Friday, November 2, 2012

DON'T PANIC

I knew I should have posted this first instead of working on my Ohio post.

Rasmussen isn't tied, despite what you are seeing on RCP.  The actual number is Romney 48.4 - Obama 47.6.  You are seeing a rounding error.

Please keep in mind that Rasmussen is using a D/R/I of 39/36/25 and is showing Independents favoring Romney by 3 points and a 5 point cross over advantage.

If turnout is Even like I expect, then Romney leads in this poll by 3.6.  If Rasmussen's own party ID poll is correct, then his lead is 5.3%.

Obama is at 47.6% in a D+3 poll 4 days before an election and his early voting numbers have crashed.

Explain to me how this points to an Obama victory?  Because if that happens, it will be the first time in history.  Kerry was at 47.4% at this point, lost the election, and won the Independents.  And he didn't have skewed polls inflating his advantage.

If the Democrats can turnout at D+3 in this environment, given the level of enthusiasm the Republicans have, then they deserve to win.  So quit panicking, go volunteer, and damn it VOTE!

Update: The ABC/WaPo tracker cancelled out Rasmussen by moving two points toward Romney.  This was mostly due to Independent preference moving from +3 to +8 in one day.  Probably a good Obama day (during Sandy) dropped off, while a good Romney day moved in.

20 comments:

  1. Well, I'll still worry since it's my nature and the future of the nation - any chance at having a future for this nation at the risk of sounding melodramatic - does hinge on this thing. But seeing that he's still using a D+3 model does help a lot. I still don't understand why he's doing that aside from my "wants invites to the best DC parties" theory.

    ReplyDelete
    Replies
    1. I've been saying for a month that he has a flaw in his methodology. He will need to fix it by next cycle. This isn't the first time either, he got 2000 horribly wrong and had to fix his model after that one too.

      His daily tracker and his monthly poll can't be both right, but right now they are sitting 5.6 points out of phase with each other.

      Since Gallup is confirming his monthly poll, but conflicting with his daily track, I tend to think the monthly poll is closer to reality.

      We will know on Tuesday.

      Delete
    2. IDK about that theory. If anything, come election day Ras wants to be right. Maybe he's seeing something others aren't or maybe he's being overly cautious. I'm thinking the latter though because based on these other polls, even if Romney wins bu 3 or 4, he'll still be on the closest (maybe second to Gallup).

      Delete
    3. BTW, that was in reply to "davidinvirginia".

      Delete
    4. I'm sure he'd like to be right...well, at least I would, but then again...do pollsters ever go out of business? The bigger, more famous ones I mean? With so much money being in politics nowadays, I would think, even if Ras - or one of the others - misses badly, they'll still have plenty of work in the next cycle.

      Delete
    5. Sadly that is true. As I said elsewhere here, there's no accountability when pollsters are wrong. In 2016 the same ones will be back on the network proclaiming they are right and the media personnel will take it at face value.

      Delete
    6. I think there would be a hit to reputation if they are the only ones to get it wrong. But if you get it wrong with the pack, then its just "oh gee, nobody could have predicted that." We may be seeing a pack mentality in the national pollsters, with nobody wanting to take a big step out. As long as they are within range of the pack, nobody gets hurt. Start to step out, and you are in trouble if you are wrong.

      Delete
  2. Thanks Dave. People need to get a grip.

    ReplyDelete
  3. I just ran numbers from Cuyoga County Elections absentee ballot requests. The numbers look really good!

    Cuyoga County Ohio
    Through Thurs before election

    year dem rep other
    2012 140,006 52,256 86,166
    2010 129,671 50,899 48,656
    2008 151,296 36,232 96,221

    So reps are up and dems are down.


    ReplyDelete
    Replies
    1. Those are nice, about a 27,000 vote swing our way in the heart of Obama's turnout base.

      Delete
  4. slight correction

    Cuyoga through thurs before election

    dem rep other total
    2012 140,006 52,256 86,166 192,262
    2010 126,728 50,609 47,824 177,337
    2008 139,977 35,554 86,238 175,531

    ReplyDelete
  5. which is consistent with what I'd expect. they are going to get out their votes, but the R turnout will be colossal.

    ReplyDelete
  6. Good article from Washington Post about the "Obama defectors"
    http://www.washingtonpost.com/blogs/the-fix/wp/2012/11/02/the-obama-defectors/

    I think the reason Rasmussen polls get conservatives bent out of joint is because they have a reputation of the "Fox News" of polling and people instinctively think they "tilt" to the right. The irony is, I think Rasmussen wants to change his reputation and not associated as a partisan pollster, so he's tilting his models to be a little more democrat friendly. I actually believe Gallup this time is going to be the closest to the truth, curious to see their new numbers.

    I still believe Romney will win solidly, the early voting numbers alone show that Dave's analysis looks dead on. The 2012 electorate will look NOTHING like the 2008 electorate, and those are Obama's die hard supporters that they've been trying to get to the polls early. you can forget about the "cool kid" flunkies that padded Obama's vote totals last time around.

    ReplyDelete
  7. Yes and absentee voting is getting more popular, so matching 2008 D count is not necessarily equal as 2008.

    How about this: 2004 John Kerry: 2,741,165, 2008 Obama: 2,708,685 votes

    Kerry lost by 2.5%, so Bush got more votes in 2004 then Obama in 2008.

    If that data about defectors is even remotely accurate, how is this thing even close? something very strange in polling this time...

    ReplyDelete
    Replies
    1. I've said it a bunch of times, but everyone misreads 2008. They say it was a huge increase in Democrat turnout. It wasn't. It was a giant drop in Republicans. They stayed home, and a pretty good percentage of them voted for Obama.

      Delete
    2. Very true. Reps were not in love with McCain to say the least. The same certainly cannot be said about Romney based on the data thus far...

      Delete
  8. I'm trying not to panic, but it's tough.

    Either way, even though I'm a libertarian that you folks probably hate, thanks to Dave for his effort in breaking down these polls that are probably bullshit.

    ReplyDelete
    Replies
    1. No need to panic. We have 275 EVs locked up. Now we are just talking about gravy.

      And who said I'm not a libertarian?

      Delete
    2. My only concern is Romney getting a clear victory as the Obama attorneys are itching to clock some expensive hours, lol.

      Delete
    3. That's the thing. Not just winning, but winning by more than the "margin of fraud".

      I'm not a libertarian, but I'm leaning that direction in more and more areas, the older I get. For that matter, I don't really consider myself a Republican either...I'm a conservative.

      Delete