Posted on 10/23/2010 8:31:40 AM PDT by Brices Crossroads
Jim Geraghty got an e-mail from a reader who crunched the numbers... in the early-voting states, and the news looks pretty encouraging... Comparing the numbers from 2008, Geraghtys reader says that in seven key states, Republicans have improved their performance in turnout for early voting thus far by almost 15 points, a stunning turnaround and one that may put the likely-voter models used by even the best pollsters in doubt. The smallest gain is 4.2% in West Virginia, where even Democrats are breaking towards the GOP, to a 27-point gain in Florida, where over half of all early voting has been Republicans.
In Colorado, where voters have to choose a governor and a Senator, Republican ballots increased over 7 points from the 2008 early voting.....
In Nevada, the Clark County vote (which includes Las Vegas) shows an increase of 7 points for Republicans, and in Washoe County (Reno, Tahoe) the GOP have increased over 11 points in early-voting turnout....
Pollsters expected a more robust Republican turnout in 2010 than in 2008, of course. In analyzing most of their likely-voter models in the last few weeks, its clear that they didnt expect a +15 in the gap from 2008. Many have been arguing that Democrats had begun closing that gap in the last two weeks. Assuming that holds up not just in early voting but also in Election Day turnout (as well as absentee ballots, which get counted on Election Day and not as early voting), then most of the likely-voter models will have significantly underestimated the Republican wave, even apart from the right turn of the independents. Even if GOP turnout on Election Day turns out to be half as significant, the likely voter models will still have left the polls short of the amplitude of the wave thats coming.
(Excerpt) Read more at hotair.com ...
What is important to take away from these early voting numbers is the following observation concerning the current likely voter polls:
"Assuming that holds up not just in early voting but also in Election Day turnout (as well as absentee ballots, which get counted on Election Day and not as early voting), then most of the likely-voter models will have significantly underestimated the Republican wave, even apart from the right turn of the independents."
In other words the GOP wave is only measuring intensity of turnout among GOP voters, NOT independents who are numerous in Delaware and who, along with all independents nationally have made a rightward turn since 2008.
Rush had the following colloquy with a listener who was obsessing about the polls in PA, in which he makes an observation about who is doing these polls and what their objective is (hint: to demoralize Republican voters and to energize Democrat voters):
"CALLER: I'm concerned; I gotta admit. I know better than to listen to the "lamestream media," but I'm doing it, and I'm starting to freak out because I'm starting to see... (sigh) From the Real Clear average polling I'm starting to see things tighten up, and I mean tight, tight, tight and I'm wondering, did the...?
RUSH: Like where?
CALLER: Pennsylvania for one.
RUSH: Don't believe that poll. You talking about Toomey and Sestak?
CALLER: Well, that's just one.
RUSH: Don't believe that poll. That's a 10-point race. That's a t10- or 12- point race for Toomey. Don't believe that poll.
CALLER: Okay. I'm seeing... I'm just looking at averages of lots of polls.
RUSH: Don't believe the poll. These polls are designed for two things, fundraising for the Democrats and they're trying to spur turnout. Remember, who's doing the polls? These polls are not reflecting public opinion, they're trying to shape it.
CALLER: Well, that's what I believe, too, and I needed to hear you say it and I think America needed to hear you say it..."
Don't Believe the State-Run Polls
http://www.rushlimbaugh.com/home/daily/site_102210/content/01125108.guest.html
The early voting numbers suggest that the strategy is failing miserably and may even be backfiring. As Rush observed on Wednesday,October 20 in speaking with a Delaware voter:
"RUSH: Well, you have to think that the regime is concerned about O'Donnell. There's no reason to send Obama and Joe Bite Me in there if it's a safe seat for the Democrats. I mean why waste their time. I mean their stated reason was, oh, it's a golden opportunity, we're gonna send Obama and Bite Me in there because Christine O'Donnell's so stupid that it will allow us to show the rest of the country how stupid all conservative Republicans are. If that was the purpose, they didn't succeed in it. They went in there and their campaign appearances were all about, "Don't take anything for granted here." They obviously don't think that Coons has this in the bag, otherwise there's no reason to go.
Now, if the experts are saying -- there are a lot of them, by the way, from Dick Morris to Charlie Cook to Rasmussen, there's a bunch of them -- if they are saying that there are at least 100 seats in play, then pretty much you have to conclude that there aren't any safe seats, if a hundred seats are in play? I mean there are some safe seats, but a hundred, this is huge. If 70 to 100 seats change hands, that is a Teanami. That is Obama's Katrina. What the Tea Party's doing is making every election about ideology, which is what has to happen. The sooner that happens, the sooner elections are about ideology, the sooner we're finished with these people as having anything like a chance to become a ruling majority, or minority. If the education is conservatism versus Democrat, liberal, socialist, Marxist, they'll never win, outside of a couple enclaves here in the country."
November 2nd: Obama's Katrina
http://www.rushlimbaugh.com/home/daily/site_102010/content/01125113.guest.html
Extremely important point.
The likely voter models are way way off.
Which means that the polls are way way off.
The Republicans are going to do significantly better than the polls indicate.
Nothing early in Delaware. The average is plus 15. That means add 7.5 to the Republican and subtract 7.5 from the Democrat. What that would mean in Delaware is that if a pollster is talking to way too many Democrats, and there’s a crazy 19 point lead for Coons, subtract 7.5 from Coons and add 7.5 to Christine. A 19 point lead becomes a 4 point lead.
>> in the early-voting states, and the news looks pretty encouraging...
I get emails from MoveOn.org every few days and they were in a panic last week. But this week, they’ve accepted their fate and talk about the election as a done deal as they speak of “dark days ahead” and “they can’t take away our sense of humor”. The emails where they screeched hysterically about foreign money “stealing” the election (as the once-and-future-Nazi Soros continues to fund Dems through progressive organizations) were funny, but these new ones where they sound like whipped puppies calling a suicide hotline make me laugh out loud.
“The likely voter models are way way off.”
Correct. And even analyses like Feraghty’s and Morrissye’ are way off becasue they don’t take into account the Independents who have swung hard right and are voting early too. In other words, especially in a state with a lot of Indies (like Delaware) there are going to be some surprised (and depressed)people on Wednesday morning.
Libs drinking hero own kool aid. LOL. Bad hangover coming. They can’t say we didn’t warn them.
Senator O’Donnell. Get used to the sound of it.
We have to win in all place. Remember when Tom Foley lost when he was Speaker of the House in 1994, that is that kinda wins we need.
I’m in Texas 17 where we have a real chance to unseat Chet Edwards after 20 years.We have a gubernatorial race this year too. Local elections administrator states that in my county there is a 45% increase in early voters over the same period in 2006 when we last also had a gubernatorial election on the ballot. I don’t know if the big turnout is for Rick Perry or a determination to oust Chet Edwards. I voted yesterday midday and had about a 15 minute wait in line to cast my ballot.
We need to take 10 seats away from the Rats to win control of the Senate. I love our chances in AR, CO, IN, MO, ND, WV, WI and PA. And we could win in IL and NV. That's 10. CA and WA are possible, although DE and CT look like real long shots.
OHIO CAMPAIGN STOP "Barack, you think these cheering suckers will vote Democrat, again, after all your screwups."
" My screwups? You the one who went globe-trotting to Spain. That dumba** trip killed my agenda for good."
“DE and CT look like real long shots.”
DE does not look like a long shot if you look at it closely, as I have done:
Why Christine O’Donnell Can Win this November (Wave Election plus Open Seat)
http://www.freerepublic.com/focus/f-news/2586363/posts
How Christine O’Donnell will Win (Has everyone forgotten Massachusetts on January 19?
http://www.freerepublic.com/focus/f-news/2608041/posts
O’Donnell versus Coons (Facebook and Fundraising versus Public Polls and the NRSC)
http://www.freerepublic.com/focus/f-news/2609964/posts
It doesn't work that way. Early Voting magnifies the swing, we think, with the energized Dems of 2008 giving way to energized Americans in 2010, but the swing is (probably) smaller than what you see in early voting. The EV swing is an indicator, but it's not linear (you can't just change the 2008 outcome by 15% and get the truth). The truth is that we don't know how big the final election turnout swing will be this year. It will be HUGE (barring an October Surprise), but we don't know how big. Donate. Vote. Get out the vote. Get trained now as a poll watcher and stop the Dem fraud machine. We can't take this election for granted, although it is looking quite nice.
Senator O’Donnell - puts a smile on my face.
I looked at the “bad” polls from a couple weeks ago -
Here are the voter registration numbers currently
In Delaware, its 47% Dems, 29% Reps, 24% indy.
Survey USA talked to 45% Dems, 32% Republicans, 21% Indys.
They talked to way too few Republicans and too many Dems and Indys.
Monmouth talked to more Indys (self identified?) than Democrats and more Democrats than Republicans.
Roughly 31.2% Republicans.
Magellan had it at 47% Dems, 38% Reps, 15% Indy.
I would argue that something like 40% Dem, 37% Rep, the rest indy would make a more accurate poll. Either way, for Delaware, a 3 point advantage for Democrats. I don’t know about the indys.
Right, but I’ve been thinking the likely voter models are screwed up for months now. Rarely do you see what they’re using for a likely voter screen.
In Maine, often they required that you voted in 2008 to be considered a likely voter now. That’s clearly wrong.
When we do have real data, like with this early voting, we’re seeing a huge swing to the Republicans.
Yes, there’s no 1:1 correspondence between early voting and election day voting, but these early voting numbers certainly do indicate that pollsters should take another close look at their likely voter screen and make sure that they talk to more Republicans.
I hope you're right, but I am using the term "long shot" in the pure, gambling sense. Over at Intrade, where people put up real money, a $7 bet on a Republican victory in Delaware will settle up at $100 bucks if she wins, a profit of $93, or odds of more than 13 to 1. Of course, one explanation of the long odds is that folks are betting based on information from the pollsters and the pollsters may be off, as you suggested, by an order of magnitude or two.
this has to be better than 94.
we won 52 seats then, it will be more this time.
http://www.realclearpolitics.com/articles/2010/10/22/whos_going_to_vote_this_year_107686.html
P.S. The article does a great job of explaining the technology of polling and how pollsters can influence the results of their poll.
You know how little faith I have placed in the polls. IN the last several weeks, among new Facebook fans, I have noticed O’Donnell has been maintaining roughly a 2.5 -1 lead, sometimes as high as 2.6-1, but never reaching 3-1.
As I get closer to election, I am focusing on this a little closer to try to gauge the intensity (and therefore the turnout). Since October 15 (when COD had 24,964 fans) she has added 1561 (as of October 23) and now has 26,524. Coons, on the other hand, had 9247 on October 15 and has added 497 as of October 23.
For the crucial home stretch, Christine O’Donnell is out pacing Coons by a factor of almost 3.2 to 1. THIS IS THE FIRST TIME SHE HAS CROSSED THE 3-1 THRESHOLD WITH COONS. Indeed, it is the first time I have seen any of the candidates I have observed (Miller, O’Donnell, Coons and Murkowski) outstrip the others by such a pace.
This indicates to me several things:
1)The polls are not reflecting the true turnout model, and are materially understating her support (a fact confirmed by the early voting analyses);
2) The attacks on her are backfiring and are hurting Coons;
3) The visits from Biden and Obama to Delaware are driving up the intensity of her support while doing nothing for Coons and perhaps even damaging him; and
4) O’Donnell is peaking at exactly the right time.
So many conservatives sat the last election out because of McLame that just having them come back to the polls in the midterm is probably enough for a Republican win. Add to the fact that many Independents are blowing back to the conservative side and even some disillusioned Dims are either switching sides for this one or will sit it out, should result in a massive Republican win.
I remember the “polls” in 1984, at this time, according to MsM - and that’s all there was) the vote was going to be a 2 to 1 blowout f or the dems.
They got half of it right, It WAs a 2-1 blowout - for Reagan
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.