Tuesday, November 6, 2012

November 6, 2012 - My Final Prediction

If you haven't voted yet, why are you reading this?  Go vote!

I will not be doing any more poll averages after this, even with final polls that will drop today, like Fox.  I'm going with the final polls that have been released so far, and making my predictions.

Polls included:

Tracking Polls
R+1 - Rasmussen Daily Track
R+1 - Gallup
O+1.6 - IBD/Tipp
O+3 - ABC/WaPo

Others
R+1 - Battleground
O+3 - Pew
Even - CNN/ORG
O+0.005 - NBC/WSJ
Even - Monmouth

O+0.51% - Current RCP Average
O+1.40% - Average using the 2008 turnout model
R+1.52% - Average using the D+3 turnout model
R+3.85% - Average using the 2010 turnout model
R+3.92% - Average using the 2004 turnout model
R+7.02% - Average using the Rasmussen Party ID turnout model

Average Democrat advantage across all polls: D+5.00
Average Independent support for Romney across all polls: +8

I thought long and hard about including the CNN poll in the final analysis, and in the end decided I would do so.  It heavily favors Romney, but I think it helps better capture the Independent support Romney has.  For most of these polls I have posted in depth analysis, and I don't need to revisit those.

As I said before, I am offering three election scenarios.  D+3, Even, and the Rasmussen Party ID.

D+3:  (likelihood 10%) One of the reasons that I think the pundits and pollsters have been expecting a D+3 turnout this year is that looking at 2008, the Dems turned out at D+2, while the Republicans were at R-5 over 2004.  I believe the thinking is that the Dems will still be able to turnout their base and execute their election strategy, but the Republicans will bring their base back to almost 2004 levels.  I believe that this is the Romney campaign's turnout model.  For weeks now, the only cold water that I have had thrown on my expectations has been from the campaign itself.  I keep thinking this will be a blowout, then they leak some internals showing them barely ahead in Ohio.  However, from their perspective this makes complete sense.  They are still polling, and must apply a turnout models to determine the results.  They have to assume that their competition will be able to execute their turnout strategy as well.  You can't go into a turnout election assuming the Democrats won't turn out.  Having said all of that, I think they have over estimated the Democrat's ability to turn out their base.  The early voting numbers are down from 2008, and all of the campaigning on the other side has been designed to appeal to the base.  Obama has made no effort to appeal to the Independents or undecideds.  On the other hand, I think the GOP has underestimated the enthusiasm of their base.  The size of the crowds at rallies has caught them by surprise, as has the number of volunteers.

Using this model Romney will win 49.21% to 47.69%

With this 1.52% win, Romney will take FL, VA, CO, NH, and OH for 279 Electoral Votes.

Even: (likelihood 65%) I view this turnout model to be the most likely.  Obama has a small lead in early voting that is overwhelmed by GOP turnout on Tuesday.  Obama is still able to get his core supporters to the polls, but it isn't enough to compensate for a very fired up GOP base and the Independents.  One factor that everyone is missing is the Evangelical vote.  As I noted in the IBD/Tipp poll, religious votes are favoring Romney by a large number.  This block is able to produce 25 million votes for the GOP, when it votes.

One of the things that everyone misses that the most played commercial of the Obama campaign is footage of Romney saying that he opposes Roe v Wade and will defund Planned Parenthood.  This might be a good commercial for the Obama base and the "War on Women", but it also drives evangelicals to the polls and makes them enthused to vote for Romney.  This ad may end up being the worst decision of the Obama campaign.

Using this model Romney will win 50.38% to 46.52%

With this 3.85% win, Romney will take FL, VA, CO, NH, OH, WI, IA, and PA for 315 Electoral Votes.
 
R+4.2: (likelihood 25%) This is the turnout to end all turnouts.  The irony is that all it take to believe this model is to believe the 3 month average of a poll of 45,000 respondents with a MOE of 1, is actually correct.  The average of the Rasmussen Party ID poll for the last 3 months is R+4.2%.  Even Gallup is finding that R+3 is possible, with leaners.  In this turnout model the GOP gets all of the self identified Republicans, and the reports that Obama has been cannibalizing his election day voters turns out to be true.  Turnout among Democrats is way down from 2008.

Using this model Romney will win 51.96% to 44.94%

With this 7.02% win, Romney will take FL, VA, CO, NH, OH, WI, IA, PA, MI, NV, OR, MN and ME-2 for 355 Electoral Votes.

65 comments:

  1. I want to believe...but it is so hard for me to look past, well...everyone else. Why isn't the Romney campaign seeing it this way? Why isn't Baghdad Nate Silver? And everyone else? Pack mentality?

    ReplyDelete
    Replies
    1. You don't know what the Romney camp is seeing, only what they're saying. And what they're saying (up by 1% in close race) is, not-coincidentally, the exact thing that will maximize GOP turnout (no Eeyores, no "in the bag" folks).

      But you know who IS seeing this? Michael Barone. The man who literally wrote the book on US electoral politics.

      Delete
  2. Your concern is noted. As I said earlier, we will know tomorrow.

    I've made this point numerous times. If we did not have access to RCP, and didn't know a single poll results, who would you think is winning?

    We focus way too much on polls, and are manipulated purposely as a result.

    ReplyDelete
    Replies
    1. That's a really good point, and Peggy Noonan made a similar one in her recent article where she said she think Romney is going to win. We've been staring at numbers on paper and ignoring what's actually occurring before our eyes.

      If it wasn't for the barrage of polls, I would think this election was 1980 all over again.

      Delete
  3. I'm excited. Nervous, but excited. All signs point to a Romney victory but I dunno. The pessimist in me won't allow me to get too excited. Here's praying America comes through today.

    Also, Dave, your D+3 prediction is close to Rasmussen. Hope you guys are right.

    ReplyDelete
    Replies
    1. Turns out my D+3 is only 100ths of a point away from Rasmussen. But I used a completely different methodology.

      Delete
  4. Holy crap, Dave, I came up with the same three scenarios with almost exactly the same odds (I have 10/60/30).

    I've been calling them 2000, 2004, and 1988. We're not winning Cali even at 1988 though.

    ReplyDelete
    Replies
    1. No. That was the sad thing looking at the 1988 map. CA was red.

      Delete
  5. Thanks for your tireless efforts Dave. I voted at 9 a.m. in my heavily democratic precinct in Cincinnati. Turnout was about average with no lines. I placed vote # 164 which is normal for when I vote. No long lines like back in 2008.

    ReplyDelete
  6. Nate Silver's model has a blowout for Obama. Nate has Colorado Obmaa 80% probability. Romney wins CO by 4 and so let's re-calibrate all the rest of his numbers.

    ReplyDelete
  7. Thanks for all your efforts Dave. Whatever the outcome, I think your approach makes total sense. It is crazy to rely on simple weightings of headline numbers given that those are dependent on assumptions on party ID which can be manipulated to produce the desired result.

    The fact that so many polls converged on a tie is deeply suspicious and makes me think that the only value that most of the polling had was hidden in the internals.

    I'm looking forward to a great night which will be even longer for me here in London.

    NJH

    ReplyDelete
  8. BTW I'm pretty sure you used the old EV numbers for your total, at least on scenario 3: I'm getting 355 with those states.

    ReplyDelete
    Replies
    1. I will double check. I have a spreadsheet that adds them up for me.

      Delete
    2. Try a quick recheck on one of the online sites?

      Delete
    3. I rechecked, I forgot to move Michigan into the Romney column. 355 is correct.

      Delete
  9. Dave, you are using the ras PID number for October for your 3 month average. The 3 month average is now 4.23.

    ReplyDelete
    Replies
    1. August D/R/I 33.3/37.6/29.2
      Sept D/R/I 34.2/36.8/29.0
      Oct D/R/I 33.3/39.1/27.5

      3 Month Average 33.6/37.8/38.6 = 4.2

      You are correct. The only error above is in the label, the calculation is correct in the spreadsheet.

      Delete
  10. I think I've been looking at numbers WAY too long.

    ReplyDelete
  11. My personal "realistic worst case scenario" was D+3. If we can win with that model, I feel pretty good.

    Even if you want to take all topline poll numbers at face value and go with an RCP average of Obama +.7%, that still tells me Obama is going to lose purely from a logical understanding of how undecided voters tend to break and the fact that there's never been a President reelected on such a margin.

    Incumbent Presidents either win big or lose big.

    ReplyDelete
  12. I just started reading your blog about 10 days ago. I want to say thank you for all of your work and effort to see through the BS that gets spewed by the MSM and the left. Here's to Romney/Ryan and 4 great years.

    ReplyDelete
  13. Dave,

    This discussion thread on skewed polls is generally skeptical of the concept, but it does have some interesting charts that plot the Obama lead vs. the D+ factor. As you would expect, the 50/50 point falls right around the D+3 threshold.

    http://nomoon.org/forum/showthread.php?tid=15

    I'd be curious to see what your data would look like plotted in such a fashion.

    ReplyDelete
  14. Thanks for all the updates and hard work that went into them, Dave. Gah, I hate this waiting. I really do. I am so ready for this mess to be over with. Honestly, I'm not sure I'll even feel much joy if Romney wins, just relief. Turnout seems to be looking good in the right places so far, but I'm still gonna keep my fingers crossed and worry until we know for sure whenever that is.

    ReplyDelete
  15. Thanks for all your work Dave, ever since I discovered this site I've been hitting it multiple times per day. I am curious if you or anyone else has put their money where their math is? On Intrade right now, Romney is trading at 28, so you'd nearly quadruple your money with a Romney win (it goes to 100 on a win). I just tried to place my own bet, but it'd take too long for the money to get into the account.

    ReplyDelete
    Replies
    1. I was looking to put my money in there, was going to pick up a Romney contract when they were at 20. Thought $2000 would be easy money. But it is run out of Ireland and is hard to get into from the US.

      It's made up of a bunch of Euroidiots who don't understand our elections.

      I figure I would have made 10 grand. Someone is going to be making a pile of cash though.

      Delete
  16. Dave,
    Thanks for all your work. I have really enjoyed the analysis and the chance to sharpen my pencil and do some math at times and for your gracious access to post my points.

    I almost wish I could be with you and watch campaign to see which way the numbers come up!

    ReplyDelete
    Replies
    1. Glad it helped. I'll miss this little band of polling misfits once the only poll that counts is posted.

      Delete
  17. A "thank you" from me, too. I lurk at Ace's site, stumbled on your link there, and have been coming here for a couple of weeks now. Back in the long ago I did grad work in market research so I get into this stuff. All along I knew the polls weren't telling the full story but couldn't find anyone who seemed willing to dig into them.

    ReplyDelete
    Replies
    1. You are welcome. I really started doing this for me, and it was easier for a few others to look at it on a blog, without scrolling off in comments.

      Yesterday, I had 5000 visits. I'm pretty humbled by it all.

      Delete
  18. So King Ewok *just* noticed the Ras party ID. LOL

    ReplyDelete
    Replies
    1. Ace is great at posting about stuff his posters fully discussed 1 or 2 days later.

      The problem is none of the cobloggers read the comments :)

      Delete
  19. It's going well with the turnout numbers.
    an example with real data:

    pasco county, Fla McCain won by 3.5 in 2008

    Numbers By Party ID

    if splitting indies
    54.15 – 45.95
    With 8% Romney indie advantage
    55.04 – 44.96

    198,231 votes total, 64% turnout as 5 pm est
    2008 had 217,115 votes and 73.74% turnout


    contrast:
    Hillsborough County, Fla (Obama 53, Mccain 46)
    Total votes as of 4:30 pm : 328,694
    Total 2008: 513,312

    ReplyDelete
  20. forget about those Hillsborough County numbers, they are not live updating

    ReplyDelete
    Replies
    1. The Pasco County statistic is the key one. That is a swing district. But last I looked, Republicans were crushing the Democrats. With an 8% edge in the "others", Florida will be called very early.

      Delete
    2. They never posted the last update, but I don't like the election results anyways. arghhh

      Delete
  21. Guys, ver depressing.. Going to sleep now

    ReplyDelete
  22. Not going well so far Dave, any comments?

    ReplyDelete
  23. This comment has been removed by a blog administrator.

    ReplyDelete
  24. This comment has been removed by a blog administrator.

    ReplyDelete
  25. Now now be gracious, Dave did a great job building and updating his blog, Go celebrate at your favorite blog.

    ReplyDelete
  26. Dave did in fact expend a lot of effort. I applaud it. But it is clear that effort was in creating an entirely fabricated system of math. And that should not be.

    ReplyDelete
    Replies
    1. that's not true, its not fabricated, everything he did was transparent, he simply took the polls and re-calibrated on different D/R/I scenarios.

      Delete
    2. "Re-calibrated" to create an alternate, self-confirming but totally imaginary reality.

      Delete
  27. Dave, where are you?

    ReplyDelete
  28. AND IT IS OOOOOOOVER! REALITY WINS!

    ReplyDelete
  29. We (conservatives) were by and large starting to sound like losers the last couple of weeks, with the excuses. I wanted to deny that fact and didn't want to concern troll, because it wasn't helpful, so I didn't.

    I applaud your effort, Dave. This country is a lot further gone than I ever wanted to admit. The thing about a 9% response rate is if you do it enough times you get a lot more than 9% of the people... and as a macro exercise, the polls reflected reality. Gotta stare at this asshole for another four years. Unreal.

    ReplyDelete
    Replies
    1. Hear, hear. I've appreciated your analysis, Dave.

      Delete
    2. It is totally unfathomable to me that anyone can have any kind of appreciation for "analysis" that is so completely exposed as utter fantasy and make believe.

      Appreciate the effort, appreciate the team-support. But do call out the product as what it really was - fake and dangerous "science." You people are not entitled to your own facts. Appreciate that more than anything.

      Delete
    3. It is not the critic who counts; not the man who points out how the strong man stumbles, or where the doer of deeds could have done them better. The credit belongs to the man who is actually in the arena, whose face is marred by dust and sweat and blood; who strives valiantly; who errs, who comes short again and again, because there is no effort without error and shortcoming; but who does actually strive to do the deeds; who knows great enthusiasms, the great devotions; who spends himself in a worthy cause; who at the best knows in the end the triumph of high achievement, and who at the worst, if he fails, at least fails while daring greatly, so that his place shall never be with those cold and timid souls who neither know victory nor defeat.

      --Theodore Roosevelt

      Delete
  30. I too appreciate your work. Unfortunately the takers have finally outnumbered the givers in this country. Why work hard to better yourself when you can just vote for what others have earned. It's a sad day in America for both parties. The dems just haven't realized it yet. They're just happy the cool guy that gives them other people's stuff won.

    ReplyDelete
    Replies
    1. Let's go back to Reagan's or Eisenhower's tax rates, those fucking commies.

      Delete
  31. What's killing the GOP is social issues.
    Across the board, we got killed despite an economy that's worse than anything we've ever seen.

    If you took abortion off the table, you'd erase the gender gap and we could address real issues.

    ReplyDelete
    Replies
    1. It's not the social issues. Blacks are anti-abortion. So, too, are Hispanics. Those are groups Republicans lose big in. The bigger issue is that both groups are a bit "tribal" and will simply vote for the party that promises them the most. There's no way to fix that unless you start playing the Democrat's game, which is NOT something I want to see happen.

      Anyway, I'm totally crushed. I haven't cried in a long time, and I actually felt like crying earlier tonight. I'm just speechless :(

      Delete
  32. This comment has been removed by a blog administrator.

    ReplyDelete
  33. Great, here come all the courageous "I told ya so!" people, who never actually "told you so" before the election. Now that it's over, they "knew it all along" and anyone who didn't is just stupid. Dave, you laid out your analysis for all to see, and you are to be commended for doing so. You (and many of us) were wrong because turnout was way more Democratic than expected. Frankly, your way of reporting the polls based on turnout models was largely validated. It is just that you didn't think 0bama could generate that degree of turnout again. Thanks for your work and interesting analysis. It is a lot more valuable than the mindless lemmings who don't even know why they think something other than somebody told them so and it is what they want to believe.

    ReplyDelete
    Replies
    1. Unfortunately, it seems the major flaw in his analysis was a failure to correctly account for Romney's lead among independents. It's apparent now that many of them were Reps who switched. Oh well, live and learn.

      Delete
    2. I don't know if that is true or not. Have to see all the data when it comes in. Right now, it is stunning to me that turnout was so massively down. Remember how Romney voters were more certain to vote in many of these polls? Maybe not. Or maybe the G0P needs to get into the 21st Century in terms of early/absentee voting. Election day voting is a shrinking subset of the pie. As long as our early voting and voting ID measures make it easier and easier to vote outside the traditional process, election day voting will continue to shrink. Republicans cannot keep counting on election day turnout to save them.

      Delete
  34. Here's the one other thing Dave couldn't have accounted for: if Romney turned out McCain's vote in the battleground states, he would have won. They could have switched, or just stayed home, but the next nominee needs to get those numbers at the very least. And yes, if the party doesn't figure out how to craft a more appealing message to Hispanics and some of the Asian-American blocks, we're in trouble.

    ReplyDelete
  35. I am still having a hard time believing that Romney had 3M fewer voters turn out for him than McCain did after 4 freakin' years of having Obama in charge. The national demographics did not change that much in just 4 years. It's damned strange. At the very least, a whole bunch of people in both parties lied their asses off about their enthusiasm level for voting.

    ReplyDelete
  36. Yeah that is a puzzle. That's why I got so excited when I looked at Obamas early voting numbers, they weren't very good, and I figured we'd at least get John McCain's turnout, nope, the Rs stayed home. Head scratcher...

    ReplyDelete
  37. Judging from the behavior of the campaigns and their actions during the last few days of the campaign, I have little doubt that Romney went to bed Monday night thinking he had it in the bag, and Obama thought he was a goner. D+6! Incredible that so many Republicans were willing to stay home and hand the election to Obama.

    ReplyDelete
    Replies
    1. No way. 0bama knew he had the ground game. I was on a Project ORCA call on Nov. 5. Neil Newhouse spoke, and he discussed the two national polls showing a 1 point lead, and referenced internal polls for 6 swing states being with MoE (did not mention the specific states). That's it. He didn't say we are tied. He didn't say we're ahead but its close, he didn't say they look good, we just need to make sure we turn our voters out. He said within MoE. I didn't go around posting this to various places, but to me this said that they thought they were slightly behind across the board in swing states. Which turned out to be the case. I don't think anyone knew for certain, but I do not think that Romney was assuming he was going to win, either.

      Delete
  38. The bottom line on polls... and I got fooled in 2008 ("the polls are crap!") and almost all the way in 2012 ("it's a conspiracy!")... but they call lots and lots and lots of people and get lots and lots of responses. A number of the polls ARE crap. But all those calls, on a macro level, speak for themselves.

    The media runs this country now. They've turned enough of us into ignorant lemmings that they can do just about anything they want.

    ReplyDelete
    Replies
    1. This comment has been removed by a blog administrator.

      Delete
  39. This comment has been removed by a blog administrator.

    ReplyDelete