"Why I Think Obama Is Toast"

Politics & Elections
1

Over at Redstate, Dan McLaughlin has put together an argument as to why he thinks Romney is a very good shot to be the 45th president of the U.S.; it’s worth a read.

In case you can’t be bothered reading, the central argument is that Obama is going to lose the Independent vote, quite possibly by double digits. The only hope he has to win, therefore, is to either “steal” some R votes from Romney or get a significant number of Democrats to show up.

The first, McLaughlin argues, is unlikely as Republicans generally get more crossover votes than Dems get (that helps to somewhat offset that Democrats generally enjoy a lead among voter registration). The second, he argues, is also unlikely to happen simply because the electorate is going to look far Republican than it did in 2008, due in part to increased Republican registration and Republicans not staying home as they did in 2008 (I’ve made similar assertions on this board). To make the latter point he has posted two graphs-- one for Rasmussen and one for Gallup-- which compare how each predicted party ID to be for 2004 and 2008 presidential elections versus actual exit numbers. In both cases, they were fairly accurate, with Rasmussen being a bit more accurate than Gallup. Both currently show Republicans with the edge going into the final week of the race, with Rasmussen at about R+3 and Gallup R+1.

(He also posts a graph for Rasmussen which show it to be less accurate in off year elections, due primarily to understating R’s.)

Summarized, assuming it true that the electorate is more R than D this year (or even slightly D), without a high Democratic turnout to offset Republicans showing up and Obama losing Independents by a wide margin, Obama has a tough road to reelection.

Discuss.

(And, yes, it does play into the whole “oversampling Democrats” that the left likes to dismiss.)

2

How many people who currently call themselves independents are people who don’t want to admit they are Republican? Are they the same people who were exit polled, said they voted for Kerry but didn’t? I guess I can relate, if I voted for Bush, I damn sure wouldn’t admit it.

3

This argument has been made before on this board, and I keep responding to it. Go back to my OP and click on the link that says “I’ve made similar assertions on this board”. There is scant to no evidence that Independents have been growing at the expense of Republicans. It is, in fact, the opposite trend that appears where Independents have grown at mostly the expense of Democrats.

4

It’s possible that you’re right and Romney will win. I think it’s more likely that the polls, taken as a whole, show the real story or something very close to it.

The polls, as a whole, show a pretty stable lead of 2-4 points for Obama in OH. They show, as a whole, a slightly larger lead for PA, WI, and NV. They show a smaller lead for Obama in IA. And they show something close to a dead heat in VA, CO, and NH, with small leads for Romney in NC and FL. And that together shows a pretty stable advantage in the electoral college for Obama. So that’s what I think is likely to happen.

5

But that’s not what the OP is arguing, is he? He (or at least the article he is quoting) seems to be saying that right now, objectively, Obama cannot win.

6

Why would you expect a low democratic turn out?

7

Well I think he (the Redstate poster) is making certain assumptions, like that the relative turnout will be R > D (or perhaps close to R = D), and if those assumptions hold true, then Obama cannot win. But that’s easy- of course Obama will lose if more Republicans turn out then Democrats. I just don’t think that’s going to happen, or at least not in the swing states Obama is leading in. So I disagree with those assumptions.

8

In case I’m not answering the intent of the OP, I’ll go further:

I think the D/R ratio will be less advantageous to the Democrats than it was in 08, but I still think Democrats will have an advantage. This is because that’s what the polls are saying, and because I believe Obama’s campaign has a superior get-out-the-vote operation then Romney’s. Romney has left that duty to the Republican National Committee- which is strong and organized in some states, but not so strong and organized in others.

9

A very important point is missed in that the the R > D only really matters in a few swing states and Obama has incredibly strong GOTV efforts in those states (OH for example). So while there may indeed be a R > d nationwide, I am not convinced it will manifest itself in these swing states. But I am not convinced it will even happen Nationwide.

10

Which, as I’ve said multiple times, is more than likely a function of polls having a large number of Democrats relative to Republicans and Independents in their samples. If, as it’s been said, that polls do not adjust for party ID, then any poll simply becomes a measure of who decides to take that poll and not a true reflection of the race. Again, you all can disagree, but do you have anything which even leads credence to the idea that we’re going to see near 2008 level turnout for Democrats and Republicans? I produced at least some evidence showing that Democrats registration advantage in most swing states are down precipitously since 2008 while Republican registration advantage has seen, in most cases, a slight uptick. Coupled with Republicans being more enthusiastic this election than Democrats, a reversal from 2008, and you have the case for a much better GOP showing.

Not necessarily a low Democratic turnout, but rather a higher Republican turnout than 2008.

[QUOTE=Gangster Octopus]
A very important point is missed in that the the R > D only really matters in a few swing states and Obama has incredibly strong GOTV efforts in those states (OH for example). So while there may indeed be a R > d nationwide, I am not convinced it will manifest itself in these swing states. But I am not convinced it will even happen Nationwide.
[/quote]

From the link:

11

It’s not going to be identical- in fact most of the polls show that more Republicans will turnout than in 08, and perhaps slightly fewer Democrats. But the evidence I’ve seen shows that the Democrats will still have an advantage, at least in states like OH, WI, and NV.

It would be a huge, huge deal if the polls were to be as wrong as you think they will be. You might be right, but I think it’s more likely that the story the polls tell is pretty close to accurate.

And yes, I do trust guys like Nate Silver- it’s in his best interest to be accurate, and he has a very good track record (not just in 08, but in 2010). I think it’s more likely that his analysis is correct then your Redstate guy’s analysis.

12

How do you know what the true proportion of the voting population is Democratic, Republican, or Independent? You have to determine it via polling!

[QUOTE=538]
But pollsters, at least if they are following the industry’s standard guidelines , do not choose how many Democrats, Republicans or independent voters to put into their samples — any more than they choose the number of voters for Mr. Obama or Mitt Romney. Instead, this is determined by the responses of the voters that they reach after calling random numbers from telephone directories or registered voter lists.

…

But party identification is not a hard-and-fast demographic characteristic like race, age or gender. Instead, it can change in reaction to news and political events from the party conventions to the Sept. 11 attacks. Since changes in public opinion are precisely what polls are trying to measure, it would defeat the purpose of conducting a survey if pollsters insisted that they knew what it was ahead of time.

…

But if there is such an error, the historical evidence suggests that it is about equally likely to run in either direction.

Nor is there any suggestion that polls have become more biased toward Democratic candidates over time. Out of the past seven election cycles, the polls had a very slight Republican bias in 2010, and a more noticeable Republican bias in 1998, 2000 and 2006.

They had a Democratic bias only in 2004, and it was very modest.
[/QUOTE]

Link

I don’t truly understand where these accusations of over-sampling are coming from. Presumably, the accusers have a preconceived notion of what the proportion should be, and when that case isn’t shown via polling, bias is assumed.

Seriously, how do you determine what the party identification proportion should be, state-wide and nation-wide? Do you just assume it must have lower Democratic identification than 2008 and higher Republican and Independent? Why? By how much? How do you determine this rigorously?

13

I went back and checked out what your buddy was punditing 4 years ago.
September 11, 2008 predicts 260 McCain, 255 Obama
Oct 7, 2nd debate “Advantage McCain”
Oct 22 “You can’t trust the polls”

Good luck, OMG!

14

Sage advice from Dan, but I think I trust Nate a bit more.

15

Just a reminder, the real world final tally for 2008: Obama 365 - McCain 173. I’m not sure he’s very good at this…

16

Still a slightly better track record than paid Fox News partisan Dick Morris.

17

Omg keeps making the same mistake in thinking that party affiliation is a demographic.

It’s not.

So any argument based Democrats or Republicans being over- or under-sampled in a Poll is bogus from the get go. As Nate Silver has painstakingly pointed out on multiple occasions, party identification is very fluid, even from week to week for a substantial portion of the electorate. It’s not like a demographic, where for the most part, for example, if you’re male you stay male for your entire life. There are only a few exceptions to that (from the point of view of demographics); unlike gender or ethnicity, many people change voting habits more often than they change glasses styles.

In short: most people change party affiliation A LOT more often than they change gender or ethnicity–to name the two most common demographics used by pollsters. For many people, enough to significantly affect the outcome, party affiliation isn’t part of their day-to-day lives at all; they just vote for whomever on Election Day and otherwise don’t think about it at all.

Thus you can’t treat party ID as a demographic.

The only possible way to accurately ascertain whom people are voting for is by polling, with adjustments for actual demographics.

18

C’mon, give ol’ Dan a break. I’m sure he understands that his “you can’t trust the polls” attitude in '08 was asinine, and he’ll do better this time.

reads the OP and the link once again…

Or maybe not. Ol’ Dan may not be the type that learns from past mistakes.

19

Both Gallup and Rasmussen actually do this, where they poll specifically for this issue. I know Rasmussen tries to control for party identification, though I’m not sure Gallup does. As it is, I refer you to the link in the OP where both actually predict that the electorate will be slightly less Democratic than Republican, a stark difference between other pollsters who are making a few non-historic assumptions (like Republican turnout will be lower than the midterm election). That’s different than simply calling people (not-so) randomly until you get a certain sample, which is what most other pollsters do, which is why you tend to see wild swings from one poll to the next (i.e., one poll is D+9, the next is D+5, the next is D+7, etc.). When you do not try to at least somewhat control for party identification, any poll is simply a measure of who decides to participate in it. It’s not really all that hard to understand.

ETA> It’s like polls which assume a D+9 electorate in Florida on election day when the actual electorate is D+4 (and usually more Republican when you use a LV model). They are clearly wrong, though the wisdom in this thread would be that “they’re right at that time”.

Not that you care or anything, but the above doesn’t have anything to do with the validity of what was stated or written in the OP. It’s nothing more than a blatant red herring. But I’m sure you knew that, so no need to point it out.

20

Oh, right. It’s like touting a stock “past performance is not an indicator of future performance”. Red herring, my ass, it’s painfully relevent to consider his track record. But, since he’s telling you what you want to hear…