November 3, 2012 Analysis - Final Weekend
Polls included:
Tracking Polls
R+0.8 - Rasmussen Daily Track
R+5 - Gallup
O+1.3 - IBD/Tipp
R+1 - ABC/WaPo
Others
O+1 - Battleground
Even - Pew
R+1 - NPR
O+1 - CBS/NYT
Even - Fox
O+5 - National Journal
O+0.05% - Current RCP Average
O+0.76% - Average using the 2008 turnout model
R+2.02% - Average using the D+3 turnout model
R+4.26% - Average using the 2010 turnout model
R+4.30% - Average using the 2004 turnout model
R+6.04% - Average using the Rasmussen Party ID turnout model
Yesterday we had a 1.2% drop in Rasmussen that lead to a collective freak out by all the nervous poll junkies, because it was reported as a tie. Later in the day, ABC/WaPo posted a gain for Romney. The most important part of the ABC/WaPo poll was Independent preference moving back to +8 and half of the 2% undecideds making up their mind and committing to Romney.
The "second look at Obama" is now over, and we won't seeing any more movement toward him in the polls. There are four reasons for this:
- The bump he was getting from Sandy is now over, and it is becoming a liability as he must campaign while everyone is seeing people homeless and in gas lines. A lot of people are now asking why FEMA can't get gasoline into New Jersey and shelters into Staten Island. The excuse that it isn't FEMA's job isn't going to work after Obama's bold statements during his photo ops. Bloomberg did Obama no favors with the NYC Marathon stupidity.
- Benghazi is now back into the news as the CIA releases timelines that raise more questions than they answer. Additionally, having two GOP Senators succeed in getting access to the prime suspect in 2 days, while the Government couldn't for 7 weeks makes Obama look ineffectual.
- Minds have now been made up. The final weekend is about turning out your base and organization. There are no opportunities left to change voter minds.
- The difference in energy levels and enthusiasm between the two campaigns is palpable. Obama was barely able to draw 2,800 in Ohio in one of his strongholds yesterday. Romney drew 40,000 outdoors in the cold in Ohio. On Saturday, Ryan will be in Minnesota. On Sunday Romney will have a huge crowd in Bucks County, PA. The visual contrast between the campaigns is impossible to ignore.
D+3: The least likely scenario is that Democrats turnout at 3% higher than Republicans. This is the "split the difference" scenario that pundits have been using all year. Take the Obama enthusiasm from 2008 and the GOP enthusiasm from 2010 and assume a turnout between them. This is now very unlikely to occur. We have specific evidence that the GOP 2012 enthusiasm is at or above 2010 levels. Obama's early voting numbers are crashing and he can't fill a venue, signs that Democrat enthusiasm isn't there. Finally Independents are supporting Romney by the same rate they supported the GOP in 2010. However, even with this scenario, Romney leads by 2 points and would get a 50-48 win. The only way that Obama could win is through massive voter fraud. This would require the production of 2,600,000 fraudulent votes and would very likely be detected. We would have a constitutional crisis. In this model, Romney wins close races in Ohio, Colorado, and Florida, winning 275 electoral votes.
Even: 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 thing to keep in mind is that the Rasmussen and Gallup party preference polls are real. They really do show, at this point in time, that the Republicans hold a 1 to 3 point advantage in party identification. In this model, both parties get their bases to the polls, and the GOP misses just slightly their high water mark. In this scenario, Romney leads by over 4 points, and would win 51-47. He takes OH, CO, FL, IA, NH, WI, and PA, winning 315 electoral votes.
R+3: After watching the Romney rally in Ohio, I am seeing this result as much more possible. If Romney's energy continues to build, and Obama continues to make gaffes like "revenge voting", then we see massive turnout on Tuesday. 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. In this scenario, Romney wins by over 6, with a 52-46 final tally or better. This would be a similar result to the 1988 election. Romney would win OH, CO, FL, IA, NH, WI, PA, NV, OR, and ME-2 for 329 electoral votes. He would also have a good shot at winning MN and MI.
Romney is going to win.
Dave, you don't seem like a guy who worries a whole lot, but are you not concerned about Rasmussen's latest Ohio poll? He has 'unaffiliated' voters up by 9 points. Are these independent voters, or is there some statistical difference?
ReplyDeleteDave, Rasmussen's Saturday daily tracking poll shows another tie at 48%. Do you know the exact numbers? Was there movement toward the President again?
ReplyDeleteI am neither worried about Rasmussen's Ohio or National polls. Rasmussen is sampling at a D+3 level and showing a tie in both polls. He is also still suffering from counting more early votes than are actually returned. His likely voter screen is being gamed like everyone else.
ReplyDeleteHe is just a data point now and is showing no movement in his polls. It is confirming my opinion.
Thanks for your response, and for your great analysis in general!
ReplyDeleteI still believe Romney carries Nevada. In Las Vegas there isn’t a single Obama sign anywhere. Romney signs are on ever corner. Crowds are not turning out for Obama rallies. Last poll taken here was D +10, but early voting is about D+5 When Obama said in 2009, “You don’t go blow a bunch of cash in Vegas”, it made having a convention in Vegas toxic. Hundreds cancelled their conventions and thousands upon thousands of people lost their jobs in the Casinos. Vegas hasn’t forgotten that.
ReplyDeleteearly voting numbers here:
http://nvsos.gov/index.aspx?page=1089
Dave, as you know I've been reading you all month and love what you're doing. I also believe that your numbers are solid and that we're going to win.
ReplyDeleteBut this "cannibalizing early voters" stuff *sounds* more like the weird excuse making of the (currently) losing side than reality.
Do we really know this to be true or is it just a "guess" to make our own suppositions about reality fit the numbers? I am not at all convinced that Dems are only getting their likeliest voters out first.
Dave isn't the only the one to say this, other professional pollsters who do the news programs, have said the same thing.
DeleteThere is a statistic that is being looked at. How many early voters have voted in 3 of the last 4 elections. Those voters are considered safe voters that would vote on election day. Voters in 2 or less of the last four are considered low propensity voters. The Dem early votes are over 90% in the first category. The GOP early voters have a lower percentage of them, but still pretty high. I think about 75% last I checked.
DeleteAlso, note that the only time I even mention "cannibalizing" is in the R+3 scenario. In the others I consider it a wash between the parties.
DeleteOk, thanks Dave. I wasn't going after you specifically, by the way. I've seen this "cannibalizing" theory a lot. I got sucked into believing the "tea leaves" on BambiCare and am seriously gun shy this time around. We're gonna do this thing.
DeleteThanks for the update, Dave. I was already feeling better about this after seeing that amazing crowd at the Romney Ohio event last night. Wow. And the blazing fast turnaround on getting Barky's latest gaffe (the revenge thing) out into a campaign ad almost instantaneously.
ReplyDeleteI sure hope we see the R+3 scenario play out...it'll take that kind of whuppin' to get much of anything done after the election. After the last 4 years, this bunch richly deserves being embarrassed on election day.
If only to see the faces on MSDNC (and specifically Chris Matthews), I hope to see an R+3, lol. Of course, any Romney victory, though a clear cut one, so voter fraud won't be a factor and won't give an excuse for the Obama campaign.
DeleteYes, we definitely need a win beyond the "margin of fraud". If the crowd in Bucks Co., PA, on Sunday is anything like that crowd Romney had in Ohio last night, I might just start thinking this is really gonna happen. (fingers crossed)
DeleteI'm going to be watching the turnout in Bucks County too. If it is like last night, then I'm going to start believing in that R+3. For reference, McCain's last minute appearance in PA was about 1500 people.
DeleteI will be looking for comparison between the two.
For the record, the word is that Bucks County will exceed 50,000. I'll wait to see it happen, but very encouraging.
DeleteGreat work, once again. So, on a station like Fox News, save for again Dick Morris and Karl Rove, are they inept at reading the numbers, lazy or want to push this as being so tight in order to drive up the ratings?
ReplyDeleteOn FNC, I'd bet on it being a combo of #1 and #3. I mean, who has time to analyze numbers when you have to find a generator to keep Geraldo's coy alive. :-)
DeleteI agree with DIV. I've gotten into arguments with Bret Baier on Twitter about them using the RCP averages all the time. He really doesn't get it. Math isn't his strong suit.
DeleteSo, they're lazy. It's easy enough to have the graphics department put up the RCP average rather than actually delving into the internals of the polls...
DeleteI have stopped watching Bill O'Reilly and Neil Cavuto for this very reason. They (and others on Fox, too) keep saying it's going to be very close. But, I really believe it's going to be more or less R+3, and a sure Romney victory. All the signs point to that. So, I just find something else to watch. I will be there Tuesday night, though, when most of them will be eating their words. (From my mouth to God's Ears)
DeleteYeah, I've stopped watching Neil, though only because his interviewing "skills" is horrible. He goes on with a long statement and then asks of his guest, "What do you think about that?" Drives me f'ing insane.
DeleteI've mostly stopped watching Cavuto, too, for the same reason as you have, but I happened to be flipping past FBN this morning, and he was doing a live show and was doing a great job of juxtaposing all the happy, every is going great statements from politicians about the cleanup and recovery work going on post-Sandy with video, pics and citizen statements that showed that the politicians are full of crap on that. I was like, jeez, Neil, why can't you be like this more often? I might start watching you again. Apparently he has lots of friends/relatives in the area were things are still a mess and seemed genuinely, personally angry about it. It made him a lot better to watch.
DeleteNice to hear. IMO, the coverage of Sandy has been good, it's only when they get into the election that things start to bug me, lol.
DeleteI've worked in the IT department for the government's library system for over 5 years, so I see a fair amount of liberals and how they're reacting.
ReplyDeleteUnder 25 college kids: The difference couldn't be further from 2008. Obama was a true cult then, so many who liked him just because it was the cool thing to do. That is completely over now. Obama will still win this naive vote overall, and I doubt many flip to Romney, but turnout will be millions less. Under 25s consist of 34% of the country, 100 million Americans. Polls that are using similar turnout to 2008 are ignoring this obvious turn.
Poor government leeching trash: Here is where I'm not so confident, Obama is still openly beloved because they think he "does things" (i.e. gives them free shit), and Romney is stupidly vilified. Most of them still rely on Obama to live, pretty much. But, on the bright side, these are where most of the early votes come from, and Obama is underperforming thus far, another thing some polls are just lying about for some reason.
And there's the general non-politic following people that I know, like most of my family. They knew Obama was going to win in 2008, people say he's smart, so why not give the black guy a chance. I know many of those squishy idiots who are changing. I don't think they account for much, but I'm very confident Obama won't be winning the independent vote by 8 points again.
Thus, I'm thinking that somewhere between Even-D+4 is about right. I don't see how Republicans can equal what they did in 2010, Presidential elections simply draw more the trash and uninformed.
Dems turned out their base really well in 2010 to avoid the full rout. That year wasn't GOP best-case scenario, not even close.
DeleteI agree with Dave's take on this, though I'm pretty sure R+3 means a full midwestern sweep (well, not IL). Would really really like Romney to take PA so we can call this sucker early.
My only quibble is 2004. That was a presidential year where Bush was fighting a huge headwind with Iraq. But the GOP still turned out and we had an even result.
DeleteIf we have a 2004 result, but with Ind at +8, rather than the -1 that Bush got, this will be an early night.
Thus, I'm thinking that somewhere between Even-D+4 is about right. I don't see how Republicans can equal what they did in 2010
DeleteOn balance I agree with you, but the paradigm we're hoping for is a hyperpartisan election like 2004, where both sides are enthused and the middle plays tie-breaker.
The 2008 electorate was D+7, but this was a composite of R-5 and D+2 vis-a-vis 2004.
Republicans saw a large scale abandonment by conservatives, but lets remember September 2008 was a pretty unique time: Lehman just collapsed, the CDO and CDS markets were unwinding, trillions in synthetic assets vanishing.
So, to get 'back' to parity we just need a 2% (plus whatever natural rate of rel. growth) decay in the democratic position from 2008.
Not unthinkable at all, especially in light of accumulating information from early voting which is showing that the rate of democratic voting is dropping after an initial burst, so his numbers are diverging wildly from 2008's trajectory. A very nice sign.
NB: It's ironic how Obama supports love to remind us how bad and what a unique time it was back when he was elected as an excuse for his economic record, but when it comes time to create an electorate model for polling -- 2008 is the new baseline.
Thanks Dave. One huge question I can't seem to get answered by anyone:
ReplyDeleteIf Ras thinks it will be R+1 turnout based on his polling, WHY is he polling at D+3?
He doesn't. His month polling shows R+2.6, but that is party identification. His tracking polls are weighted based on the responses he gets over the last 7 days of his tracking poll. He is getting D+3 in the tracking poll, so that is why it stays at D+3.
DeleteIf he is rationalizing, then he is thinking that a lot of those R+2.6 folks will stay home. However, I can think of multiple reasons that his response rate is coming in with too many Democrats.
Very true, especially after 2010, Walker, and dare I say... Chick-Fil-A..
DeleteWhat about "Romney Democrats"? That seems to be a factor this year.
Well, there will certainly be far more Romney Democrats than Obama Republicans, lol.
DeleteWhen some of my relatives get nervous, I like to ask them....what has possibly happened since the Republican tidalwave in 2010, the Scott Walker vote, the Chick-Fil-A protest, Benghazi, the debates, etc. that has made obambi more popular, more attractive? Nothing, zilch, zero....then they start to feel better.
ReplyDelete