It's really too bad that the folks behind Five Thirty Eight.com have gone and created such a compelling website based around state-by-state general election polling. It's all really well done and, as such, I can't really bring myself to look away. But this stuff is all really and truly meaningless. Six months ago, no polling showed Barack Obama winning the Democratic race, and no polling showed John McCain winning the Republican race and the general election is about six months away.
« Many Causes! Many Solutions! | Main | Appeasement »
The Map
22 May 2008 12:13 pm
Comments (52)
Useless!? Poblano makes predictions based on demographic data as opposed to polling that implies that demographics is destiny.
Matthew Yglesias is unbelievable.
I didn't read any criticism of fivethirtyeight.com from Matt when it was showing Obama doing better than Clinton in the electoral college.
O Canada!
Our home and native land!
True patriot love in all thy sons command.
With glowing hearts we see thee rise,
The True North strong and free!
From far and wide,
O Canada, we stand on guard for thee.
God keep our land glorious and free!
O Canada, we stand on guard for thee.
O Canada, we stand on guard for thee.
The time looking ahead is long, but the circumstances differ in a fundamental way. In the primary, voters were mostly of the same party and issue differences between the candidates were small and exaggerated to mobilize a following. Predictions were indeed hard to make. Thus, in the summer of last year McCain was given less than a one in ten chance, even a one a twenty chance, of winning.
In the general election, differences are large and well known by virtue of parties and their history, and voting patterns well established. We already can be reasonably sure that the winner of the election will secure less than 55% of the popular vote, probably less than 53%, and that the outcome will turn on the results in a dozen or so states. That is, we can be reasonably sure which candidate will win in three fourths of the states. We can also predict a probable winner come November, albeit with less certainty than we can predict the Belmont or the Eastern Conference playoffs.
Ad hominem attacks, how refreshing.
phil is correct.
The best thing so far about 538 has been its prediction of about a 14 point win in North Carolina and a 2 point loss for Obama in Indiana. Then Mickey Kaus mocked it because it didn't account for Rev. Wright, and that it would be completely wrong. Then it was exactly right.
Joel:
Exactly right for the primaries, but pure garbage for the general election, right?
Tim K, you know you can always write-in Hillary during your next parliamentary election. Most Americans would be quite grateful if you'd take her off our hands. Thanks in advance.
[with gusto]
O Canada, we stand on guard for thee.
It should be noted that Poblano was able to develop a pretty good model for the Clinton-Obama contest only after the contest was well under way and he could analyze the results in the earlier states. He will never in fact get the same kind of data for the McCain-Obama contest, because it is all going to happen on one day.
Ummm, Timmy Baby, the primaries kind of just happened, you know, in a short time frame between the polls and the actual voting. Six months from now is a lot different than a week before voting. Really, the fact that you are so obsessed with one candidate who rode to power on her husband's name just because she feigns right on Israel is just laughable. You amuse me, my little monkey.
Question: when has Obama underperformed polls?
Matt,
The value of those numbers depends on how you choose to view them. If you're looking at the numbers to gauge the outcome of an election 6 months away, you're right, the numbers are meaningless. But, if your questions run beyond simple matter of an outcome, there's a wealth of information with which to evaluate the state, and direction, of the race.
Matt,
Are you saying an Electoral Friedman is a useless unit?
I'm shocked.
And Matt has a blog about sports, his book, and his constant denial and deceptions about his pro-war stance he had in 2003.
A post about another blog, their wp-model and their "usefulness" is ridiculous. Especially coming from someone who's blog seems to be more about his own vanity than anything else.
this stuff is all really and truly meaningless.
Michael Dukakis supposedly had a 17-point lead over George H. W. Bush in July 1988.
He lost in November by 8 points.
For what it's worth electoral-vote.com has had something like this in place for years. To be sure, it indicated a Kerry victory the day of the election.
DTM is right of course about there not being time to make a model in the same way the primary worked. (Though straight demographics seems to be a residue of the last few states to vote.) But in terms of a great site for collecting numbers, or insisting that people put some numbers to their airy "Hillary would win all the states Obama could win in the general" assertions, it's quite useful.
Reality Man and DTM:
This isn't just Pablano's analysis. It's his analysis, it's Karl Rove's analysis, it's the summary at electoral-vote.com, and it's the judgment of many independent analysts.
Of course, despite believing Clinton is the stronger general election candidate in terms of the electoral map, I know that Obama almost certainly is going to be the nominee. I think the relevant question remains how to deal with Clinton and her supporters, how to unite the party, and what role Clinton should play in the campaign.
Should Obama should her for VP? Should he help her pay off her campaign debt? Should she be given the keynote address to the convention? A pre-announced place in his cabinet, perhaps along with Edwards?
I'd be glad to hear others' thoughts.
Tim, I don't think making her the VP would suffice. It's not just winning back her supporters, it's winning over voters who would like to vote for a Democrat but don't want to vote for a black guy who nearly a tenth of the electorate thinks is a Muslim. He's just not a very good general election candidate, and you Democrats will have to hope he sweeps the Southwestern states because he won't win Ohio.
Tim K,
You're drawing conclusions from current polling Poblano himself has not drawn.
As for Clinton, I'm not sure why offering her something personally is necessary, or wise. Indeed, the most important thing Clinton can do for Obama is campaign for him during the general election, but it would undermine those efforts if it appeared that he had bought her support with some sort of personal favor.
So I'd suggest following the Edwards model instead: Edwards asked Obama to promise to go on a "poverty tour" during the general election, which Edwards apparently believes will further a cause in which he is interested. Similarly, Clinton could ask Obama to do something which she believes would further one of her causes. And that avoids the problem above, because it is natural to campaign for someone who is willing to do things to further your causes.
Tim K: Interesting discussion topic, albeit not the topic of this thread.
I have a question to add to your list: Should Clinton shut the hell up?
DTM:
Clinton is going to have 47-48% of the delegates in Denver, and (to be charitable) half of the popular vote from the primaries. You cannot see why it is necessary to offer anything to her personally? Half the Democratic party wants her to be the nominee!
John Edwards didn't win a single primary and commands a whopping 18 delegates, along with less than 1% of the primary vote. It's insulting to compare Clinton and Edwards in that regard.
I know if the positions were reversed I wouldn't be this petulant towards Obama and his supporters. I'd be advocating Obama as V-P, or at the very least a highly prominent role in the general election campaign and at the convention.
Leo:
Do you actually want the Democrats to lose? Because the absolute best way to make sure that happens is to dis half the Democratic party.
I want to read Poblano's take on the new Survey Usa poll which puts Obama ahead in Va. by 7%.
(Suffice it to say that if Obama pulls ahead in Va., the electoral map for November is going to be VERY interesting).
Tim K,
So it is insulting to Clinton and her supporters to treat her as someone who cares about actual causes, as opposed to caring about how she can personally benefit from this situation?
Interesting theory.
Anyway, as noted I think Hillary Clinton (and Bill Clinton too, of course) should indeed have a prominent role in Obama's general election campaign. But again, to the extent it appears they are playing that role only because there is something in it for them personally, it actually undermines their value in that role.
And yes, I would say the same thing if the situation were reversed. I would expect Obama to campaign vigorously and sincerely for Clinton, with no expectation of some personal reward in exchange. Now if she thought he would in fact be the best choice for VP, that would also be fine, but I would be very disappointed in Obama if he demanded that as a condition of his support.
Does Clinton want the Democrats to lose? Because comparing the primaries to an election in Zimbabwe seems to be a clear attempt to make that happen.
This is a dumb post. Why would you compare a choice among over a dozen alternatives six months previous to a choice between one of two alternatives that will be preswented six months from now? And the most recent historical precedent (while also quite flawed as a data point) directly contradicts your assertion. If you look at the polling on Bush/Kerry six months out, it was pretty much 50/50 or even had Kerry as a slight favorite (this was just after Condi's impressively bad performance at the 9/11 Commission hearings and the Abu Ghraib photos came out), so it tracked pretty closely to the final result. Meanwhile, if you looked at polls in September showing Kerry losing by 10+ points, you would be mislead as to the final outcome.
In any case, national tracking polls are already a stupid measurement of the state of the race. Look at the CO polling if you want a sense of bell-weather states in this match up, and which way they trend.
Because the absolute best way to make sure that happens is to dis half the Democratic party.
Then someone should tell Clinton and her supporters to quit doing it. I guess in Tim K's fantasy world, it's not dissing half the party when you tell everyone in states that voted against you that they don't matter because: they're a Caucus state; or they have too many black people; or they aren't big enough; or they tend to vote Republican in general elections.
Tim K is a really disingenuous person.
Tim K.: By saying that McCain is more qualified to be Commander in Chief than Obama, Clinton has DQ'd herself from consideration for VP. It's that simple. If Clinton truly cares about the party and the country, she will campaign hard for the Dem nominee and urge her supporters to support the Dem nominee rather than use her position to extort something for herself.
DTM:
No it is insulting to Clinton supporters to treat her as an also-ran candidate, as opposed to the runner-up who happens to be just as popular as the actual nominee.
Clinton is going to support Obama whether he offers her something or not. What's best for the party though is if her, her team, and her supporters feel they have been given the respect they have earned through the millions of votes and nearly half the delegates they command.
Maybe she should be offered the VP slot, and maybe she shouldn't. I can see both sides of that argument. But what I cannot see is this arrogant attitude whereas she is seen as somehow equivalent to a John Edwards.
You people behave like naive children sometimes.
"Clinton is going to have 47-48% of the delegates in Denver, and (to be charitable) half of the popular vote from the primaries. You cannot see why it is necessary to offer anything to her personally? Half the Democratic party wants her to be the nominee!"
How is that going to bring in all of those people who were just voting for her as a proxy for Bill, which was probably not an insignificant amount of the men who voted for her (though as far as I know has never been quantified)? In addition, key parts of her support, such as Latinos, have actually already switched to preferring Obama over her in polls, so this process is already happening without offering her the spot. The VP spot does little to bring in support at the polls, thus limiting her ability to help him in places like PA that are somehow supposed to be her country now, while likely driving up Republican turnout to vote against her. She is also unpopular out west and would possibly bring the ticket down there. If we want to target Appalachia, we could just put somebody from Appalachia on the ticket and it could be a prominent Clinton supporter like Strickland.
You haven't answered the question yet Tim. How can we talk about bringing the party together without discussing Clinton's clear rhetorical push to tear it apart?
Leo:
First, it isn't Clinton's sole responsibility to unite the party. If she were the nominee it would be her problem, but she's not. Both of these candidates tried to tear each other down. Clinton did it by attacking Obama for lacking the qualifications and experience to be president, and Obama did by calling her divisive, untrustworthy and disrespecting the Clinton Presidency at every turn.
If you want to dwell on her attacks, while ignoring his on her, then go cry about it in the corner. It's not constructive.
"You people behave like naive children sometimes.
Posted by Tim K | May 22, 2008 3:11 PM"
Maybe if you didn't stoop to this whenever anyone challenged your arguments, people wouldn't take so much joy in dissing you. You bring it on yourself. In fact, one could say your inability to deal with criticism makes you a child.
Reality Man:
Oh please. I deal with more than my share of criticism on here with no response even in the face of the most petty and vitriolic invective. I'd love to see how you'd deal with it any better, except you're just one clapping seal here among many and wouldn't know a dissenting opinion if it bit you on the ass.
Tim K, thanks for proving my point for me. The criticism of you on this blog only turned personal when you started calling people naive and stupid for backing Obama, questioning their intelligence when they disagreed with you, resorting to strawmen when you got backed into a corner in an argument and so on. You aren't exactly the paragon of upright discourse you pretend to be. You may not realize how you seem to other people, but you come off as someone who has no ability to argue intelligently for an extended period of time without the vitriol coming out. You dish it out but you can't take it because you simply lack the emotional maturity, which also suggests you don't have faith in your own intelligence and arguments. I would feel sorry for you if you weren't just so fun to smack around.
Obama hasn't said a single nasty thing about Clinton in weeks. I thought Clinton was on board with that program, which was definately a good one for the party. It appears I was wrong. If Clinton is determined to attack Obama's legitimacy, nothing Obama can do will heal the party. He certainly can't put someone on his ticket who will say in public that he obtained his nomination through election fraud.
What's amazing about the fivethirtyeight map is that it shows Obama losing both Ohio and Florida but still pulling off a narrow win. (Actually if you run the numbers it's a 269-t69 tie--what a mess that would be.)
I think it's obvious that Obama will certainly win all of the states the map currently shows in blue except for Colorado and New Mexico. Yes, he might win both of these as well, but I can't imagine him losing any of the others. I'd also give him a very strong chance in both Florida and Ohio. I think the race in both states will be much closer than this site indicates. And he actually stands a chance in Virginia.
So I calculate that the worst result Obama might get is a 255-283 loss. The best I can see him doing is a 325-213, which would involve him winning in CO, NM, NV, OH, FL, NH, and VA. The chances of such a blowout are pretty slim, though.
Tim K,
Your argument frankly doesn't make any sense. I will happily agree that Clinton did better than Edwards. But why does that mean that Clinton should get a personal reward, whereas Edwards instead gets Obama's help on a cause he supports? Again, to me it is very odd that you think "respect" for Clinton implies treating her as if she is only out for personal gain.
And before you stomp your feet about Edwards too much more, may I remind you that in your 2:01 post, you proposed for Clinton yourself "[a] pre-announced place in [Obama's] cabinet, perhaps along with Edwards." It is amazing to me that in just over an hour you went from proposing yourself that Clinton in fact get the same position as Edwards to: "But what I cannot see is this arrogant attitude whereas she is seen as somehow equivalent to a John Edwards."
But I guess consistency is not your strong suit.
First of all, I think that Matt's right that's it too early to draw sweeping conclusions this early, but it was wrong of him to tout 538 when Obama was up in its analysis.
That said, it's not what every independent analyst is coming to, and I do think that while there are some broad trends to keep in mind, it would be a mistake to make a sweeping conclusion such as Tim K.
It's true that 538 does show Clinton to be doing better, but since it's gone back and forth throughout the process, is it really appropriate to broadly deem Clinton more electable or a stronger opponent. One could certainly do so, but it might be akin to assuming that she was the overwhelming favorite to win the nomination last year.
One factor to keep in mind is that Barack Obama's numbers might be weighed down in the short term by Clinton loyalists. (That's what the Nevada Rasmussen poll indicated.)
Lastly, Clinton's strength among certain parts of the electorate doesn't necessarily make her an ideal VP. She may ultimately help marginally in Ohio, but hurt more decisively in places like Virginia or Colorado.
John Edwards didn't win a single primary and commands a whopping 18 delegates, along with less than 1% of the primary vote. It's insulting to compare Clinton and Edwards in that regard.
John Edwards only has 18 delegates because he pulled himself out of the race before Super Tuesday. He could have made a much better case for staying in the race at that point (when less than 5% of the delegates had been selected) than Hillary Clinton can today.
Reality Man:
I'm through responding to you. It isn't worth my time.
DTM:
It's not your solution I have a problem with vis-a-vis Clinton so much as your dismissive attitude.
Poblano's doing good work, because he's attuned to the polling models and the demographics, and that's going to be an issue in what I suspect will be a very different electoral battleground, with a different turnout model.
For what it's worth electoral-vote.com has had something like this in place for years.
And deserved props go to EV.com, but 538's doing something slightly different. The more the merrier.
One factor to keep in mind is that Barack Obama's numbers might be weighed down in the short term by Clinton loyalists.
Both Clinton and Obama poll more poorly against McCain than they otherwise would, because they're currently involved in a separate election. Each candidate's most ardent supporters refuse to even consider the thought of voting for their Democratic rival (because doing so would imply that their favorite might lose). This happened in the Republican race as well and is completely normal.
Once the Democratic nominee is chosen, these maps will change radically.
The election isn't for six months, it's true, but right now, both campaigns are deciding how they're going to allocate resources, trying to figure out where they're strong and where they're weak, and compensating as needed, especially wrt VP picks. So today's polls are very relevant. A few close polls in Texas could get Obama to sink a great deal of his financial advantage there and force McCain to compete. Maybe it works, maybe it doesn't, but it's worth knowing whether it's an option.
Tim K-
You do have a habit of dishing out insults disguised as campaigning for some sort of higher moral virtue. You called me a sexist, which when you did, was the very first time in my 29 years that anyone ever threw that particular charge at me.
Perhaps you get too much static here but you ask for it in part because of your never-ending support of Hillary Clinton. It boggles my mind. If you were a woman and over 50, I would understand. But I’m assuming you’re not, so your level of support makes no sense. She has never done anything to improve your life personally, you didn’t grow up in the same neighborhood, and you never worked for her campaign…or even voted for her. I just don’t get it. So you think she is a great candidate. So what? For example, I really like Sen. Bobby Casey of Pennsylvania for a number of different reasons. But I do hope that I would never get so emotionally attached to him as you are to Sen. Clinton. She lost fair and square, it’s over.
"Should Obama should her for VP? Should he help her pay off her campaign debt? Should she be given the keynote address to the convention? A pre-announced place in his cabinet, perhaps along with Edwards?
I'd be glad to hear others' thoughts."
If she gets out the day after the final primary, June 4th in other words, I'd offer her anything AND everything she wanted except the VP slot. (All the other things you mentioned and/or my full support of her becoming NY Gov or Senate Majority Leader) And not giving her a spot on the ticket is only because I'd think that there is a better VP pick out there. No malice, just business.
But that's IF she gets out by the 4th.
Led @ 3:01 bears repeating:
By saying that McCain is more qualified to be Commander in Chief than Obama, Clinton has DQ'd herself from consideration for VP. It's that simple.
But it doesn't let you split Nebraska to settle ties. :(
Comments closed June 05, 2008.
| 
Heads in the Sand
Poblano's current work seems mostly to be about developing and tuning his methodology, not making accurate forecasts of the election six months out. (Plug in current poll numbers and it'll output garbage, but plug those same numbers in in October and it'll output a genuine clue.) His primary analysis has also been top notch.
Posted by phil | May 22, 2008 12:20 PM