Rasmussen’s 2012 Polling Has Had A Republican Bias All Year

As in 2010, Scott Rasmussen's polling in 2012 has shown a distinct bias in favor of Republicans.

To add some more fuel to the “Skewed Polls” debate we have this observation about Rasmussen’s polling so far in the 2012 election cycle:

Enough presidential polling data is now available to analyze Rasmussen’s data. Here is the methodology. The database contains 119 Rasmussen state polls from Jan. 1, 2012 until yesterday.. For each poll, a check was made to see if at least one poll from a different nonpartisan pollster was in the data base within a week either way from the Rasmussen poll. For example, for Rasmussen’s poll of North Carolina on Oct. 2, a check was made for any other polls of North Carolina whose midpoint was between Sept. 25 and Oct 9. In this case, polls from PPP, ARG, SurveyUSA, and High Point University were found. For 82 polls, comparison polls within a week were found. For the other 37 Rasmussen polls, no other nonpartisan pollster surveyed the state within a week of Rasmussen’s poll, so those polls were not used in this analysis.

For each remaining poll, the Obama – Romney score was computed. The arithmetic mean of the other polls’ scores was then subtracted from the Rasmussen Obama – Romney value. Ideally, the result should be zero, but statistically that is very unlikely. A positive result means Rasmussen is overestimating Obama’s standing and a negative one means he is underestimating it. For example, for the North Carolina poll cited above Rasmussen said Obama was 4 points behind but the average of the other pollsters put Obama 0.2 behind, so Rasmussen gets a bias score of -3.8 here. Averaging all 82 polls, Rasmussen’s mean bias is -1.91 points, that is, Rasmussen appears to be making Obama look almost 2 points worse than the other pollsters.

(…)

Note that this does not necessarily mean Rasmussen is wrong and the others are right. It could be that Rasmussen is right and the others are painting too rosy a picture for Obama. There is no way to know now.

On Election Night 2010, Nate Silver observed that, based on the election results that had come in, it appeared that Rasmussen’s 2010 polling was biased against Democrats by three to four percentage points. In a follow-up post, noted this about Rasmussen’s performance that year:

The 105 polls released in Senate and gubernatorial races by Rasmussen Reports and its subsidiary, Pulse Opinion Research, missed the final margin between the candidates by 5.8 points, a considerably higher figure than that achieved by most other pollsters. Some 13 of its polls missed by 10 or more points, including one in the Hawaii Senate race that missed the final margin between the candidates by 40 points, the largest error ever recorded in a general election in FiveThirtyEight’s database, which includes all polls conducted since 1998.

Moreover, Rasmussen’s polls were quite biased, overestimating the standing of the Republican candidate by almost 4 points on average. In just 12 cases, Rasmussen’s polls overestimated the margin for the Democrat by 3 or more points. But it did so for the Republican candidate in 55 cases — that is, in more than half of the polls that it issued.

If one focused solely on the final poll issued by Rasmussen Reports or Pulse Opinion Research in each state — rather than including all polls within the three-week interval — it would not have made much difference. Their average error would be 5.7 points rather than 5.8, and their average bias 3.8 points rather than 3.9.

Nor did it make much difference whether the polls were branded as Rasmussen Reports surveys, or instead, were commissioned for Fox News by its subsidiary Pulse Opinion Research. (Both sets of surveys used an essentially identical methodology.) Polls branded as Rasmussen Reports missed by an average of 5.9 points and had a 3.9 point bias. The polls it commissioned on behalf of Fox News had a 5.1 point error, and a 3.6 point bias.

Rasmussen’s polls have come under heavy criticism throughout this election cycle, including from FiveThirtyEight. We have critiqued the firm for its cavalier attitude toward polling convention. Rasmussen, for instance, generally conducts all of its interviews during a single, 4-hour window; speaks with the first person it reaches on the phone rather than using a random selection process; does not call cellphones; does not call back respondents whom it misses initially; and uses a computer script rather than live interviewers to conduct its surveys. These are cost-saving measures which contribute to very low response rates and may lead to biased samples.

Rasmussen also weights their surveys based on preordained assumptions about the party identification of voters in each state, a relatively unusual practice that many polling firms consider dubious since party identification (unlike characteristics like age and gender) is often quite fluid.

Rasmussen’s polls — after a poor debut in 2000 in which they picked the wrong winner in 7 key states in that year’s Presidential race — nevertheless had performed quite strongly in in 2004 and 2006. And they were about average in 2008. But their polls were poor this year.

The question, of course, is why Rasmussen’s polls have shown this bias for two consecutive election cycles now, although it’s worth noting that they seem to be slightly less biased against Democrats than they were in 2010. The natural conclusion many on the left will draw is that Scott Rasmussen is deliberately cooking his books to favor the GOP but I think the answer is likely more mundane than that. Yes, the fact that Rasmussen weights by Party ID, something that almost no other pollster does, is going to have an influence on the numbers, but  Nate Silver noted on his blog before his big move to The New York Times that Rasmussen’s polls seem to show the same bias even when they’re not using the Likely Voter model.  So, what’s going on?

As James Joyner observed in November 2010, there seems to be another factor at play:

 [A]nother — not mutually exclusive — explanation exists.  It turns out that Rasmussen’s results, which are robocalled, don’t include cell phones.   Neither do the Fox numbers.  That alone could account for a 3-4 point Republican skew.

The cell phone issue is an important one. As I noted last month, there is clear evidence that polls that don’t include cell phones tend to understate the President’s support and, indeed, we saw that very phenomenon in a series of NBC News polls that were released back in May. More and more, it’s becoming quite apparent that polls that exclude cell phones, like Rasmussen’s do, aren’t getting a clear picture of the electorate. I’m not sure that the cell phone issue alone would account for Rasmussen’s bias toward the GOP, though. Public Policy Polling is also a robocalling firm and they most assuredly don’t display a bias toward the GOP, indeed if they do have a bias it’s toward the Democratic candidate more often than not. Of course, PPP doesn’t weight for Party ID in the manner that Rasmussen does so that likely explains why they are usually so dissimilar.

Scott Rasmussen runs his polls based on the theory that he, and only he, knows what the makeup of the electorate is going to be, and most especially exactly how many self-identified “Republicans,” “Democrats,” and “Independents” will show up on Election Day. As we learned during September’s “Poll Denialist” kerfuffle, though, his polling model is based on assumptions that most other pollsters don’t accept. There is weighting that goes on in polling, and some of it is absolutely necessary. If a particular set of raw poll numbers seem to have drawn, at random, an oversample of a particular demographic group such as race, gender, or age, then what most pollsters typically do is adjust those numbers by using the data provided by the United States Census Bureau. If they are dealing with a likely voter poll, they will also likely use Exit Poll information from previous elections as a guide as well. For example, Exit Polling seems to consistently show that women vote in a slightly larger amount then men, and that older people vote in a large percentage than younger people. However, the data for Party ID isn’t nearly as well-established. People tend to change their Party self-identification depending on the political mood of the country  This is especially true of that large group of people who call themselves “Independents,” many of whom are actually Republicans or Democrats who, for whatever reason choose to use the “Independent” label. Often, those people will move back toward their party as their support for a candidate becomes more solid. So, weighting your poll on something that seems to fluctuate as much as Party ID doesn’t seem to make a whole lot of sense, which is why most pollsters don’t do it.

Perhaps it will turn out that Rasmussen’s model is correct this year, we won’t know that until Election Day. At the moment, though, it’s simply undeniable that this model is producing results which are giving a boost to Republican numbers in a manner that none of the other pollsters are.

H/T: Taegan Goddard

FILED UNDER: 2012 Election, Public Opinion Polls, US Politics, , , , , , , , , , , ,
Doug Mataconis
About Doug Mataconis
Doug Mataconis held a B.A. in Political Science from Rutgers University and J.D. from George Mason University School of Law. He joined the staff of OTB in May 2010 and contributed a staggering 16,483 posts before his retirement in January 2020. He passed far too young in July 2021.

Comments

  1. PJ says:

    One major issue with Rasmussen, it uses a likely voter screen more than a year before an election, and then it is using the screen for any kind of questions asked.

    Rasmussen is push polling Republican views.

  2. Geek, Esq. says:

    Yet Rasmussen models get added instantaneously to the RCP averages, but PPP polls only get added when they’re favorable to Romney.

    If only there were a site that had RCP’s page layout and ease of use, without its rightwing slant.

  3. anjin-san says:

    Rasmussen is a Republican PR firm that uses polling as a marketing gimmick, nothing more.

  4. PPP’s state-level polling is added to the RCP averages on a regular basis. The only thing I’ve seen not be included is the separate polling they do for the DailyKos and the SEIU

  5. Let's Be Free says:

    Left-leaning media sponsor polls that by their sheer number dominate the averages. Exposing that Rasmussen polls to the right of their average is Geraldo Rivera journalism. There is nothing behind that wall.

  6. Geek, Esq. says:

    @Doug Mataconis:

    They haven’t included the last two PPP polls for your home state, Virginia. Both of which showed Obama ahead (+2 last week, +5 today).

    http://thehill.com/blogs/ballot-box/polls/264007-poll-obama-leads-in-virginia

    But today’s Rasmussen Virginia poll (showing Romney up) of course is already in their averages.

    Surely a coincidence that a site founded with an explicitly pro-conservative Christian Republican agenda would make such an oversight two weeks in a row.

  7. Anderson says:

    We really need somebody to produce an EC map that uses only polls that include cellphones.

  8. Jr says:

    I love how PPP puts out there VA poll that favors Obama and then a few hours later Rasmussen puts out his to favor Romney, they are a narrative pollster whom shouldn’t be taking seriously.

  9. Geek, Esq. says:

    @Jr:

    Are you suggesting that weekly/monthly polls asking “should Obamacare be repealed” have an agenda?

  10. John425 says:

    “Note that this does not necessarily mean Rasmussen is wrong and the others are right. It could be that Rasmussen is right and the others are painting too rosy a picture for Obama. There is no way to know now.”

    Actually, Doug-there is a way to know. The so-called “mainstream polls” consistently weight Democrat participants more so than Republicans. For example they’ll use 39% Democrats, 32% Republicans and 11% independents.

    Gee, lawyering must be hard.

  11. Geek, Esq. says:

    Also, remember when the guy from Suffolk said that Obama had zero possibility of winning Virginia, based on his firm’s polls that showed Obama ahead but with only 46% of the vote?

    That guy belongs in the Pollster Hall of Shame.

  12. Geek, Esq. says:

    @John425:

    Math is even harder. 39+32+11=92% of voters.

    Also, those firms don’t weight their polls. Those just happened to be the numbers they came up with when asking a random sample of voters.

  13. Jr says:

    @John425:

    Party ID is fluid, stupid.

    And besides the reason party id for Democrats looks higher is due to the fact many Republicans call themselves independents now……much like the Tea Party in 2010.

  14. Geek, Esq. says:

    @Geek, Esq.:

    Oops. Math really is hard. 39+32+11 is only 82% of voters.

  15. Wayne says:

    Accuracy ranking of pollsters

    http://www.breitbart.com/Big-Government/2012/09/27/Flashback-rasmussen-most-accurate-pollster-in-2008

    Mason-Dixon & Rasmussen have been two of the top ten pollsters over the past twelve years
    http://www.pollster.com/blogs/pollsters/

    Yes one can cherry pick an election year or group of races to claim a pollster is accurate or not. However, it is their record overall that should be look at.

    The way many put so much faith in any poll especially when they agree with it amazes me. Polling is an art and is often wrong. Also, public polls are often manipulated for an agenda. It is only when an election get close that many of them “adjust” their sampling in an “attempt” to be right.

  16. David M says:

    @Wayne:

    Also, public polls are often manipulated for an agenda. It is only when an election get close that many of them “adjust” their sampling in an “attempt” to be right.

    The complaints about Rasmussen are his results in 2010 and numbers so far in 2012, and setting an agenda appears to be what he is doing.

    One other note about cell phones, a decent number of them have the “wrong” area code as far as polling is concerned. Plenty of people now keep the cell numbers after they move, so their cell number could be from another state.

  17. The Q says:

    Does anybody know the actual nuimbers from the 04 and 08 elections as to the registered voters and how they in reality turned out?

    What I mean to ask is, if Dems are 39 and Repubs are 32% of the registered voters prior to the election, what are the post election results?

    Did Dems comprise 39% of the voters in 04 and 08, or were they more (08) or less (04) represented.

    Similarly with the Repubs, do they turn out more or less than their 32% registered numbers.

    If we know that on average the Dems turn out 5% less (or more) than their registered numbers indicate, we can assess these polls much more critically.

    For example, if Obama leads in a poll that is weighted 39/32/11, but dems on election day only turn out 35%, well then, those polls inaccurately reflect the Repub candidate’s support.

    Any data?

  18. David M says:

    The Q:

    Does anybody know the actual nuimbers from the 04 and 08 elections as to the registered voters and how they in reality turned out?

    It varies greatly by state. CNN has them for 2008 and 2004

  19. Geek, Esq. says:

    Also note that RCP includes the Iowa poll from PPP showing Romney ahead from ten days ago, but neither of its subsequent polls showing Obama ahead there.

    RCP makes a good enough effort at pretending to be unbiased to snooker a lot of people.

  20. Wayne says:

    @David M
    And those complaints have been level at all the other polling firms as well. Many of the polls have used sampling margin of +Democrats that is greater than what it was in 2008. Many on this site made fun of those who pointed that out or many other sampling issues. Some other issues such as certain pollsters past history of suddenly adjusting their sampling % and methods in when the election gets close.

    Most polling firms have an off election year. That why it is better to look over multiple election years to determine a polling firm reliability. They should always be taken with a bucket of salt.

    It just funny how many liberals think when the left complain about a polling firm it is great, legit, and an intellectual conversation. However when the right does it, it is whining, silly, etc.

  21. Barry says:

    @John425: “Actually, Doug-there is a way to know. The so-called “mainstream polls” consistently weight Democrat participants more so than Republicans. For example they’ll use 39% Democrats, 32% Republicans and 11% independents. ”

    No, they don’t. They might have more self-identified Democratic (can you spell?) participants than you’d like.

  22. Tsar Nicholas says:

    My head actually exploded while I was reading this post and the items to which it makes reference. No, seriously, I’m crippled. I might never recover.

    Where do you even start with this?

    Ahem.

    The notion that if Rasmussen says X and if the average of all the other polls says Y means ipso facto that Rasmussen is + or – Z “biased” is so numbingly cognitively dissonant it could actually substitute for anesthesia. All it means is that Rasmussen is using a different weighting methodology. Either he’s right or he’s wrong. We’ll know after Election Day. But what takes it from the absurd literally to the comatose is that most if not all of these other polls to which references are made laughably are rigging their own samples to boost Obama. That’s the most surreal part about all this nonsense.

    But before we get to what the liberal media polls are doing let’s take a step back. Using a likely voter model weighted by party ID not only is not suspect as a methodology it’s common sense. It’s not as if there are 20 possible categories of voters. People either are Democrats, or Republicans, or Independents. People either are more motivated or less motivated to vote. People either are more likely or less likely to vote. In large part that breaks down along partisan party lines. This isn’t rocket science. It’s not a closely guarded state secret. Geez.

    Getting back to the Rasmussen vs. other polls dog & pony show, again, what numbs the skull is the fact that so many of these media polls blatanly and brazenly are stacking the decks in favor of Obama. And it’s not even about partisan party “skew.” It’s a lot simpler than that. It’s a lot more basic. As basic as black and white, male and female. Literally.

    Check out the internals of some of these other polls. PPP and Gravis Marketing stand out. Marist too. Compare the racial breakdowns to historical exit data info. Let’s say for example, just humor me, white people were 83% and black people were 11% of the OH electorate in ’08. Of course that was a “transformative” wave election cycle for Obama, so it goes without saying that this time around relative black turnout won’t be higher. Can we at least agree on that? And let’s say, again humor me, that if a poll, oh, “hypothetically” PPP, has included 80% white people and 13% black people in a survey of OH, well, then guess what, Cochise, that means they oversampled blacks and undersampled whites. For obvious reasons. The same analysis would hold true for women and men. If a polling service is including significantly more women in a survey of OH than the actual number of women who voted last time around do we really need a flow chart and a puppet show to explain to us why they’d be doing that?

    I’ll go ahead and let people in on final dirty secret of the recent media polling. In recent weeks it’s become obvious even to liberals in the media that Romney is winning with Independents. Every poll says so. So what have the media polling services done? Yeah, you guessed it, they’ve started to undersample Independents. Right now SurveyUSA is the most noteworthy culprit. Check out the internals of their recent polls. Notice that they clearly state Romney is winning the Independent vote. Usually ranging from 7-9 points. But then juxtapose the percentages of Independents in their surveys with the percentages of Independents who actually voted in ’08 and in ’04. Connect….the….dots.

    Look, in the end none of this will matter. If Obama wins this election, even if by exactly one dead felon in Cleveland, nobody is going to go back and say “Hey, wait a minute, the likes of Time, NBC/Marist, CBS/NYT, PPP, SurveyUSA, etc., all had him up 3 or more points in OH a week or so before the election, what gives?” The left will be too busy preening and beating their chests and the right will be too busy engaging in a circular firing squad. And if Romney wins the election, even if by two full percentage points in OH, the right will move on to complaining about his presidency and the left instantly will go into full litigation and recount mode and will spend the ensuing four years claiming that Diebold did it, or Halliburton, or Rove, or Bush, or whatever. C’est la vie.

  23. Barry says:

    BTW, IIRC Rasmussen will drop their bias in the past few weeks (we should see a shift soon), just to make sure that their end figure does end up close.

  24. David M says:

    @Wayne:

    And those complaints have been level at all the other polling firms as well. Many of the polls have used sampling margin of +Democrats that is greater than what it was in 2008. Many on this site made fun of those who pointed that out or many other sampling issues. Some other issues such as certain pollsters past history of suddenly adjusting their sampling % and methods in when the election gets close.

    How many times does it have to be pointed out the Rasmussen is pretty much the only pollster who alters the reported Party ID to match what they think the electorate will be. The polling firms you’re complaining about are just reporting what the respondents are saying, they are not adjusting the results to meet a predetermined partisan breakdown.

  25. Jr says:

    @Barry:

    Yup, they did the same thing in 2008.

    If they didn’t. they would go the way of Zogby and no one would take them seriously for the 2014 mid terms and the 2016 elections.

  26. mattb says:

    Rasmussen? Bah!
    PPP? Hacks!

    The only two pollings firms that can be trusted to absolutely positively get it right are Paladin//CFP (with their CTV or Certain to Vote measure) and Tristero/Rosicrucian polls (featuring their revolutionary CTVAWNDOBIBE — Certain to vote and will not die or be imprisoned before election — model).

    [for the ironically/fact challenged, background can be found from here on down]

  27. MBunge says:

    @Tsar Nicholas: Pretty much everything right wingers are saying about polls this time around confirms the old adage “A little knowledge is a dangerous thing”. They’ve looked into polling techniques JUST enough to find an argument to dismiss any polls they don’t like.

    Mike

  28. jan says:

    @Tsar Nicholas:

    You are one of the most interesting and realistic posters on this site, saying it like it is, willing to show the petticoats of both party’s shortcomings, rather than just harp on the misfortunes of those not clinging to Obama’s reelection.

    Whatever direction Rasmussen leans, he at least doesn’t completely fall over from the weight of bias, like some pollsters, such as PPP do — at least in the early stages of polling. PPP usually gets a more objective hat on during the closing days of any given election period.

    However, Rasmussen is a rather steady pollster, using a methodology that has been finessed since the mid-1990’s. He is usually one of the first pollsters to use the more reliable LV screen, which also seems to weed out some of the blue-sky democratic stats seen in an RV or adult’s only screen. Gallup, for instance, didn’t change his screen to LV until October, which also saw changes in his polling, going more in the direction of the R’s.

    As far as Rasmussen’s polling aptitude, he came within 1% point in the 2004 Bush/Kerry race. In 2008 he predicted a 52/46 outcome, as compared to the final margins being 53/46. People who follow polling also tend to say that Rasmussen pins national polling better than he does the one’s done in the states.

    A final note on the vagaries of polling: with the advent of early polling in a majority of states it remains to be seen how accurate any of these polls are going to fare. For instance, it is well known that dems comprise a greater portion of early voters. When a pollster calls people, screening for LVs, they will accept all people who have already voted, which may be giving these polls a sampling edge to Obama that will not be realized on the final day, November 6th, which is usually denoted by more R’s voting. Consequently, once 2012 is over, there may have to be some serious reevaluations by pollsters, taking into account the evolving way voting is going.

  29. Septimius says:

    Exit Polling seems to consistently show that women vote in a slightly larger amount then men, and that older people vote in a large percentage than younger people.

    Exit polling also seems to consistently show Democrats performing better than the actual election results. Why? Exit polling consistently oversamples Democrats.

  30. jan says:

    Here is a Romney endorsement in words, from the NY Post, along with their front cover picture, that says it all.

    A potential problem in Ohio, regarding Ohio law and how it may slow up the final vote count:

    Ohio is a key state in this election, but according to USAToday a majority of absentee ballots (around 800,000) have not been returned yet. If someone requests an absentee ballot and then votes in person instead, they must use a provisional ballot so election officials can verify they didn’t also cast an absentee ballot. By Ohio law provisional ballots can’t be counted until at least Nov. 17. So according to the article, there could be a high number of provisional ballots cast because of this reason (the usual reasons involve problems with addresses and names). In 2008 there were around 200,000 provisionals and about 20% were ruled invalid. So there will probably be more provisionals this time around. To complicate matters, absentee ballots postmarked by Nov. 5 have 10 days to arrive at the election office. I also believe that if you do not produce acceptable ID when voting (doesn’t have to be photo ID), you must produce it within a certain time frame or your vote doesn’t count. So if it all comes down to Ohio, we may have to wait awhile to know who won. In fact Ohio doesn’t have to come up with final official results until Nov. 27. I think Ohio may contribute to a national mental breakdown as people say “I can’t take it anymore.”

  31. David M says:

    @Septimius:

    Exit polling also seems to consistently show Democrats performing better than the actual election results. Why? Exit polling consistently oversamples Democrats.

    You clearly don’t understand what exit polling is, or how it works. The final exit polls that Doug was referring to are corrected to match the actual results in the race, so errors from over-sampling are removed.

  32. anjin-san says:

    One thing we can count on is increasing shrillness from Jan, Jenos, et al if momentum continues to swing back towards Obama.

    In other news – GO GIANTS!

  33. sam says:

    @Wayne:

    http://www.breitbart.com

    I stopped reading right there.

  34. jan says:

    @anjin-san:

    “One thing we can count on is increasing shrillness from Jan, Jenos, et al if momentum continues to swing back towards Obama.”

    Shrillness, or is it simply denial coming from your side?

    BTW, Obama is rooting for the Tigers. You know that thing about political calculations, and Obama needs MI, as the last poll was 47/47. Like you, though, I am backing the Giants.

    Other tidbits to deny:

    1) Here’s an interesting trend. Polls with larger samples show Romney in the lead, versus the smaller samples where Obama is ahead.

    2) WAPO poll showing 50/47 Romney, that can be added to the RCP list above.

    3) Also worth noting: In Defiance, Ohio (home to GM plant), GOP leads early vote by 13%…which is now a larger lead than 2008. Maybe, voters are finally getting beyond Romney’s ear;ier GM op-ed by-line, created by the NYT, which lays out a proposal very different than the one coughed up by Obama in the last debate.

  35. Cycloptichorn says:

    I think most here are really misunderstanding the purpose of Rasmussen polls.

    They aren’t meant to accurately report public opinion, they are to CREATE public opinion. Ras floods the zone with more polls, earlier in the day, than any other pollster, period. They only need a very slight, consistent bias on each poll to be able to consistently shape the narrative – people like a winner, and when they see that someone is ‘winning’ in polls all over the country, they are more likely to become pumped/depressed.

    That’s why Ras to this day still polls on the health care bill; that’s why they came up with that BS ‘Presidential approval index’ which measures strong approval v. strong disapproval, a metric only designed to make Obama look bad.

    If nothing else, the fact that they don’t call cell phones, and THEN run the results through their preconceived partisan splits, which are always +2 GOP or so, is proof enough that they are intentionally skewing their polls to produce a desired result.

  36. C. Clavin says:

    Jan calls Tsar…

    “…one of the most interesting and realistic posters on this site..”

    Seriously…has anything funnier ever been posted on the OTB site????
    Jan wins the thread.

  37. David M says:

    @jan:

    Maybe, voters are finally getting beyond Romney’s ear;ier GM op-ed by-line, created by the NYT, which lays out a proposal very different than the one coughed up by Obama in the last debate.

    There’s no doubt that Romney will continue to try and hide his positions on the rescue of GM & Chrysler. The record is clear: Romney did not support rescuing the auto makers and ensuring they were not liquidated.

  38. C. Clavin says:

    The Princeton Election Consortium…as of 5:00 this evening has Obama 297 v. Romney 241
    They hit the Bush Kerry EV exactly. Exactly.
    They missed the Obama McCain EV by 1 vote.
    If you are watching National polls you are just wasting your time.
    Which explains why Jan is watching National polls.
    http://election.princeton.edu/

  39. Wayne says:

    @David M
    Most pollsters weight their polling. They may not weight purely by party but faulty weighting other categories can result in an oversampling of Dems and it could be on purpose. Therefore, you are only assuming they don’t adjust results for partisan reasons.

    It looks like ABC and CBS do weight by party in some of their polls. In addition, most polls change from register voters to likely voters when the election gets closer. The early polls are often used to mislead early on.

    Weighting polls can be used to improve polling results. That why most pollsters do weighting of some sort. Weighting by political party seem to be reasonable to me. However, like any weighting it can be abused.

    Is Rasmussen abusing the weighting technique this time? Argue on. However, he has been one of the more accurate pollsters in the past. Perfect, no but better than about everyone else has been over several elections.

  40. jan says:

    @C. Clavin:

    Currently RCP has the race, without toss-ups, as 281/257, Obama. However, that can easily change. And, to add perspective, The University of Colorado, Boulder, has another forecast pointing to a 330/208 win for Romney.

    According to their updated analysis, Romney is projected to receive 330 of the total 538 Electoral College votes. President Barack Obama is expected to receive 208 votes — down five votes from their initial prediction — and short of the 270 needed to win.
    The new forecast by political science professors Kenneth Bickers of CU-Boulder and Michael Berry of CU Denver is based on more recent economic data than their original Aug. 22 prediction. The model itself did not change.

    The state-by-state economic data used in their model have been available since 1980. When these data were applied retroactively to each election year, the model correctly classifies all presidential election winners, including the two years when independent candidates ran strongly: 1980 and 1992. It also correctly estimates the outcome in 2000, when Al Gore won the popular vote but George W. Bush won the election through the Electoral College.

  41. jan says:

    @C. Clavin:

    “If you are watching National polls you are just wasting your time.
    Which explains why Jan is watching National polls.”

    I posted a list of national polls, indicating a pattern. However, as usual you presume more than is stated by someone, as indicated by your statement above. National, state polls, their internals, dealing with favorability, etc, plus polling models like the one you posted, as well as the one from Colorade, all give a spectrum of data and analysis to be taken into consideration.

  42. David M says:

    @Wayne:

    Most pollsters weight their polling. They may not weight purely by party but faulty weighting other categories can result in an oversampling of Dems and it could be on purpose. Therefore, you are only assuming they don’t adjust results for partisan reasons.

    It looks like ABC and CBS do weight by party in some of their polls. In addition, most polls change from register voters to likely voters when the election gets closer. The early polls are often used to mislead early on.

    Most of that is wrong, and you have shown absolutely no evidence for your allegations.

    Mark Blumenthal is probably the expert , so what he says here seems relevant:

    “Claims that media polls “assume” a specific partisan or demographic composition of the electorate are mostly false. The pollsters behind most of the national media surveys, including those who conduct the CBS/New York Times/Quinnipiac, NBC/Wall Street Journal/Marist and Washington Post polls, all use the same general approach: They do not directly set the partisan or demographic composition of their likely voter samples. They first sample adults in each state, weighting the demographics of the full adult sample (for characteristics such as gender, age, race and education) to match U.S. Census estimates for the full population. They then select “likely voters” based on questions answered by the respondents, without making any further adjustments to the sample’s demographics or partisanship.”

  43. Cycloptichorn says:

    @jan:

    That analysis is a joke – there have only been 7 elections since 1980, and at least two of them were absolute blowouts. That’s a less than significant sample size to base a model upon.

    Their ‘model’ relies upon economic conditions only, and the current climate we find ourselves in is unprecedented during the time period they claim to cover. Romney would have to win every single swing state out there in order to get 330 EVs, including every one he’s behind in right now in polling. The odds of that happening are exceedingly low. For that to happen, Romney would have to win every state RCP projects him to win, plus OH, PA, MI, WI, NV, IA, and some other state that Obama is currently leading.

    To believe that model, you have to believe that all the polls – ALL of them – are simply incorrect across the board. It’s one-dimensional and not very good analysis.

  44. KansasMom says:

    Jan wants to talk about polling? BWHAAAAAA!!!!! I don’t think she realizes that she is the inspiration for Tristero/Rosencrucian and ground breaking CTVAWNDOBIBE — Certain to vote and will not die or be imprisoned before election — model.

  45. KansasMom says:

    @jan: Sweet baby jeebus, just when I think you can’t get any dumber. Obama is rooting for the Tigers for the same reason I am, they play in the same division as his favorite team. I am KC fan, he is a White Sox fan, they all play in the AL Central. This is nothing new to sports fans we do it all of the time. Go Big 12, especially my school’s in-state rival K-State!

  46. David M says:

    @jan:

    And, to add perspective, The University of Colorado, Boulder, has another forecast pointing to a 330/208 win for Romney.

    A model that has Romney over 300 electoral votes has some serious problems. Obama 243 / Romney 295 is probably the worst case for Obama, while Obama has a much easier path to 300+ EVs.

  47. bk says:

    @KansasMom:

    @jan: Sweet baby jeebus, just when I think you can’t get any dumber.

    She can,and she will.

  48. KansasMom says:

    @bk: Since she started posting here she has never failed to disappoint. I actually get a little sad when she disappears for days at a time.

  49. Ed in NJ says:

    @John425:

    Aren’t you in the least bit embarrassed to be this ill-informed while polluting this website?

  50. Ed in NJ says:

    @PJ:

    To add to your point, the “push-polling” point is even more insidious. If you look at some of Rasmussen’s poll questions, they are often framed from the right, to elicit an emotional response. They also frequently use “priming”, asking a series of negative or positive questions regarding a certain current situation first, before getting to the candidates. Obviously if the Dem is the incumbent, you get the bad scenario, and if the Republican is the incumbent, you get the good scenario. All designed to elicit a biased response.

  51. C. Clavin says:

    Jan…
    Romney has never…never…led this race.
    The odds are 90%+ that Obama wins.
    I personally believe that low-information voters like you and Tsar and Doug will win the election for Romney.
    But to date there is no factual basis for that belief.
    It’s just me being afraid of the stupid vote.

  52. anjin-san says:

    Keep talking about “the big mo” Jan – the more time you spend making noise, the less time you have to waste on things like trying to actually understand what is going on around you.

    While there is no doubt that Romney’s strong performance in the first debate altered the trajectory of the race, people who are serious about this thing know that the Romney surge ran out of steam about a week ago. That does not mean that Romney can’t win, because he still can. It just means that the race has stabilized, and the fundamental dynamic of the election still favors Obama. It was always Obama’s to lose, and he did nearly fumble it away, but the bleeding stopped a while back.

    The Romney campaign, like you, is not starting to sound a bit desperate in its attempts to maintain the illusion of continuing momentum.

  53. Jr says:

    Yeah, the Romney momentum theme is a load of nonsense.

    The truth of the matter is the race has stabilized and the President still has a slight…..but very significant lead in the key swing states.

    Mike Allen summed it up best, Obama has a bigger lead in states like Iowa, NH, and Ohio then Romney has in NC.

  54. C. Clavin says:

    Glenn Beck and Romney appearing together tonight…with Dick Cheney.
    Who believes his tack to the center???

  55. KariQ says:

    I actually like Rasmussen polls, although they do unquestionably have a GOP lean to them. Compare their swing state (not national) results from 2008 and their polling for Senator races in 2010 to the actual numbers and you will see that Rasmussen is consistently 2-4 point too favorable to the GOP candidate.

    That consistency is actually a good thing. You can get a very good feel for where the race stands, just by Rasmussen’s polls, as long as you remember to shift it about 3 points to the Democrat.

  56. jan says:

    @anjin-san:

    “While there is no doubt that Romney’s strong performance in the first debate altered the trajectory of the race, people who are serious about this thing know that the Romney surge ran out of steam about a week ago. “

    …..???? Maybe, according to the pundits at MSNBC, they may be tellling you this. But, you wouldn’t know it by the large, turn-away, bumper-to-bumper and enthused crowds attending rallies in swing states like Iowa, CO, even Ohio. National polls are inching their way towards Romney, which usually leads in front of state polls.

    Of course, one event could change the course of all this, and it could turn the other way. But, as of today, the mischaracterization of Romney, that Obama painted throughout the spring/summer all but disappeared when people saw the R candidate, unfiltered and able to represent himself, during the debates.

  57. jan says:

    Here are some twitter visuals of two recent R rallies:

    Today in Defiance, Ohio, a crowd of 12,000 in a town of 16,400.

    Red Rock, CO, a few days back. People had to be turned away, some coming from as far away as 500 miles. Rumor has it that people are being paid to attend Obama’s rallies.

    This indicates ‘Mo’ for Romney/Ryan, IMO. Whether it’s enough to counter the Obama machine, who throws out racial epithets, women wars, class warfare, binder, big bird, bayonets, bulls**t, glossy last-minute 4-year plans, and people under the political bus in order to win, remains to be seen.

  58. KariQ says:

    @jan:

    Rumor has it that people are being paid to attend Obama’s rallies.

    I’d ask where this ridiculous rumor got started, but I think I know. It’s the same place that created the rumors about his college transcripts and his birth certificate and the apology tour that never happened.

    Meanwhile, in the real world, this is what an Obama campaign rally looks like: http://www2.timesdispatch.com/news/virginia-politics/2012/oct/26/tdmain01-obama-rallies-15000-at-carillon-in-richmo-ar-2312500/

  59. David M says:

    @jan:

    Obama machine, who throws out racial epithets, women wars, class warfare, binder, big bird, bayonets, bulls**t, glossy last-minute 4-year plans, and people under the political bus

    This is just random words and incoherent blather now.

  60. anjin-san says:

    @ Jan

    Rumor has it that people are being paid to attend Obama’s rallies.

    Taking those crazy uncle chain emails to heart, I see.

    Maybe, according to the pundits at MSNBC, they may be tellling you this.

    Doubtful, it has been years since I tuned into MSNBC.

    people saw the R candidate, unfiltered and able to represent himself, during the debates.

    Romney had an excellent first debate, then lost the next two by a fairly large margin. I give him credit from pulling out of the death sprial he was in, but the fact is momentum is back with Obama. Given that the fundamentals have always favored Obama, that is a steep hill to climb.

    I do see a lot of excitement about Romney coming from low information voters. Hey, Jersey Shore is a hit too. Despite this, as an Obama supporter, I am happy with where we are today.

  61. Scott O says:

    @jan: Reality called, wants to know why you don’t visit anymore.

  62. Herb says:

    @jan:

    “Red Rock, CO, a few days back. People had to be turned away, some coming from as far away as 500 miles. Rumor has it that people are being paid to attend Obama’s rallies.”

    You confused me with that Red Rock thing….but then I realized you were talking about Red Rocks, which is a concert venue in the town of Morrison. It’s an amazing place, for sure, a great venue for any occasion….rock show, open-air movie, church service, political rally.

    I am not surprised people were driving 500 miles to see Romney there. You should have seen the place when Mumford and Sons came through.

  63. sam says:

    @KansasMom:

    Jan wants to talk about polling?

    Poll question: Jan is the Kathryn Jean Lopez of OTB

    Yes_________

    No__________

  64. john personna says:

    I”m sure many got bored out of the election sooner, but this week was it for me. Wake me November 7th, and I may have a comment.

  65. mattb says:

    One theory for the disparity between Obama’s lead in many swing states (something that even Rasmussen continues to show*) and Romney’s slim lead in many national polls is the nature of ad buys in this election.

    Very little money has been spent on national advertising and advertising in safe states. Ungodly sums are being spent in the battlegrounds (to the tune of an estimated $100K per undecided voter according to NPR).

    The result is a highly divided media ecosystem that leads to the potential for very different views of both candidates depending on where you live.

    (* – The latest Rasmussen map puts the race at a dead heat with both candidates with 257 electoral votes and the entire she-bang coming down, according to Ras, to Iowa and Ohio where the candidates are tied. Here’s Rasmussen’s map, with no toss up (other than ties), via RCP.)

  66. john personna says:

    I do like this, quoted approvingly by Tyler Cowen, as a summary of the campaign:

    The right wing has lost the election of 2012.

    The evidence for this is overwhelming, yet it is the year’s best-kept secret. Mitt Romney would not be throwing virtually all of his past positions overboard if he thought the nation were ready to endorse the full-throated conservatism he embraced to win the Republican nomination.

    …The right is going along because its partisans know Romney has no other option. This, too, is an acknowledgment of defeat, a recognition that the grand ideological experiment heralded by the rise of the tea party has gained no traction.

    Maybe that’s part of what makes me think things are settled.

  67. mattb says:

    BTW, another factor to consider in these remaining days — since we’re looking at lots of factors — is the pretty extreme differences in the two candidates ground games. Given that many swing states have extended voting, the Obama campaigns emphasis on field offices and getting people to the polls make sense.

    What we won’t know — until the election is over — is how much of an influence this will have on the outcome. That said, there’s a strong argument that this strategy helped Obama flip three states in 2008.

  68. J-Dub says:

    Ho-hum, there is a statistical tie among people who answer land-lines from a blocked number.

    Maybe because its a close election, but I don’t remember seeing so much emphasis put on polls before. There are new and conflicting poll numbers out daily, or even several times daily.

    11 days until the only poll that matters.

  69. J-Dub says:

    @mattb: The economy would get a real boost if that $100k could have just been given to the people. Instead, my donation went to some rich fuck that owns a tv station.

  70. DRE says:

    @Tsar Nicholas:

    My head actually exploded while I was reading this post and the items to which it makes reference. No, seriously, I’m crippled. I might never recover.
    Where do you even start with this?
    Ahem.
    The notion that if Rasmussen says X and if the average of all the other polls says Y means ipso facto that Rasmussen is + or – Z “biased” is so numbingly cognitively dissonant it could actually substitute for anesthesia.

    I’m afraid that any mental illness you might have in regard to this post preceded your reading of it. Here is the final paragraph of the quoted article:

    Note that this does not necessarily mean Rasmussen is wrong and the others are right. It could be that Rasmussen is right and the others are painting too rosy a picture for Obama. There is no way to know now.

    Most of the rest of the post discusses difference between poll results and actual election results. Are you questioning which one of those was correct?

  71. mantis says:

    @jan:

    Rumor has it that people are being paid to attend Obama’s rallies.

    Rumor has it Romney runs a secret cloning lab to produce attendees for his rallies.

  72. C. Clavin says:

    “…Rumor has it Romney runs a secret cloning lab to produce attendees for his rallies…”

    And sons.

  73. mantis says:

    @C. Clavin:

    And sons.

    With serial killer eyes.

  74. DRE says:

    @jan:

    For instance, it is well known that dems comprise a greater portion of early voters. When a pollster calls people, screening for LVs, they will accept all people who have already voted, which may be giving these polls a sampling edge to Obama that will not be realized on the final day, November 6th, which is usually denoted by more R’s voting.

    Dems generally comprise a larger portion of in-person early voting, but Reps are overrepresented in absentee early voting. The balance of the two elements varies greatly from state to state, but I believe that as a general rule Dems are more likely to be overrepresented in Election Day voting, and Reps in early voting, but it would take a detailed analysis of rules and campaign activities in each contest to be able to make any meaningful assumptions about the effect on likely voter screens.

  75. C. Clavin says:

    I have heard a rumor that Romney cloned hair samples from Charles Manson in order to produce his sons…it must be true.
    Or at least as true as the crap Jan spews forth.

  76. Scott says:

    @KansasMom: Agree. But you lost me at K-state. Come on, get yer guns up. Go Tech.

  77. PJ says:

    I’m still waiting for jan to give us a link to the site where she found the Paladin//CFP poll…

  78. C. Clavin says:

    Just out of morbid curiosity I looked at the UC Boulder election forecast that Jan mentioned.
    It’s a forecast based on economics.
    Earlier today I noticed that the Oak tree outside my kitchen window is redder than usual this time of year. Therefore I predict Romney wins 400 EV’s.
    Jumpinjehosaphat…you can’t make up how ridiculous that woman is.

  79. MBunge says:

    Here’s the difference in this election.

    If Nate Silver turns out to be wrong in his forecast, he’ll go back to the numbers and his process to figure out what went wrong and why.

    If Rush, Sean and company turn out to be wrong, they won’t spend a single minute re-examining the data to find out why. They’ll be too busy losing their shit and openly speculating about insurrection and revolution.

    Mike

  80. mattb says:

    @PJ:

    I’m still waiting for jan to give us a link to the site where she found the Paladin//CFP poll…

    Actually, I’m pretty sure she got it from the “home” of Paladin/CFP, The Hedgehog Report (http://www.hedgehogreport.com/).

    She just missed the fact that the entire thing is an inside joke among site commentors.

  81. mattb says:

    And on that note, @mantis, has Tristero/Rosicrucian polling released their latest data yet?

  82. jan says:

    @PJ:

    @mattb:

    I understand why you are such Obamabots, as you share with him the tendency to rivet yourself on pettiness and trivia.

    In the meantime there is this commentary dealing with liberal denial will only get worse. They must be getting their psychological reading from posters on this blog.

    ….. but other voices, including polling analysts like the New York Times Nate Silver and Mark Blumenthal at the Huffington Post, are on slightly firmer ground when they claim that their reading of the polls tells them that Romney’s momentum is over. In a race this close, one has to admit the possibility that they might turn out to be right. But these frantic denials of a Romney surge not only contradict the clear trend of the polls. They smack of the sort of desperation that is often in evidence as candidates who were once thought in a commanding position start slipping. After months of liberals telling themselves that Romney was a fake or a fraud that no one could possibly take seriously, they are having a hard time coming to grips with the possibility that he might be elected president in 10 days. If denial is the first of five stages of grief, liberal mourning about the possible end of the Obama presidency can be said to have begun.

    Feeding this denial is the widespread oversampling of Democrats in polls that still show the president leading the race. The assumption that the turnout of the president’s supporters will match or exceed those that lifted him to a historic victory in 2008 seems to be based more on a leap of liberal faith than evidence, but it is statistical tricks like that that are keeping Obama’s head above water in the polls. Partisans always tend to believe polls that tell them what they want to hear, but systems that weigh polls in an arbitrary manner such as Silver’s forecast seem to be similarly positioned to keep Obama ahead for as long as possible.

    Furthermore, where Tobin is spot-on, is pointing out the intense dislike libs have for their political counterparts, which ultimately effects their objective ability to discuss this race with even civility.

    But even looking beyond the biased analyses being published by liberal sources, the refusal of many Democrats to accept the reality of the Romney surge may be rooted in something more emotional than just skewed poll numbers. Many if not most liberals share the attitude of contempt for the Republicans that were so easily discerned in the attitudes of both President Obama and Vice President Biden during the debates. Though most Americans have rejected the attempt by the president’s campaign to define Romney as a heartless plutocrat or a monster, liberals bought it hook, line and sinker. The idea that such a person could have caught and passed Obama in the space of a few short weeks seems impossible to them not so much because they think the numbers don’t support this thesis but because they just don’t want it to be so.

    Rather than debunking Romney’s wave, liberal analysts who seek to deny it are merely confirming their inability to look dispassionately at what has occurred. Democrats living in liberal echo chambers need a reality check.

  83. mantis says:

    @jan:

    In the meantime there is this commentary dealing with liberal denial will only get worse.

    Trying to unskew reality, jan?

  84. KariQ says:

    Rasmussen is a very good pollster. He is. Once you remember his 2-4 point GOP tilt relative to the actual voting totals (based on both 2008 and 2010 final polls compared to vote totals), you can see that he really does put out excellent polls. Democrats who don’t recognize this are allowing the house effect of his polls to obscure their consistency in reflecting the actual movement, or lack thereof, in the election.

    Case in point: Today, Rasmussen released a new poll of Florida showing the race there tightening to Romney 50, Obama 48. This is down from a 5-point lead for Romney in their previous poll of Florida on Oct 12, when it was Romney 51, Obama 47. This does contradict the “Romney Momentum” argument, but given Rasmussen’s excellent record at following swings in an election, it’s safe to say that Florida is tightening. You’d be foolish to dismiss their findings.

  85. anjin-san says:

    @ Jan

    intense dislike libs have for their political counterparts, which ultimately effects their objective ability to discuss this race with even civility.

    As opposed to your classy, reality based assessment of “libs” as being petty, in denial, obsessed with “trivia” and in general “bots”?

  86. anjin-san says:

    @ Jan

    frantic denials of a Romney surge

    Can you show were anyone denied the Romney surge? As I pointed out upthread, there was a surge after his strong performance (and Obama’s weak one) in Dever.

    There was a surge – it ran out of gas 8-10 days ago.

  87. anjin-san says:

    It seems that Romney, like Bush before him, has put Photoshop to work to make his rallies appear larger than they are in real life:

    http://www.buzzfeed.com/zekejmiller/romney-campaign-appears-to-exaggerate-size-of-neva

  88. datechguy says:

    Seriously this is what we are reduced to?

    Let me say it one more time. Romney is going to win this election and it’s not going to be close

  89. Jr says:

    @datechguy: Romney is losing in Ohio, WI, and NV…….but yeah he is going to win.

  90. KariQ says:

    @datechguy:

    Let me say it one more time. Romney is going to might win this election and but it’s not going to be close

    Edited for accuracy. You’re welcome.

  91. Herb says:

    @jan: When I saw this story, I thought of you.

  92. PJ says:

    @datechguy:

    Let me say it one more time. Romney is going to win this election and it’s not going to be close

    Ok, which states that Obama won in 2008 is Romney going to win this year?

  93. jan says:

    @Herb:

    When I saw this story, I thought of you.

    Always nice ‘to be thought of,” Herb. I’ve seen variations of those kind of photo distortions with some real estate photographs, in depicting room sizes, too.

    If somehow your thoughts of me revolved around those two rally photos that were posted — those were from cell phones and posted on twitter, making it more difficult to alter images. No matter how you want to parse crowd sizes, though, the R/R rallies have been covered by personal comments, newspapers stories and photos, cell phone pics and the enthusiasm for the R ticket is just not isolated to one or two rallies, but among numous functions throughout the swing states.

    Unlike you guys, who have constantly crowed about Obama winning, I don’t take that same posture with Romney. IMO, he has been increasing his support among people, newspaper endorsements and so on, since that first debate. But, it remains to be seen if he can catch up and unsnarl the early deluge of negative ads Obama threw out to damage his character, inappropriately defining him.

  94. jan says:

    @anjin-san:

    “As opposed to your classy, reality based assessment of “libs” as being petty, in denial, obsessed with “trivia” and in general “bots”? “

    If you read, with any objectivity, the responses of most here, to the handful of posters who see this election differently, the ‘denial’ and ‘Obamabot’ comments fit the words and tone used by the OTB majority. I think when you’re in the midst of ‘clan politics’ it’s difficult for you to see the trees through the forest kind of thing.

    Mind you, the same kind of commentary goes on in extreme right wing sites as well, where there is no tolerance or even processing of comments that differ with their ideological preferences. The left and right wing sides of the spectrum simply close their mind-doors to any outside interference with what they want to believe.

  95. anjin-san says:

    @ Jan

    no tolerance or even processing of comments that differ with their ideological preferences.

    No one here has a problem with your “ideological preferences” – it’s the fact that you are a misinformed serial liar that leads to the derision that is heaped on you. I would love to have some Republicans to argue with that were worth the trouble.

    right wing sides of the spectrum simply close their mind-doors to any outside interference with what they want to believe

    Well, you prove this on a pretty much daily basis.

  96. al-Ameda says:

    @jan: .

    @Tsar Nicholas:
    You are one of the most interesting and realistic posters on this site, saying it like it is, willing to show the petticoats of both party’s shortcomings, rather than just harp on the misfortunes of those not clinging to Obama’s reelection.

    Come on Jan, after all I’ve done for you, you betray me like that?

  97. DRE says:

    @jan:

    If you read, with any objectivity, the responses of most here, to the handful of posters who see this election differently, the ‘denial’ and ‘Obamabot’ comments fit the words and tone used by the OTB majority.

    Maybe you should spend more of your time responding in a substantive way to the many people who make substantive comments and responses to yours, and less time engaging in name calling battles.

  98. Herb says:

    @jan:

    “No matter how you want to parse crowd sizes,”

    Yeah, I’m not so big on parsing crowd sizes. I’m fully confident in both candidates ability to draw large crowds….and millions of votes.

    The rest I thought of you is that because if this photo was altered (not saying it was) it was altered for people like you. As Nick Nolte said in the Thin Red Line: “It bucks the men up. Makes the Japs look like they’re catching hell.”

  99. Geoff says:

    @PJ: Let’s take a look at the recent nyt quinipiac poll. The pollsters at quinipiac (pollster for NYT) weights all of its results by looking at party ID… In fact, it gave all results a +7 or 8 democrat weighting which in this istance means that the pollsters had initial results that unanimously favored Romney in all of the key swing states but then went back and added 8 percentage points to Obama’s tally. Why do they do this? They weighted their results to reflect the party turnout of the 2008 election. Besides the numerous studies and polls that would refute this assumption, does anyone in their right mind think that the 2008 turnout is relevant in this election or that it is the best way to predict the results of the coming election? I would think not. Rasmussen has traditionally been the most accurate pollster (along with Gallup) because they are truly INDEPENDENT organizations… Regardless of what you want the polls to say, these pollsters are accurate because they are constantly evaluated on the accuracy of their polling, not the likability of the polls within a political party. Back to weighting- Gallup has done little to weight their polls in this race, and in fact Rasmussen has given a weighted edge to democrats, albeit much smaller than the +8 from the NYT poll. Take a look at the breakdown of the electorate in this years election (gallup does a nice breakdown) and make your assessments of the accuracy of polls based on this, do not rely on a poll to tell you the whole story.

  100. Geoff says:

    @Jr: PPP is a narrative pollster, it has all of its roots in liberal media, and has had some very small sample sizes as well

  101. Gregg says:

    Well written. This fact can be seen so clearly with all the battleground states polling on the RCP website. Rasmussen had Romney either tied or ahead. This may explain why the FOX news people were so wrong with there predictions. I am pleased someone else noticed this and detailed it so well.