Is there a good timeline of which states' polls close when?

This is generally a solid election data site, period.

CNN/ORC just in with their final national poll. They have it 49-49.

The sample is D+11 (!) and independents go for Romney by 22 points.

Topps baseball cards put out an "update" series this year that - among the usual hundreds of traded, midseason callup, and All-Star cards - includes 50 different Obama cards and 50 different Romney cards. Each are identical, except that they list one of the 50 states on the card.

If the candidate on the card wins the listed state, you are in some sort of sweepstakes.

I bought 2 packs a month ago and got one of the Romney cards.
My state? Utah.

Felt great about it until the Salt Lake City paper endorsed Obama.
Ok, that part really happened, but I still like my chances.

I don't think it's a rout by any means but I think he gets enough of the swing states that California pushes him over the top.

The question is how quickly they call swing states. I think most of the networks are still a bit gun-shy after the Florida 2000 debacle, plus I think they mostly really like the idea of a neck-and-neck anybody-could-still-win-this-thing horse race, that I wouldn't be surprised if the networks are reluctant to call states like Virginia and Ohio until well after 11:00 even if one of the candidates is holding a steady 2 or 3 point lead through the first couple of hours of official vote totals.

If I had to pick a number, I'll say the first official declaration of a winner is at 11:30 EST.

. . . hey, nobody's going to remember that I was the idiot who picked Obama to win Arizona

Protip from the 480: No way.

Okay, if we're just making predictions for fun, I don't much see the point of just doing a boring, "yeah, this is what I think". Nate's model is as good as anything, so I'd be inclined to just adopt that as my own, and what's the fun of that?

Your post reminds me: one slightly-counterintuitive-until-you-think-about-it result of Nate's model is that the most likely single outcome is not 303 electoral votes, but 332 (Obama wins Florida but loses North Carolina). So if we trust his model as the best guess available on the state of the race, and we're only concerned about getting the exact number of electoral votes right, 332 is really what we should be betting on, not the mean/median expectation of 303.

A while back:
The editorial I quoted above was from the News, not the Post. The News is not some rightwing paper; they slant left. And they endorsed Obama in 2008.

The NY Daily News, particularly its editorial board, has leaned conservative for many decades. Any impression of liberalism is because the News shares a market with the Post. But they're a meat-and-potatoes, law-and-order, remember-the-good-old-days tabloid.

Besides, do newspaper endorsements carry any weight in 2012? The only one that was even marginally interesting was the Salt Lake paper shunning Romney.

Daily Kos' wet-dream timeline.

But it's interesting to see that at 2030 EST, Romney will lead 142-74.

The sample is D+11 (!)

Where are you getting that number? The poll itself (PDF) refers to 418 Democrats (307) or Independents who lean Democrat (111) (p. 6) and 445 Republicans (293) or Independents who lean Republican (152) (p. 7). I get that being a partisan split of 36-34-30 D-R-I, and an I split of 42-58 D-R in terms of lean, which lines up pretty well with Romney's 59-37 lead among Independents.

I think most of the networks are still a bit gun-shy after the Florida 2000 debacle, plus I think they mostly really like the idea of a neck-and-neck anybody-could-still-win-this-thing horse race, that I wouldn't be surprised if the networks are reluctant to call states like Virginia and Ohio until well after 11:00 . . .

The close race narrative is also better for the networks' ratings. They don't want folks switching to some entertainment program because the result is obvious. On the other hand, no network wants to be the last to call a state, since that may cause viewers to switch to the networks that are getting their calls out quicker. Getting it right counts for a lot, but without Tim Russert, election night coverage seems rather vanilla.

MSNBC, for Obama, at 8:05 PM on May 12, 2011

Fox News for Romney, at 7:00 AM on November 5, 2008, based on a hot tip from Mitch McConnell.

Sorry Andy but Joe's was funny. Yours, not so much.

I know a political scientist who picked Obama to win Indiana and a squeaker in Arizona. Accordingly, we now have a bet going...

The SLC endorsement wasn't a shock either.

They didn't "call" 2008 until 11:00, when California was called for Obama as the polls closed. I remember Charles Gibson counting down the seconds until 11:00, it was very emotional for my father and I (my father said he remembered when there was a time when some black people weren't allowed to vote, in his lifetime).

You must have a young father!

I'm not trying to be snide; I'm just genuinely surprised anyone would think this is so far in our past, unless you are way younger than I am (quite possible.) I am 40--younger than half the Knicks roster-- and my older cousins of my generation went to strictly segregated movie theaters. We went to desegregated schools but our parents--my aunts and uncles--all went to segregated schools, the younger ones were in school when my home county desegregated post-Brown (not massive resisters so earlier than most of the South; people my mom's age farther south would have likely graduated from still-segregated schools in areas that waited post-Brown II.) My grandfather paused in front of bathrooms all the time to make sure he hadn't misread the sign (not my uncles who were younger when the colored signs came down so had transitioned more quickly.) Disfranchisement wasn't the norm--I'm upper South--so the voting thing wasn't as crucial in people's memories (whites were a solid majority.) My mother and all of her older siblings voted for George Wallace for president. Later I moved South, and there lots of people of my parents' generation remembered the first time they saw black people voting. In 1964, 6% of black people in Mississippi voted; by 1966 the number was 66%. The intervening act was obviously the Voting Rights Act. (Numbers are less pronounced for other states.) Barack Obama's father, were he a citizen, would not have been able to vote in much of the South when BHO was born.

Where are you getting that number? The poll itself (PDF) refers to 418 Democrats (307) or Independents who lean Democrat (111) (p. 6) and 445 Republicans (293) or Independents who lean Republican (152) (p. 7).

Those are the registered voters. The likely-voter breakdown is on the Methodology page (p. 29).

My father was born in 1951, so he was a teenager when the Voting Rights Act was passed.

I remember he said he had never seen a black person until he saw one at the Bronx Zoo when he was 8 or 9, and was completely amazed. That's what happens when you grow up on Long Island.

The sample is D+11

The RV sample is just about even, slightly more R in fact. What are they supposed to do if all the Republicans say they aren't going to vote? I guess their LV screen is just way too stringent; would have been better to project 96% turnout. ;-)

I believe the Houston and Orlando papers flipped from Obama 2008 to Romney 2012. Which also means nothing.

as for NY Daily News - dammit, lost the tweet earlier in the day. think I know which credible source had it, but won't risk erring. concept was that DN owner Mort Zuckerman is angry with Obama for not being supportive enough of Israel in his opinion.

on another note, Nate tweeted yesterday that this was a best-case on the right-wing side for how Romney might yet win. Don't see uber-anti-Nates here, but this would undermine a biased argument: he's sending his followers to an alternative point of view:


http://www.pointoflaw.com/archives/2012/11/my-election-prediction.php

My father was born in 1951, so he was a teenager when the Voting Rights Act was passed.

Our parents are only 3 years apart in age but worlds apart in geography! But I can totally see why it would feel so much more distant to him since it was only something on TV.

"I remember he said he had never seen a black person until he saw one at the Bronx Zoo when he was 8 or 9, and was completely amazed."

I've mentioned before that the opening scene in Chuck Berry's autobiography, iirc, is of a Chuck of age 3 or 4 and a bad fire in the neighborhood in East St. Louis. When Chuck saw the white firemen, he was horrified because he thought the fire must have burned the color off them. Granted that's a much longer time ago.

Chuck has a new 'record' coming out soon, by the way.

Those are the registered voters. The likely-voter breakdown is on the Methodology page (p.29).

I interpreted this differently than you did:

Respondents who reported that they had already cast an absentee ballot or voted early were automatically classified as likely voters. Among those likely voters, 41% described themselves as Democrats, 29% described themselves as Independents, and 30% described themselves as Republicans.

To me, that doesn't say that the breakdown of "likely voters" is 41-30-29, it says that's the breakdown of "those likely voters" which I take to be in reference to the specific subset of "likely voters" referred to in the previous sentence, i.e., "[r]espondents who reported that they had already cast an absentee ballot or voted early".

I believe the Houston and Orlando papers flipped from Obama 2008 to Romney 2012. Which also means nothing.

Maybe next to nothing. Around 30 large newspapers flipped to Romney (maybe more with today's additions), but the one state that might be most impacted by newspaper endorsements could be Iowa, where the Des Moines Register flipped to Romney, and the newspapers in the four largest cities all backed Romney. If Iowa is close, that might matter.

Kiko, it's a waste of time to argue the poll truthing. Really all the way around.


That said, in your map, you should probably have flipped Nebraska-2 before you flipped Arizona.

I do think that the odds in AZ are slightly better than the polling indicates, but Obama might lose by 5 points rather than 7. AZ might very easily flip sides in 2016.

I remember he said he had never seen a black person until he saw one at the Bronx Zoo when he was 8 or 9, and was completely amazed.

By any chance, did this "black person" live inside a glass tank meant to represent Antarctica, and eat raw fish?

There are rumors of NY and NJ not stopping voting on Tuesday - giving polls extra time under the circumstances.

If it really is Obama in the 270s or 280s and those two states aren't done until Wednesday, Obama can't win until Wednesday. Could really throw things into some disarray.

EDIT: Sorry - it would be another day of voting in the next 20 days in NY. If Obama needs NYs electoral votes, that would really throw things up in the air.

To me, that doesn't say that the breakdown of "likely voters" is 41-30-29, it says that's the breakdown of "those likely voters" which I take to be in reference to the specific subset of "likely voters" referred to in the previous sentence, i.e., "[r]espondents who reported that they had already cast an absentee ballot or voted early".

I read it a few times myself because it's somewhat ambiguous, but it seems unlikely they'd give the LV breakdown for only early voters rather than for the entire sample. In a 57-page poll report, that would be an odd piece of info. to omit.

There are rumors of NY and NJ not stopping voting on Tuesday - giving polls extra time under the circumstances.

The Brietbartians are going to go ape #### if that happens.

I read it a few times myself because it's somewhat ambiguous, but it seems unlikely they'd give the LV breakdown for only early voters rather than for the entire sample itself.

I understand what you're saying, and it's weird that they don't actually give a breakdown of likely voters if I'm reading it right, but the results just make no sense otherwise. The poll has Obama leading 50-48 among RV with a partisan split of 36-34-30 (D-R-I and w/ the 30 split about 40/60 in lean D/R). If anything, that's a very pro-Obama result, given the Independent split. Then, they do an LV screen that (a) excludes 3-4 times more Republicans than Democrats, which just goes against everything everybody knows about LV screens (Republicans vote more than Democrats), and (b) turns a 50-48 Obama lead into a 49-49 tie despite a wildly more pro-Democrat sample. I don't buy either of those things, especially the second one.

I'm fairly sure if Obama is already 270 EC without NY finishing, then Romney would have long conceeded .

I think Baseball Crank's article is about the best Conservative (sane and logical) Argument that I've seen made, that all models are built on some fundemental assumptions, and if those assumptions are flawed (his example is good, aka the pre 2008 derivative models) then the whole thing can be thrown out the window.

Still, that argument make sense for why it MIGHT be wrong, but also just as likely to mean that it might not be. I'll trust the models until they're clearly proven otherwise.

I don't buy either of those things, especially the second one.

It's a really weird poll; no doubt about that. The RV poll makes no sense unless they found a huge pocket of pro-Obama Republicans and/or independents. The LV poll makes no sense unless those pro-Obama Republicans don't plan to vote, which would be quite un-Republican of them.

Math to come.

A late, game-changing endorsement.

The poll is not D+11. Anyone that can add 2+2 can see it isn't D+11. The poll is skewing Republican. (#1730)

You're looking at the registered voters.

"Anyone who can add 2+2" can see it's simply not possible for this poll to "skew Republican" and also have independents go +22 for Romney but only yield a 49-49 tie.

A late, game-changing endorsement.

Game. Over.

There's been bickering here for months, let's have some fun while we wait. What would be a game changer with 36 hours to go? A legitimate Kenyan birth certificate? Romney converting to Islam? A Joe Biden bender? Paul Ryan burning Ayn Rand books?

From the Florida secretary of state, total early and absentee voting this year was 4,421,453. GOP = 1,733,065, Dem = 1,892,677, Other = 795,711.

Dem advantage is down over 200,000 from 2008, when Obama won the state by just 236,000 votes.

— source

So the math is that the poll has an almost dead even distribution when only looking at RV and ignoring the true nature of independents. When you look closer into the number you discovers that almost none of the "independents" are truly independent. They are either disguised GOPers or Dems which again is why the so called independents are breaking heavily for Romney. Tea partiers voting for Romney is the same as Republicans voting for Romney. When you factor the Conservative-Liberal-Moderate into it you discover that it isn't D+11 in the LV but almost dead even.

On the surface it is 41-29-30 for LV but the reality is that it isn't really that. Again, as several polling sites have said or shown party affiliation is a red herring.

Obama is gobbling up almost all of the people who consider themselves liberal, +21 on people who consider themselves moderate, and Romney is +55 on people who consider themselves conservative.

There's been bickering here for months, let's have some fun while we wait. What would be a game changer with 36 hours to go? A legitimate Kenyan birth certificate? Romney converting to Islam? A Joe Biden bender? Paul Ryan burning Ayn Rand books?

The whole concept of game-changers is interesting. In 2004, Obama made no secret of his drug use, and people couldn't care less. But just four years earlier, Bush almost lost the election because of a DUI from 24 years earlier. It's hard to believe that something like an old DUI can change votes at the presidential level, but it obviously happens.

From the Florida secretary of state, total early and absentee voting this year was 4,421,453. GOP = 1,733,065, Dem = 1,892,677, Other = 795,711.

Florida has just under 12 mil registered voters. Those numbers say that 37% of registered voters voted early or absentee. That's amazing. Anyway, breaking it down by affiliation:

R's - 41% voted early or absentee
D's - 40%
Other (I's and minor) - 29%

already I can see another need for a FLA recount, based on that 109 pct analysis
:)

On Election night we'll see how faithful those voters are to their parties. There is no guarantee that a Dem will vote for a Dem nor a Rep vote for a Rep. Republicans in most polls have appeared to be more faithful to their party than Democrats.

Since I don't want to junk up the other relevant thread, here's campaign contributions by NBA people of note.

A late, game-changing endorsement

The last time Pee Wee Herman had anything to do with presidential issues was when he was arrested doing his John Wilkes Booth impersonation. ;-)

My prediction is Obama 328-210. Earlier on this page Austin mentioned the most likely single outcome from 538's model is Obama 332-206. That's my map, except flipping New Hampshire to Romney. Romney winning NH is certainly not out of the question and him winning here would not really mean anything is regards to the polling in the other swing states.

In 2004, Obama made no secret of his drug use, and people couldn't care less. But just four years earlier, Bush almost lost the election because of a DUI from 24 years earlier. It's hard to believe that something like an old DUI can change votes at the presidential level, but it obviously happens.

Obama let it be known he had done drugs in his first book published in 1995. Bush's criminal record did not come out until just before the election because he did not disclose it, the media found it out. And then he claimed he didn't remember it happening.

already I can see another need for a FLA recount, based on that 109 pct analysis

I appreciate good snark, but my numbers above say that 41% of registered R's voted early, 40% of D's, and 29% of I's. They could add up to as much as 300% and not violate any mathematical principle.

FWIW, here's Brian Burke of Advanced NFL Stats on mixing politics and analytics.

It's a really weird poll; no doubt about that. The RV poll makes no sense unless they found a huge pocket of pro-Obama Republicans and/or independents. The LV poll makes no sense unless those pro-Obama Republicans don't plan to vote, which would be quite un-Republican of them.

Okay, working through the math, lining up the O/R splits by partisan ID with the final results, it does look like they have an LV split of 41-30-29. Which means that their LV screen is knocking out approximately 3.7 times more Republicans than Democrats (and more Independents than D and R combined, but I can kind of believe that), and that the Independents getting knocked out favor Obama 85-15. But then that latter result implies that Obama's leading among Independents in the RV sample (~55-45), but earlier, they asked Independents (in the RV sample) how they leaned, and they leaned Republican 58-42. So, you have a group of right-leaning Independents who prefer Obama but aren't going to bother voting. The results just don't seem to fit together in a way that makes sense. And yes, I put way more time and energy into all of this than it warranted.

Bush's criminal record did not come out until just before the election because he did not disclose it, the media found it out. And then he claimed he didn't remember it happening.

The last part is entirely plausible.

I hate math, despite the fact that I'm probably better at it than 95% of the general population.

math is easy. i'm really good at math. calculus is what's hard, and that's where the power is.

I remember he said he had never seen a black person until he saw one at the Bronx Zoo when he was 8 or 9, and was completely amazed.

Could have been in a cage in 1906.

Obama let it be known he had done drugs in his first book published in 1995. Bush's criminal record did not come out until just before the election because he did not disclose it, the media found it out. And then he claimed he didn't remember it happening.

I agree it wasn't a 1:1 comparison. I just find it amazing that some people apparently switched their votes or stayed home because of Bush's 1976 DUI. It seems like a person would have to be totally apolitical to base their decision on such a thing.

There's been bickering here for months, let's have some fun while we wait. What would be a game changer with 36 hours to go? A legitimate Kenyan birth certificate? Romney converting to Islam? A Joe Biden bender? Paul Ryan burning Ayn Rand books?

It's obviously too late for anything to happen now, as "deny deny deny" would be enough to hold off the hounds until the end.

If something were to have happened this week (assuming Sandy doesn't bury the story), here are the ones that might make a difference:

- Romney's tax returns are uncovered, and he was granted amnesty for shady tax shelters AND not paying any taxes for a large stretch of time.
- Obama's name is found in some old police records as either a potential informant (talked to a cop, once) OR as a possible dealer (gave some of his coke to someone for a few bucks, once).
- confirmed audio/video tape of Romney using a racial slur OR Obama saying "f*cking white people"

Really? Nobody has said it yet?

Dead hooker found in candidate's bed
Dead boy found in candidate's bed

FWIW, here's Brian Burke of Advanced NFL Stats on mixing politics and analytics.

Per usual with Brian, that is a good read.

- Romney's tax returns are uncovered, and he was granted amnesty for shady tax shelters AND not paying any taxes for a large stretch of time.

Oh man I forgot about the tax returns (mission accomplished Mitt!), that is a good one.

There's been bickering here for months, let's have some fun while we wait. What would be a game changer with 36 hours to go? A legitimate Kenyan birth certificate? Romney converting to Islam? A Joe Biden bender? Paul Ryan burning Ayn Rand books?

Joe Biden picked up an underaged girl on the way to a campaign event today. Seriously.

Dead hooker found in candidate's bed
Live boy found in candidate's bed

That's how I've learned it...

Ok, what are peoples' projections for when the first network calls the race?

This is important. I need to know how late I have to stay drinking in a dark, empty bar. I don't think I can handle watching this in real time.

- Romney's tax returns are uncovered, and he was granted amnesty for shady tax shelters AND not paying any taxes for a large stretch of time.

Daily Kos has had this annoying thing since this morning, as per a facebook link flying across my feed.

"I don't think I can handle watching this in real time."

We are electing a President, not a king or an emperor.
Either way, the republic and the democracy will still stand.

It's "dead girl or live boy" -- respect the parallelism!

It's a new world we live in. We're much more tolerant of differences. Romney might even pick up a few votes with a live boy in his bed.

You mean our elections aren't like this?

My (worthless) prediction was made back in October, before the debates. No point in recapping it here.

The first indication that Obama is in trouble will arrive the moment someone on MSNBC utters the word "racism."

It's "dead girl or live boy" -- respect the parallelism!

correct. it's from one of edwin edwards' gubernatorial campaigns. he said the only way he could lose the election is if he was found in bed with a dead girl or a live boy.

The NY Daily News, particularly its editorial board, has leaned conservative for many decades. Any impression of liberalism is because the News shares a market with the Post.

And because, you know, not many conservatives endorsed Obama in 2008, or have largely endorsed Democrats for president.

Besides, do newspaper endorsements carry any weight in 2012? The only one that was even marginally interesting was the Salt Lake paper shunning Romney.

My argument was not that it carried any weight, but that it might indicate a strain of sentiment among some who voted for Obama in 2008.

FWIW, here's Brian Burke of Advanced NFL Stats on mixing politics and analytics.

Meh. He says this:

My only commentary is a warning about how certain we can be about projections of unbounded, real world systems. In an election this close, a swing of just a few percentage points of enthusiasm and turnout would change the projection from 90% Obama to 90% Romney. There seems to me to be too much leverage in something so uncertain to be confident of anything beyond the 60% level.

After the Denver debate, a change of a few points switched 538 from 80% for Obama to about 60%. If a few percentage points threw the model from 90% one way to 90% the other, then yes, the model would be useless since it would be too vulnerable to error variance. I just don't agree that sports and the "real world" are that different. Sure, the "dead hooker/live boy" situation could totally throw the election in the other direction immediately. I don't think we need Nate's model to tell us anything if Obama had a scandal of that magnitude. In football, the odds of winning may be very good if you are lining up for a 30 yard field goal with 3 seconds left down by two, but if the kicker pulls a muscle running onto the field, and the left tackle has to kick it, the odds will change pretty drastically. I would argue that given the amount of polling data available, presidential elections should be one of the easier tasks to predict (not to demean Nate's work at all).

He also says this:

I don't usually do this, but I'm going to go out on a limb and make a prediction in the presidential race. It's 50/50. Prove me wrong!

I know this is meant as a joke, but it's kind of trite. CNN has Obama with 237 EC votes in the bag, and Romney with 206. Other news outlets show similar patterns. Statistically, Obama has a big advantage that he only needs 33 more electoral votes, while Romney needs 64. There is close polling in many of the swing states, but Romney needs almost all of them, while Obama has room for some mistakes. Any model that doesn't see that as an advantage for Obama is totally broken.

The 538 model is predicting Obama will win. The fact that it says there is a 15% chance that Romney could win is meaningless. If Romney wins, this model (likely because of biased polling data) was wrong. Statistically, it likely will be wrong a few times here and there, but the model is only useful if the candidate with the higher percentages wins the election most of the time.

The first indication that Obama is in trouble will arrive the moment someone on MSNBC utters the word "racism."

And the first indication that Romney is in trouble will arrive the moment someone on FOX utters the phrase "And the polls have closed...".

:)

Actually, I'm going to guess the FOX keywords of despair will be "voting irregularities".

I only just noticed that the prediction by Cokie Roberts on ABC's Sunday morning program had an incorrect graphic.
It says "Obama 294 Romney 234", which doesn't add up to 538 electoral votes.
Unless she thinks a third party is going to steal Maryland or Missouri from the big boys, someone done goofed.
I'm guessing she meant 294-244.

**and who knows--maybe he should. He was certainly accurate in '04 and '08, missing by a total of one electoral vote in those two elections *combined*.

So... you're saying Nate was in error, then? :)

Well, even if your head tends to believe that Silver's explanations for his projection make sense, that still leaves a 3 in 20 chance that we'll wake up one morning soon and be faced with Romney / Ryan / a pair of 45 year old Scalias replacing Ginsburg and Kennedy / the regulatory agencies packed with a Club for Growth clones / and the realization that a sizable part of the incoming president's electoral base is certifiably crazy. That's just a partial listing, and that's more than enough to make me a bit nervous, regardless of what my head tells me the odds are of this actually happening.

The Bush presidency badly damaged a huge part of the country's cushion, economically, militarily, and so on. We survived as well as we did in large part because there was a long way to fall. Now there isn't. A Romney presidency could easily cost us in the same way Bush resuming the presidency would.

A bit nervous? With good reason.
A McCain presidency would have put Phil Gramm in the Treasurer's chair. He might have actually been a good bit worse than Tim Geithner. Whatever you think of HRC, John Bolton as Romney's Secretary of State could induce convulsions in the Dalai Lama.

Just to add evidence for the closeness of the race, new polls today have PA and MI tied, and OH, O+2.

This is going to come down to enthusiasm and turnout.

Why does this sound like nothing so much as, 'it'll come down to whichever team wants it more'. As for 'PA and MI tied'? That's funny stuff, right there.

For instance when silver has someone with a 70% chance of winning it doesn't mean he's winning in a landslide it means that the candidates he have up 70% are going to win 70% of the time meaning given 10 such elections he's going to be "wrong" 3 times out of 10 yet people act as if he's saying O is up 70 to 30.

Yes indeed. I'm no more confident of an Obama win than I am when there's one out left in Game 7, men on 2nd and 3rd, and the opposing team's weak hitting MI comes to the plate. Like everyone else, I'm on the edge of my seat, and not remotely certain my team's going to get the out.

But 'Obama wins' is a binary issue, 100 pct or 0.

A coin flip is a binary event too. We can still state the probability.

Yes it is. Of course it is. Anyone who needs reminding of that should simply absent themselves from the conversation until they've read a book on elementary statistics.

In a sane world, people who aggregate polling data should have essentially nothing to either win or lose on election day. Everybody loves to go on about this or that person's secret sauce, but the reality is that they're all essentially just reporting aggregated data.

I don't think this is right. I mean, were I an aggregator in this election I'd also have spent the last year racking my brains to figure out how to adjust polls to account for cell phone users. I've read at least a couple of articles suggesting that typical adjustments STILL underrepresent cell users. Anyone who gets this one right has a big leg up on the competition. There's no way as an aggregator I'd simply total the LVs in, say, Ohio state polls, divide, and use that as my prediction. I'd be tweaking the data in all kinds of ways in order to get the most accurate possible representation of who's pulling the level on who is actually voting.

It's still funny, in a weird way--it's as though most people here are saying that the guy hitting .200 or so probably won't get on base in this at bat, and a few people are yelling, 'are you crazy? There's no way to tell! It's who wants it most. That's what this AB comes down to.'

I admit to some pleasure at seeing Repub pundits and pols scattering like cockroaches as the light of Election Day nears. 'Sandy interrupted Romney's momentum...'

I also think it's a live possibility Christie, by merely appearing bipartisan for a few days, scuppered his national prospects. There's going to be a desperate need to salvage reputations by scapegoating. Christie's an obvious target, and we've seen Repub pols run off the reservation for not savaging Obama enough, never mind being complimentary towards him. One good thing, maybe: the election has been so close there won't be any calls within the GOP to significantly change course. Bad for the country, but good for the Dems presidential prospects in 2016.

Obama does seem to have drawn the largest political crowd ever in New Hampshire. Good to see.

I also think it's a live possibility Christie, by merely appearing bipartisan for a few days, scuppered his national prospects. There's going to be a desperate need to salvage reputations by scapegoating. Christie's an obvious target.

Something tells me that if the GOP tries to blame a Romney loss on Christie, he's going to publicly tell party leadership to pound sand, perhaps in those exact words. And it's going to hurt the GOP more than it hurts him.

Daily Kos has had this annoying thing since this morning, as per a facebook link flying across my feed.

Here's the original Bloomberg story from last Monday. So far the Romney campaign has clammed up entirely about it. I'm not sure what's so annoying about it, although it does seem a bit curious that the MSM hasn't pressed Romney for any answers.

Something tells me that if the GOP tries to blame a Romney loss on Christie, he's going to publicly tell party leadership to pound sand, perhaps in those exact words. And it's going to hurt the GOP more than it hurts him.

Exactly. If Romney loses, Christie will have FAR more political/public capital than anyone else in the GOP. He'll be the disaster leader, and he didn't sully himself with a failed 2012 bid (in the primaries or general election).

I'm not sure what's so annoying about it, although it does seem a bit curious that the MSM hasn't pressed Romney for any answers.

I've always found Daily Kos annoying, and the fact that they couldn't bother to pay some people to pursue this more fully also annoys me.

I'm not sure what's so annoying about it, although it does seem a bit curious that the MSM hasn't pressed Romney for any answers.

I've always found Daily Kos annoying, and the fact that they couldn't bother to pay some people to pursue this more fully also annoys me.

That's the first Daily Kos story I've probably read in four years. I'm no particular fan of theirs. But the meat of the story was in that link to Bloomberg, and while your dig at Kos's cheapness is likely well founded, it still seems strange that the MSM has let it sit. If the MSM have checked the Bloomberg story and found the facts wanting, I'd think they'd have an obligation to say so. But one way or the other, they should have at least contacted the Bloomberg author and editor to question them, especially considering that both of them published their direct contact e-mails right at the end of the article. And it's not as if the story, if substantiated, has no bearing on Romney's character, or lack of it.

Perhaps you missed the part about only a small part of Romney's wealth being in the charitable trust? As did the Kos Klowns. No story, here unless you think giving to charity is bad.

Yeah, Christie is coming out of this smelling like roses, regardless of which way it goes.

Something tells me that if the GOP tries to blame a Romney loss on Christie, he's going to publicly tell party leadership to pound sand, perhaps in those exact words. And it's going to hurt the GOP more than it hurts him.

This year, yes. But in the primaries? The way the GOP is now (and granted, things can absolutely change in 3 years but it's not trending that way), it'll hurt Christie more than fighting back will help him. Bipartisanship doesn't win you primaries.

This story seems entirely at odds with the idea of an Obama win on Tuesday, let alone a big Obama win:

Democrats' drive to retake House falters

Nancy Pelosi has spent much of the past two years proclaiming that Democrats had a great shot at reclaiming the House and returning the speaker’s gavel to her hands.

But her drive to regain the majority for Democrats is on the verge of a complete collapse. Democrats are expected to pick up five seats at best — a fraction of the 25 they need. On the eve of the election, some party officials are privately worried that Democrats might even lose ground and drop one or two seats to the Republican majority.

I know about the power of incumbency, but unless the Dems picked the worst crop of House candidates in decades, they should be making major House gains in any scenario that includes Obama winning reelection.

Christie's non-adherence to "rape pregnancy is God's plan" will hurt him *far* more in any upcoming Republican primary than any post-Sandy support for Obama will ...

The NY Daily News, particularly its editorial board, has leaned conservative for many decades. Any impression of liberalism is because the News shares a market with the Post.

And because, you know, not many conservatives endorsed Obama in 2008, or have largely endorsed Democrats for president.

Ray, I'm just telling you what the Daily News is, which is a "leans conservative" newspaper. Other conservative papers, such as the Denver Post, the Houston Chronicle and the Chicago Tribune also endorsed Obama in 2008. Or compare the last 10 years of the Washington Post's editorial page with the text of the Obama endorsement it published last week, and you'll fear for the whereabouts and safety of its editors.

Exactly. If Romney loses, Christie will have FAR more political/public capital than anyone else in the GOP. He'll be the disaster leader, and he didn't sully himself with a failed 2012 bid (in the primaries or general election).

But none of that capital will be with Republican primary voters. He'll do well in the blue state that is NJ, but there's no way the last week works remotely to his advantage in a primary campaign.

This year, yes. But in the primaries? The way the GOP is now (and granted, things can absolutely change in 3 years but it's not trending that way), it'll hurt Christie more than fighting back will help him. Bipartisanship doesn't win you primaries.

Obviously, I agree. If the rest of the Sandy cleanup goes well Christie's rep in NJ gets even better. The idea that the helps Christie nationally in a far right primary campaign baffles me.

Christie's non-adherence to "rape pregnancy is God's plan" will hurt him *far* more in any upcoming Republican primary than any post-Sandy support for Obama will ...

Iirc, he's pro-life with exceptions. It might come down to how much he's willing to wiggle to wherever the far right is in mid-2015, but I still think he's a much longer shot than he was in mid-October. Btw, if the leaks are true and Christie was a first or second choice as veep, it's likely he was thought to be able to skate through the abortion issue.

No story, here unless you think giving to charity is bad.

Well, I think my subsidizing Romney giving to the Mormon church is bad, as I would think of my subsidizing snapper giving to the Catholic Church.

I know about the power of incumbency, but unless the Dems picked the worst crop of House candidates in decades, they should be looking at major House gains in any scenario that includes Obama winning reelection.

Any scenario except for most of them.

1968: Nixon wins, GOP picks up 5 seats
1972: Nixon wins, GOP picks up 12 seats
1976: Carter wins, Dems pick up 1 seat
1980: Reagan wins, GOP picks up 34 seats
1984: Reagan wins, GOP picks up 16 seats
1988: Bush wins, GOP loses 2 seats
1992: Clinton wins, Dems lose 9 seats
1996: Clinton wins, Dems pick up 2 seats
2000: Bush wins, GOP loses 3 seats
2004: Bush wins, GOP picks up 3 seats
2008: Obama wins, Dems pick up 21 seats

Well, I think my subsidizing Romney giving to the Mormon church is bad, as I would think of my subsidizing snapper giving to the Catholic Church.

As an atheist, I'm willing to let you subsidize me. I promise not to give the money to anyone.

http://xkcd.com/1130/

An Obama victory is good news for Mitt Romney!

Any scenario except for most of them.

The party of the winning presidential candidate gained seats in 8 of the 11 elections listed and made major gains in four of them. In 2000, Bush lost the popular vote, which is consistent with a slight loss of House seats, and 2004 was a very close election.

If Obama is supposedly cruising to reelection while the Dems are down 50 seats in the House, the Dems should be making major gains. It's not like Dem numbers in the House are already at 1976 levels, leaving the Dems with limited upside potential. (The Dems already held 291 seats when Carter won in '76. This year, the Dems hold just 191.)

I actually agree with Joe, to an extent. At the same time, winning 51% of the popular vote among house races will still, probably, leave Democrats in the minority. The simple fact is that it's easier for Republicans to win the House.

No story, here unless you think giving to charity is bad.

I think giving to charity is great. Giving to the Mormon Church so that they can drop millions upon millions of dollars to defeat a gay marriage bill is not quite the same as giving to the Red Cross.

Can't you pretty easily ascribe the supposed discrepancy to Romney's weakness as a candidate? It isn't too hard to imagine that an average-ish Republican candidate would be neck-and-neck with Obama, and under those conditions you would indeed expect little movement in the House.

The party of the winning presidential candidate gained seats in 8 of the 11 elections listed and made major gains in four of them. In 2000, Bush lost the popular vote, which is consistent with a slight loss of House seats, and 2004 was a very close election.

If Obama is supposedly cruising to reelection while the Dems are down 50 seats in the House, the Dems should be making major gains. It's not like Dem numbers in the House are already at 1976 levels, leaving the Dems with limited upside potential. (The Dems already held 291 seats when Carter won in '76. This year, the Dems hold just 191.)

Democrats have more seats now than the GOP did in 1988. Just like then, the out party may have won if they didn't nominate such a weak candidate.

Democrats have more seats now than the GOP did in 1988. Just like then, the out party may have won if they didn't nominate such a weak candidate.

In 1988, the Dems were in Year 34 of their stranglehold on the House of Representatives. In 2012, the GOP is in Year 2 of its majority. Again, I understand the power of incumbency, but it's hard to believe 20 to 40 first-term GOP House members, many of whom are running in districts that are projected to be won by Obama, have developed a tight hold on their districts, especially in the first election after redistricting (and with Congress' approval at just ~20 percent).

Can't you pretty easily ascribe the supposed discrepancy to Romney's weakness as a candidate? It isn't too hard to imagine that an average-ish Republican candidate would be neck-and-neck with Obama, and under those conditions you would indeed expect little movement in the House.

An "average-ish Republican" from years past would be beating Obama. But those people are being run out of the national picture by the Tea Party and others on the extreme right. An average-ish Republican today is someone like Paul Ryan. In some respects Romney is above average in that he can carry swing voters, in other respects he's below average in that he struggled to carry the base.

The party of the winning presidential candidate gained seats in 8 of the 11 elections listed and made major gains in four of them. In 2000, Bush lost the popular vote, which is consistent with a slight loss of House seats, and 2004 was a very close election.

I only listed as many results as I did so as to avoid cherrypicking/end point issues. The fact is, there's not a ton of applicability between the House races of the 60s/70s/early 80s and today.

For the last six elections in a presidential year, the average House gain for the winning party has been 2 seats. Before Obama's win, the average over the previous five elections had been a 2-seat loss for the winner's party. Your "should be major in any scenario" premise is simply incorrect.

Among the last several midterm elections, however, there have been swings of 27, 31, 49, 54, and 63 seats. That's where the volatility is.

If Obama is supposedly cruising to reelection while the Dems are down 50 seats in the House, the Dems should be making major gains.

No. To the extent he's supposedly cruising to victory, it's because of his string of incremental leads in key states. The popular vote is another thing entirely; it's possible that you might have read about this discrepancy somewhere.

Obama won by a healthy 7% last time, and "brought along" 21 seats. If he wins by just 1% or 2% (or minus-1%) this time, how vast should his 2012 coattails be?

No. To the extent he's supposedly cruising to victory, it's because of his string of incremental leads in key states. The popular vote is another thing entirely; it's possible that you might have read about this discrepancy somewhere.

Obama won by a healthy 7% last time, and "brought along" 21 seats. If he wins by just 1% or 2% (or minus-1%) this time, how vast should his 2012 coattails be?

I'd like to agree with all of this.

I only listed as many results as I did so as to avoid cherrypicking/end point issues. The fact is, there's not a ton of applicability between the House races of the 60s/70s/early 80s and today.

For the last six elections in a presidential year, the average House gain for the winning party has been 2 seats. Before Obama's win, the average over the previous five elections had been a 2-seat loss for the winner's party. Your "should be major in any scenario" premise is simply incorrect.

It's unclear why you're calculating averages instead of looking for similar scenarios — i.e., elections in which the winning presidential candidate is from the minority party in the House. 1988 is the only election on your list that really bucks the claim I made. In the other applicable elections, when the winning candidate's party had the opportunity to make major gains, it made major gains.

(I suppose 1996 is another exception, although the GOP majority was relatively slim, the 1994 "Contract with America" was still relatively popular, and the Dems were still feeling the effects of the GOP's increasing strength in the South.)

No. To the extent he's supposedly cruising to victory, it's because of his string of incremental leads in key states. The popular vote is another thing entirely; it's possible that you might have read about this discrepancy somewhere.

Obama won by a healthy 7% last time, and "brought along" 21 seats. If he wins by just 1% or 2% (or minus-1%) this time, how vast should his 2012 coattails be?

With only a slight deviation, House seats are apportioned in the same manner as Electoral College votes. If, as Nate is projecting, Obama wins by over 2 points and wins well over 300 Electoral College votes but the Dems make only token gains in the House (or even lose seats), it will be a substantial under-performance and a major missed opportunity. The Dems will likely have lost House elections in 20 or 30 districts won by Obama.

Kevin Drum, slightly altered:

the Romney campaign Joek is spending a lot of time complaining that the polls are wrong, and that's usually something that only losing campaigns losers do. In any case, we'll know in 48 hours.

(I suppose 1996 is another exception, although the GOP majority was relatively slim, the 1994 "Contract with America" was still relatively popular, and the Dems were still feeling the effects of the GOP's increasing strength in the South.)

In other words the last time an incumbent President ran for reelection two years after losing a majority in the House...Keep whistling past that graveyard.

The Onion's issue by issue guide to the candidates is pretty good

In other words the last time an incumbent President ran for reelection two years after losing a majority in the House...Keep whistling past that graveyard.

Sure, but Clinton basically ran for reelection based on the agenda Gingrich & Co. jammed down his throat. It was shameless, but it was effective. (Clinton was also aided by Dole being a far worse candidate than Romney, and by Perot pulling votes away from Dole.)

Strictly speaking, the 1996 House elections were consistent with the voting in the presidential race.

Josh Marshall had an interesting take on this. I've been in California for the last 10 days tending to a sick parent,...

Best wishes to you and your family, Shredder.

Waits in many places to early vote are hours long. It's disgraceful, and it's vote suppression. After 2008, cutting early voting hours is shameless.

Several counties in Florida came up with over-the-counter, in-person absentee balloting as a way to work around limited early voting. I think, acc to the Miami Herald, that county election HQ in Doral locked their doors, preventing same. Citizens protested, loudly, forcing the HQ to open their doors and resume allowing in person, absentee balloting.

Funny how most of these things are happening in swing states with Republican governors, eh? Sickening. Ohio's Gov lost his bid to entirely eliminate early voting there, but did manage with Republican Secy of State John Husted's help to both limit early voting hours (to fewer than 2008), and increase the complexity of provisional voting

From the WSJ online:

What if in Ohio next week the opposite is true? There and in other swing states—Wisconsin, Iowa, North Carolina, Florida—the evangelical vote is flying beneath the media's radar. It's a lot of voters not to notice. In the 2008 presidential vote, they were 30% of the vote in Ohio, 31% in Iowa and 26% in Wisconsin.

So, evangelical voters aren't answering their phones? Aren't responding to pollsters? How would they 'fly beneath the media's radar'?

I look at what Nate's doing as similar to James's Favorite Toy, or whatever James's system was for projecting careers out (Brock2?). A fun tool for sh^ts and giggles. Not something to base any serious argument on, as people are doing here.

Well, that's credible.

With a race this close, any prediction is literally a guess.

This is not true. And the race really isn't all that close. It's only close if you avoid/discredit large amounts of data.

Correct.

No offense, but why would someone with the initial claim post to this kind of numbers-oriented website? You've just said, numbers don't matter, in the sense that huge amounts of data are meaningless or are impossible to interpret, and nothing within a handful of percentage points across polls including hundreds of thousands of voters is other than a complete coin toss.

Fascinating: the final email to Romney supporters from the campaign was a fundraiser. From the Obama campaign, it was 'Get Out the Vote!'

With only a slight deviation, House seats are apportioned in the same manner as Electoral College votes.

This is simply wrong. And backwards, besides.

ETA: This is from September, but the points still hold,

"The bottom line: our model predicts Democrats will win 194 seats (44.6%), one more than they currently hold, with a one in four chance that they will take back the House. Democrats do better in terms of vote share—48.9%, or almost two percent more than they got in 2010—but the extra votes don’t make a difference for the expected seat gain."

and

"It’s hard to miss the 5-point gap between vote and seat share in 2012. This gap is not uncommon in House elections, but it is on the high side (the average gap over the past 60 years has been about 3 points)."

http://themonkeycage.org/blog/2012/09/17/forecasting-house-elections/

and the followup: http://themonkeycage.org/blog/2012/09/19/is-the-2011-redistricting-hurting-democrats-no-but-republican-incumbents-are/