the race between Republican Richard Burr and Democrat Elaine Marshall.
This poll makes Burr’s last big polling jump look small and timid.
Real Clear Politics showed 21 polls yesterday, plus we’re already at 5 today, so I’m grateful that some of them can be done in batches as with this bunch on the Nevada Senate race from Ipsos for Reuters, Rasmussen Reports, and POR for Fox News.
If we color Republican advantages in red, leads for Democrats in blue, and ties in green, then this set is a rainbow of results. Not a full rainbow or a double rainbow though, sorry.
I’m beginning to think that it may be time to scale back my near-obsessive coverage of the Ohio Senate race. It’s been a rollercoaster, but Democrat Lee Fisher has not shown a lead over Republican Rob Portman since June.
The three new polls out, from Rasmussen Reports, SurveyUSA, and Fox News/POR don’t change that, either.
Colorado is one of those states where the Republican Party had done pretty poorly lately. It looked like this year was the year of mending though, but it’s hard to see that in the Governor’s race. Democrat John Hickenlooper leads Republican Dan Maes easily in the new Rasmussen, but that’s because of a third party protest run going on this year.
The last time Public Policy Polling hit the Kentucky Senate race, Republican Randal Paul had squandered a huge lead. Democrat Jack Conway had pulled within 7 in the Rasmussen poll back at the end of June, and he was even tied in the PPP poll of Registered Voters.
PPP (this time for Daily Kos) is polling Likely Voters now, and it’s not good news for Conway.
With Christine O’Donnell’s victory tonight in the Republican Senate Primary, my last Senate projection’s hedge of a 50% Republican victory chance is obsolete. O’Donnell is the winner, and now I’m looking at how that number would look with the straight up matchup of her against Democrat Chris Coons.
Quinnipiac has released this month’s poll for the Connecticut Senate race. Now we turn to those who say Rasmussen’s “House Effect” is repsonsible for Democrat Richard Blumenthal’s surprisingly weak leads against Republican Linda McMahon, because Quinnipiac has the race even closer than Rasmussen does.
Fox News has a set of polls out, some of which I will be posting on in depth as appropriate. What’s interesting about them though is they are apparently conducted through a new service by Scott Rasmussen of Rasmussen Reports. The service is called Pulse Opinion Research, and reportedly anyone can do a poll through it for $600.
This could change the polling landscape.
With the House update done, it’s time to move on to the new Senate update. Last week the pace of Republican gains slowed, as the GOP gained less than a full seat in my projection since I started running it. But the chance of a Republican majority has climbed every time, so let’s see if that still holds.
Now available: a version of the Swingometer for the iPhone® and iPod Touch®! Get it in the App Store℠!
House of Representatives Swingometer
Electoral College Swingometer