Friday, October 19, 2012

Romney Nearly Tied with Women in Swing States

Barack Obama's leads are evaporating with key demographics, at least if you believe the polls. (I for one believe that the polls are now beginning to reflect reality after months of cooked polls.)  According to a USA Today/Gallup Poll published earlier this week, Mitt Romney leads the President 50 to 46% in the Swing States. (Nationally, Gallup's 7-Day Tracking Poll has Romney leading the President by 7%).

But there are some further key demographic details where Romney is really setting himself up for victory.

- Romney maintains his solid lead with male voters in the Swing States, ahead of the President 52% to 44% (that's a 8% lead for those of you from Palm Beach County, FL).

- Romney is now within 1 point of the President with female voters, trailing the President by merely 49% to 48%.


- Romney is leading Obama 46% to 36% with Independents according to the other most recent swing state poll, POLITICO/George Washington University Battleground Poll (since the USA Today/Gallup poll did not give this particular breakdown). Just to give you a rough idea, the President would need to win 83% of the undecided Independents to win Independents (and that by 1%). That's simply not going to happen. (Remember, historically undecided voters this late break hard for the challenger).

Remember when women were Obama's victory firewall? Yeah, about that. The President is ostensibly tied with Governor Romney. He's also losing men by 8% and Independents by 10%. This is bad, bad news for the President.

I will go on record right now: If the President only wins women by 1%, loses men by 8% and loses Independents by 10%, do you know what we'll be calling Mitt Romney come November 7th? Mr. President-Elect.

7 comments:

  1. Obama is now back in the lead:

    http://www.realclearpolitics.com/epolls/2012/president/us/general_election_romney_vs_obama-1171.html

    And is comfortably ahead in: Penn, Wisc, Mich, IA, NV, NH, and OH.

    On top of that, the RAND corporation also has Obama quite strongly ahead.

    ReplyDelete
  2. Yeah, except for that includes the Tipp Online poll that oversamples Democrats by 7% and the UConn poll that oversamples Democrats by 8%. In 2008 there was an 8% Democrat turnout advantage. Absolutely everyone with a clue agrees that won't happen this year.

    He is not "comfortably in the lead" In PA, WI, IA, NH or OH.

    Let's go alphabetically:

    In Iowa, only ONE poll shows him with a "comfortable lead" and that is a complete anomaly compared to the others. Lets see we have Tied, Obama +2%, Romney +1%, Obama +3% (all four of those within MOE) and then wait a minute...+8% Obama? Lets go back to Sesame Street since Obama loves it so much... "One of these things is not like the other..."

    http://www.realclearpolitics.com/epolls/2012/president/ia/iowa_romney_vs_obama-1922.html

    In NV all four polls are within MOE. Not a comfortable lead.

    New Hampshire doesn't even HAVE Obama in the lead according to the RCP average. We have Romney +4, a tie, Obama +2% and and a Romney +1%.

    Now let's talk about Ohio. Before I continue, I'd like to state for the record that Ohio's party identification is +1% Republican. There are 1% more self-identified Republicans than Democrats in Ohio.

    http://en.wikipedia.org/wiki/Political_party_strength_in_U.S._states

    Now to the polls: Obama is winning in the polls from Survey USA (+3%), PPP (+5%), NBC (+6%) and CNN (+4%) and Rasmussen (+1%). Romney is up +1% in both ARG and Gravis.

    Except if you look at the polling samples you find the following:

    Survey USA sample is +7% Democrat (a total +8% oversample), PPP uses a +4% Democrat sample (a total +5% oversample), NBC/WSJ uses an absolutely insulting 11% Democrat oversample (a total +12% Democrat oversample) and CNN uses a +2% Democrat oversample (while this is at least believable, it's still a +3% Democrat oversample).

    ReplyDelete
    Replies
    1. First of all, according to Gallup, looking at "party affiliation" is completely useless. They blatantly state this. Second, Pew's poll shows that there are more people who affiliate as democrat than republican. Your whole notion of party affiliation is completely false at best.

      Second, Rasmussen has it even, and TIPP has it at +2. Both are well within the margins of each other. YOu can't dismiss TIPP without dismissing Rasmussen. Rasmussen doesn't even list the party affiliation associated with their daily tracker polls, anyways.

      Third, I see nothing discussing part affiliation of the voters in your Wikipedia link.

      Fourth: "In NV all four polls are within MOE. Not a comfortable lead." Oh, man, that's good. Romney hasn't lead outside the MoE yet, but you're fully convinced of a Romney victory. That's good and unbiased "logic" there.

      Oh man, this is going to be good. When Romney loses this election and all of your dilettante analysis is shown to be completely dilettante and wrong, I'm going to make sure every liberal I know comes here and rubs it in to the point where you quit your blog.

      Seriously, you need to read a book or two on election forecasting. I doubt you'll really understand the math or logic behind it though.

      Delete
    2. "First of all, according to Gallup, looking at "party affiliation" is completely useless. They blatantly state this. Second, Pew's poll shows that there are more people who affiliate as democrat than republican. Your whole notion of party affiliation is completely false at best."

      Except that a) Pew's poll is of the nation, not the state of Ohio b) Rasmussen shows a +3% Republican affiliation advantage. So you're basically picking and choosing the one you like. c) Show me the turnout in the past 3 elections in Ohio. Not one of them, including '08, involves a +7 Democrat turnout or greater.

      "Second, Rasmussen has it even, and TIPP has it at +2. Both are well within the margins of each other. YOu can't dismiss TIPP without dismissing Rasmussen. Rasmussen doesn't even list the party affiliation associated with their daily tracker polls, anyways."

      Tipp is using a turnout model based on last election which nobody expects to happen. Rasmussen has been on record many times saying they are using a +3% Democrat turnout model.

      "Third, I see nothing discussing part affiliation of the voters in your Wikipedia link."

      Check the links at the bottom. Sheesh.

      "Fourth: "In NV all four polls are within MOE. Not a comfortable lead." Oh, man, that's good. Romney hasn't lead outside the MoE yet, but you're fully convinced of a Romney victory. That's good and unbiased "logic" there."

      "Oh man, this is going to be good. When Romney loses this election and all of your dilettante analysis is shown to be completely dilettante and wrong, I'm going to make sure every liberal I know comes here and rubs it in to the point where you quit your blog."

      BAAAAHAHAHAHAHAHAHAHAHAHAHA! That's rich. When Obama wins the election. Yes. Because historically undecided voters break hard for the incumbent when they're undecided this close. I can't wait to write my November 7th post. Seriously.

      "Seriously, you need to read a book or two on election forecasting. I doubt you'll really understand the math or logic behind it though."

      Honestly, you almost got this comment deleted on that alone. I left it up because I want your prediction of an Obama victory on the record. It WILL be quoted on November 7th after Mitt wins. Anyway, enjoy living in Imagination Land. Your wakeup call is a mere 18 days away.

      Delete
  3. By the way just to give you an example of unskewing that poll: I'll take Survey USA and be generous and say it's a +1% Democrat turnout in November. So now we're at a 36% Democrat, 35% Republican and 26% Independent.

    Now we're just taking the data right from the poll, Romney wins 89% of Republicans, 6% of Democrats and 41% of Independents (Straight from the poll). Obama wins 88% of Democrats, 5% of Republicans, and 31% Independents.

    The result is Romney 45%, Obama 42%.

    I can do precisely the same thing with the rest of these polls. My friend, you're living in a fantasy world.

    ReplyDelete
  4. You're welcome to live in Happy Imagination Land where a 2% lead is solid and Obama's set to win. You'll be shocked on election day. I won't be.

    ReplyDelete
  5. To the individual who has been debating with me on this post:

    I am not publishing your last comment. If you read the Rules for Comments you will find that you have violated almost every requirement listed.

    I will however answer one issue you raised. You claimed that in close elections the incumbent wins, historically.

    Clearly the only history you know is 2004. Let me correct you.

    2000 there was no incumbent (although Gore could be considered the incumbent since he was the incumbent VP).

    1996 was a comfortable 9% victory for Clinton.

    1992 was a comfortable enough 5% victory for Clinton, but if we want to call 5% close, the incumbent lost.

    1988 had no incumbent, but the incumbent VP won by 8%.

    1984 was a landslide victory for Ronald Reagan.

    1980 the polls had Carter winning by about 3-5% only a few weeks before the election. Reagan won.

    1976 Jimmy Carter won by 2%.

    1972 Richard Nixon won re-election in a landslide defeating George McGovern by 23% (which as a side note makes the Watergate Break-In one of the stupidest acts of cheating ever. It's like Major league ballplayers corking bats to play a junior college team.

    1968 there was no incumbent but Richard Nixon defeated the incumbent VP Hubert Humphrey by 1%.

    1964 Lyndon Johnson won handily by 32% of the vote.

    1960 there was no incumbent but John F. Kennedy beat the incumbent VP by less than 1% of the popular vote.

    1956 Dwight Eisenhower won by over 15% in landslide.

    In 1952 there was no incumbent nor did the incumbent Vice President run. For the record, Eisenhower won in a landslide.

    So basically the incumbent lost in close elections in 1976 and 1980. The incumbent Vice President lost in close elections in 1960, 1968, and 2000. Counting incumbent VPs with incumbent Presidents, that's five losses for incumbents in close elections to one win, and the 2004 Election wasn't as close as you'd like to believe (Bush won by 3%).

    ReplyDelete

All posts will be reviewed subject to the Rules for Commenting. Any post that does not abide by these rules will not be posted, entirely at the discretion of the blog editor.

Commenters who repeatedly violate these rules will be permanently banned from commenting, and thus none of their comments, regardless of content, will be posted.