« BACK  |  PRINT

RS

FRONT PAGE CONTRIBUTOR

Gallup retreats and I claim victory

By: Neil Stevens (Diary)  |  August 3rd, 2010 at 12:00 PM  | 

RESIZE: AAA

Pelosi Boehner

From Unlikely Voter: Some may recall when I questioned the recent Gallup generic ballot results with sharp language. I caught them passing off a poll of adults, with the shift toward the Democrats that usually entails, as a poll of registered voters. It got national media attention.

It’s clear to me the message was received, because now in the first release after my criticism, the poll has in just two weeks shown a remarkable 9 point swing toward the GOP.

Where in the week I complained, Democrats were ahead 48-44, projecting only a tiny Republican gain in November and continued control of the chamber by Democrats, now the poll shows Republicans ahead 48-43, which projects a much different result:

Category D R
2008 two party split 56 44
Gallup two party split July 19 53 47
Gallup two party split August 2 47 53

It’s a complete reversal. The latest results swings the electorate from D+12 to R+6, and that net 18 point swing projects to a 51 seat Republican gain from 2008 in the Swingometer, and a 229-206 Republican majority in the House.

Now don’t try it again, Gallup. We’re watching. The Internet is funny that way.

P.S. Rasmussen has a new generic out, too. It never made a radical turn toward the Democrats, and now has Republicans up 46-38. That translates to a 55-45 two party split, a 22 point swing from 2008, and a 60 seat Republican gain.

Tags: , , , , , , ,
Get Alerts

COMMENTS

POSTING RULES  |  LOGIN TO COMMENT  |  REGISTER TODAY
  • sacody

    that I don’t have to hear from liberal columnists about how they thought Democrats were going to get crushed in November until they saw the recent Gallup generics showing Democrats up. Now they can resume their original thoughts of unmitigated doom.

  • AngryMatt

    But I think 52-48 is probably on the mark for this cycle. That still has the GOP winning the House with a 45 seat gain. I’d set that as probably the “average” gain for the number we’re seeing right now. With the prognosticators always a bit behind but even Cook predicting 32-42, I’d say 45-50 seats sounds about right.

  • msctex

    I’d pay to have seen the conversation that took place at Gallup, where they had to acknowledge amongst themselves that someone is actually watching, and understands what they do. If you continue to insist upon accuracy, you may just put them put of business altogether.

  • http://www.twitter.com/AveSharia evanm

    Funny stuff- I actually assumed the change was due to Research 2000 getting sued by Daily Kos. I should have known you were on the case. (Warning: Link is shameless self promotion, though relevant.)

  • downcold

    the Swingometer is wrong. In a R+130 (!) scenario, the Democrats still control AR-01 and AR-02 (!), two of the most likekly seats to flip. In a R+5 gain, (very small) Democrats still control LA-03 (almost a sure Republican get), TA-06 (almost a sure Republican get), and AL-02, where Bright will be hard to knock off. ID-01 is listed as a pickup in a 4 point swing, and we all know it is much lower than that, due to the campaign we are running there, and Minnick’s popularity.

    And there are some Democrats (CT-04 and AL-02), listed as the most likely gains who are likely to be immediately entrenched.

    So I guess you need to factor open seats into the equation to make it more reflective of the election.

  • sacody

    Have you been watching these congressional polls? They are random samples of congressional disticts, but in the cases of NM-1, VA-5, and NC-11, we really start to see the true extent of the building wave. It is becomnig more and more evident that those races listed by Cook as Likely-D are actually Lean-D, those races listed as Lean-D are really Toss-ups right now, and those listed as Toss-ups are Lean-R. When Cook is predicting 32-42 seat gain for Republicans, that should send chills down the spines of Democrats. Realize that Cook is being as conservative as possible in moving races to different categories. Hey, you would be too if you charged that much for political analysis. Just watch; he’ll flip a lot of these in September and October like he has in the past.

  • http://www.hakubi.us/ Neil Stevens

    I’ve been saying that all along.

    At this point last time he projected the Democrats to gain half the seats they did.

  • tngal

    Always plant your flag when you gain territory. Those are the rules. Where you plant your flag is up for debate. (although I’d love to see Toby draw up a cartoon for a visual)

  • http://www.spartanburgteaparty.org karenmartin

    You’re doing a great job keeping them honest. The skills needed to sort out polls/percentages/numbers of elections … well not everyone’s got ‘em. We the mathematically challenged salute you!

  • http://www.veronicaestrada.com Veronica
  • partyof1

    Did Gallup hire former Research 2000 staff?

    Just asking.

  • clintonformccain

    The Dems only needed one “courtesy flush” from Gallup. All the JournoListers have written their articles proclaiming a massive turn in public opinion. The pols will descend upon their union halls and other stocked ponds during recess, citing the Gallup poll as evidence that Americans far and wide love their policies.

  • The_Rebel

    But if the Swingometer is even close to being accurate, we don’t have much to look forward to in MA. It will take a 42 point swing, which gives the Republicans a 313-122 advantage, before a Democrat congressman is defeated here. The Swingometer shows district 6, John Tierney’s seat, as the first to go.

    But it will only take 3 more points for two more seats to go, ha ha.

    We will do our best here to prove the Swingometer wrong. I would think it would only take a 25 point swing to get the first Dem defeated.

  • http://www.hakubi.us/ Neil Stevens

    Swingometer takes the 2008 recruitment/retirement situation and applies the swing.

    So in a big wave… it underestimates. :)

  • http://theminorityreportblog.com Repair_Man_Jack

    The smell of burning journOlistas feels me nostrils with a righteous stink…

  • indyjohn

    are a pack of partisan hacks. They have been guilty of this sort of chicanery for a very long time. Many thanks to Neil, not only catching them with their hands in the cookie jar, but to calling public attention to it.

  • congressworksforus

    Doesn’t that mean by November we could have a 1 in front of those numbers?

  • rdelbov

    11 point shift in two weeks–this historic or perhaps Gallup was just off when it did +6D when others showed republicans ahead?

  • IJB
  • IJB

    As per Jay Cost’s analysis here.

    As such, you can’t put *any* stock in a *single week* result, or in the week-to-week fluctuations, in Gallup. (Something which I find also to be true about Rasmussen’s month-to-month “issues” and “partisan ID” polling…)

    In the case of Gallup’s weekly tracking, or Rasmussen’s monthly polling, I think the best way to handle it is to not look at an individual result, but to look at a polling *average* of several results (e.g. monthly average instead of weekly result, or quarterly average instead of monthly result as Rasmussen does with their “partisan ID” polling).

  • IJB

    Therefore, the Swingometer can’t really account for what will happen in ‘open seats’ AR-01, AR-02, WI-03, MI-01, etc.

    However, you’re all alone in thinking AL-02 is “out of reach” – most people have it down as either “Toss Up” or “leans GOP”…

  • http://www.hakubi.us/ Neil Stevens

    Are they bouncy because they’re changing up the methodology or what?

  • downcold

    Cook has it at Lean D. Having lived in the district until last year, I probably understand Bright better than a lot of people too.

  • downcold

    that the AL-SEN and AL-GOV coattails should bring Roby in, but she starts at a slight disadvantage due to Bright’s personal popularity. As they become better known and it becomes obvious Roby is better for an R+16 district, since she will win.

  • stephaniet

    They just got busted. Nicely done. :D

  • ihateliberals

    just don’t get it. If the Dem’s continue to take away our liberties it won’t be long until, just like Chivas did n Venezuela, they take over the Media. Then they will decide who can report and what they can broadcast. There won’t be a need for Poll takers anymore because the party in power is always right.

  • barleycorn

    +6 Dem to +6 GOP in two weeks is not “bouncy”. A swing that violent would cause severe whiplash.

    As I’ve said before, I enjoy Jay’s insights and commentary, but chalking up Gallup’s irregularities to quaint and benign bounciness just doesn’t wash.

  • barleycorn

    (How’s that for a mixed metaphor?)

    I suspect the Swingometer has a “sweet spot” or a range in which it is the most accurate and I would further guess that that spot is roughly between a 5 & 35 seat change.

    As the climate for one party gets worse and worse the wheels can come off in ways unexpected.

    Barring a game changer such as an October Surprise (Bin Laden captured, the Talaban surrenders) type event, and assuming Republicans do their homework and their footwork, they will pick up 50 – 75 seats.

  • gemimail

    First of all, Scott Brown carried 7 of the 10 Congressional districts including MA-4, Barney Frank’s district, although just barely (50.1%). Most analysts think that just MA-10 (Cape Cod) is in play because it is an open seat. We strongly disagree since Scott Brown carried 6 of the 10 districts by over 56%. Only MA-9, which Brown carried by 56.9%, do we think the seat is safe for the incumbent because Rep. Lynch was politically astute enough to vote against Obamacare.

    In MA-5, Rep. Tsongas won the special election in 2007 by a measly 6.21% margin and Scott Brown won with 56.2% of vote. Explain why this is a safe bet for the Democrats. We believe districts 2, 3 and 6 are also n play. The bottom line is that the Republicans should pick up 4 seats MA if the generic Congressional ballot in likely voter samples reaches 9.2 or better in favor of the Republicans.

    Right now that number is 8.0 which would give the Republicans just MA-10 and MA-5. See: http://www.marstonchronicles.info/index.php?option=com_content&task=view&id=308&Itemid=1144.

  • gemimail

    Notice in the USA Today/Gallup poll done from 7/27 to 8/1, Obama is 12 points negative. The previous Gallup poll done from 7/11 to 7/13 has Obama down by just 2 points. Since a 9 point swing in the generic poll and 10 in the Obama approval poll is not likely due to statistical “noise”, it is only fair to suggest that Gallup has adjusted its partisan weighting to favor the Republicans more than they had been previously. Surely Neil gets credit for that, too.

  • lukematthews

    Maxine Water’s little ethical lapse is quickly moving to other pols like Frank. He is in charge of the whole monstrosity and there are questions about the banks he bought off, I mean funneled money to. Some of these journalists are beginning to enjoy ratting out some of these Democrats. The fact the Globe kept haranguing His Worship John Kerry is almost a fricking miracle.
    Nature abhors a void and there has been such a journalistic void it must be filled. I think Ole Barney may have some ‘splainin’ to do. He may just use his usual ‘nothing to see here’ tactic, but there has been some annoyance over his dollar discount fit of pique. It usually only takes a chink in the armor to give their opponents a voice in the lame-brained media. Once they smell some blood, they can’t help it.

  • Michael Dugas

    Lol I know you meant Chavez. The left Wing Media seems to be ignorant of the fact that once they achieve their goal of helping to create a Marxist / Socialist utopia they will no longer be “needed” and will just become just another bunch of peons under elitist rule.