Articles and Analysis


Guest Pollster: The SurveyUSA 50 State Poll and the Electoral College

Topics: Barack Obama , Hillary Clinton , John McCain , Karl Sigman , Robert Erikson , SurveyUSA

(Today's Guest Pollster's contribution comes from Professors Robert S. Erikson and Karl Sigman of Columbia University.)

In late February, SurveyUSA interviewed 600 registered voters in every state for a total of 30,000 interviews, ascertaining preferences in a McCain-Obama and a McCain-Clinton race. The focus was a new set of electoral maps of red and blue states based on who led each state in the survey. Based on who won each state in the SurveyUSA survey, Obama defeats McCain 280 to 258 while Clinton defeats McCain 276 to 262 in the Electoral College.

Of course SurveyUSA's mammoth undertaking at best presents a snapshot of the states at one point in time. And even if all the niceties of polling were perfectly met, the allocation of states as "red" or "blue" is problematic due to sampling error. Here, we take the analysis of the SurveyUSA 50 state polls one step further. Rather than assign states based on who leads in the state surveys, we assign states probabilistically to the Democratic or Republican candidate based on the SurveyUSA state polls. Then, based on these probabilistic estimates, we ask the question, given the SurveyUSA results, what are odds of an Obama or Clinton victory in the Electoral College?

To do this, we conducted one million simulations (in MATLAB) of the Obama-McCain contest and then one million more simulations of the Clinton-McCain matchup. In each case we assume that the state estimates were correct except for sampling error. Using sampling theory and the assumption of simple random sampling, we draw one million estimates of the vote for each state. In each case we draw from a normal distribution with the observed mean (percent Democratic vs. percent Republican) and the standard deviation determined by the number of respondents in the state reporting a preference (always slightly under 600).

What do our results show? First, we pooled the state polls to ascertain the national vote, weighing each state's percent in proportion to the size of its House delegation. We also assign the District of Columbia as a 436th district and assign each Democratic candidate 85 percent of the vote to McCain's 15 percent. With these assumptions, the national popular "vote" is tight as of late February. Obama wins 51.5 percent versus McCain's 48.5 percent. Clinton also wins by an even razor thin margin, 50.7 to 49.3. With 30,000 cases, both estimates are statistically significant. McCain would be in the actual popular vote lead less than one time in 20.

That being said, our simulations yield a 88% chance of Obama beating McCain (with 306 Electoral College votes on average versus 233 for McCain), and a 74% chance of Hillary beating McCain (with 285 Electoral College votes on average versus 253 for McCain). About one percent of our simulated outcomes were Electoral College ties. (We ignored within-state variation in Maine and Nebraska, which divide their electoral votes by district.)

On the one hand, we find the expected numbers of electoral votes (the average from the simulations) for Obama or Clinton to be slightly higher than SurveyUSA reports. On the other hand, there is sufficient variance in the outcomes, so that McCain wins a nontrivial portion of the simulations, even with Obama as the opponent. Our two million simulations remind us that the popular vote winner is not always the Electoral College winner, although probably due mainly to chance -- the lottery aspect of the Electoral College -- and not any identifiable partisan bias in the 2008 Electoral College.
We thank Linda Liu for her technical assistance.



Sorry to say, but an absolutely useless bit of academic analysis with no new insights. I hope our professors find something more challenging than running MATLAB simulations with SurveyUsa data.




Excellent analysis! Any chance of linking the statistics on the results of the simulations?



chandra - i enjoyed reading the analysis....even though it wasnt too enlightening.

also -- i think these surveys just show that the dems are more popular RIGHT NOW due to the attention given our primaries. novemeber is a long time from now in terms of shaping voter opinion.


I couldn't follow the technical aspects.

What I would like to see are the poll numbers from March to November 2000, and from March to November 2004, with an analysis of how much volatility there was over those time periods in key measures like Gore / Bush favorability, head to heads, etc.

Because by any other name, a snapshot taken 7 months out is still a snapshot taken 7 months out.



Ciccina - there is always a convention bounce...and i think Bush started out way ahead of gore in 00 - but basically tied him.



kerry had a 7 point lead over bush in a cnn poll in feb. 04. http://www.cnn.com/2004/ALLPOLITICS/02/02/elec04.poll.prez/
that was the first one i found by simply googling "kerry leads bush"


Tony - I think you missed one subtle point of mine.... don't make me have to do the work myself! ;-) But I've gotten so used to coming here for all the data + links - the information is so up-to-the-minute and comprehensive - that I've become spoiled.

But I do wonder how this would look plotted on a graph so one could see the degree to which the numbers shifted over time, and whether there is any resemblance between 2000 and 2004. I guess 1988 and 1992 would be interesting too.



I do not think it was necessary to do all that complicated stuff.
Why the hell simulate the election 1 Million times??
The random effect is down to less than 0,1% when you simulate that 5000 times, like I did, using not the SurveyUSA, but the


averages (made by Poblano). His results of about 59% for Obama and 41% for Clinton look way more realistic than this results.

The only thing it would be necessary for to do that is, that you want to get a result that has a deviation of 0,0001% or so. But then you cannot ignore the Nebraska and Maine voting system, that has a LOT more influence as the random effect in 10000 runs.

"About one percent of our simulated outcomes were Electoral College ties."
Yes, but the scale could have been tipped to one side or the other by a congressional district in Nebraska going to Obama or a Maine district going to McCain (unlikely).

An 88% or an 74% chance for Obama/Clinton is MUCH to high. Why? You "assume that the state estimates were correct except for sampling error."
That is nonsense. When they were, all the Rasmussen , Mason-Dixon, Quinnipac, local pollster surveys that show other results than the SUSA 50 state poll must be wrong. This is absurd.

This sounds good on the first sight, but when you think about the procedure, then there I see too often "we ignore" and "we assume".


Chris G:

this analysis is okay, definitely a step in the right direction, but assumption of simple random error is wrong. why not devote more resources to integrating uncertainty about turnout, etc?

also, totally agreed w/ Ciccinia about snapshots being irrelevant at this point. although we can't predict the future, we *can* forecast uncertainty by estimating how much opinions tend to change month-to-month (and historically how much have they in a general). the lower that number is, the less likely McCain is to overcome Obama's lead. it's a very tractable question, we don't just have to shrug our shoulders

eg, Donald Green and Alan Gerber at Yale did a great analysis of public opinion on a particular issue a while back (to time-series people reading this they used a Kalman filter i think), they seemed to do a good job getting at this issue.

in any case, it's really a misnomer to call it the probability of winning if it doesn't take time into account.



And I don�t see why you did that.
This is useless analysing, because it doesn�t analyse the results of the US Presidential Election but those of the SUSA Presidential Election, and for what is this useful?

You CAN simulate the General Election- when you include at least 3 random numbers that represent
a) poll errors
b) statewide movement until November
c) national movement �til November.

This polls are not accurate for the November election, they shouldn�t be. But the only reason I see to simulate the outcome of an election using polls for that election is, that you want to know how the election could end.

But this is the wrong way to do it.



I too agree that these polls are meaningless come November.

We can not underestimate the grass-roots effect of the Obama campaign.

Think about this: Obama's campaign has taken down a candidate with unreal name recognition and organization.

The good thing for him is that he has college kids on his side that can work tirelessly for months on end while McCain/Clinton has old grandmas that go to bed at 8 pm.

What has been happening this campaign is that not only has Clinton voters come out in force , Obama's voters are new and have come out in greater numbers. If, and this is a big if, the Democrats unite as they should, there is no chance for the Republicans come Fall. But the 64,000 dollar question is this: Will one woman's quest for power ruin the Democrats chances in November?



Nadie~ You may be correct. But no poll can measure the predictability of a new voter coming out to the polls if there is an individual or cause they are excited about. The African American pool is just about tapped out, so Obama may not be able to gain many more voters by intensifying his efforts in that arena. The youth vote, however, while higher than it has every been (up from about 15% to about 22%, pretty incredible for a Primary) still has a great pool that could be tapped. If Obama can actually get them out in numbers higher than any other demographic that will be amazing. I think that in a General Election, with an individual in which an even greater contrast can be made, and where the stakes are higher, that's a very real possibility.

But the "snapshot" by SURVEY USA, as flawed in many respects as it is, does show that Obama can win far more "Swing States" and comfortably wins the solid "Blues". Hillary also wins most of the time (but fewer Swing states and by fewer electoral votes).

What that means is that Obama benefits down ticket match-ups far more than Clinton. Hillary mainly assists candidates in solid Blue States who already have fairly assured vistories. Obama has the chance of campaigning and helping (and being helped) in Swing States. And if he actually does have a higher turnout in underrepresented demographics those voters will likely vote straight ticket.


Nadie, what you are bragging about when you say Obama "has taken down a candidate with unreal name recognition" is essentially a re-tread of the Bush campaign's negative message strategy against Gore in 2000.

I don't know if your memory goes back that far, but that's exactly the negative message strategy Obama has used.

Does this sound familiar? Hillary is too ambitious, soul-less, corrupt, associated with Bill Clinton, in the pocket of shady donors. She's a liar who would do anything to get elected. All these accomplishments she talks about? Exaggerations, or worse - fabrications. Besides, the country needs to put the Clinton years and all their scandals behind us, we can't have three terms of Clinton, its time to bring honor to the White House, we need a fresh face, a guy who hasn't been corrupted by years and years in the beltway. Sure, he doesn't have a lot of experience, but neither does she - being First Lady doesn't mean you have anything to do with important decisions. Most of the time she wasn't even in the room. What a liar!

Substitute "Al" for "Hillary" and "Vice President" for "First Lady." Substitute Governor Bush, man of the people from outside the Beltway, for Barack Obama, man of the people from outside the Beltway.

And there you have Karl Rove's message strategy from the 2000 election down to the dot and jiggle. That's how the GOP and a cadre of attack pundits (Dowd, Matthews, Rich, Limbaugh) destroyed Al Gore.

Effective? Yes. Something to be proud of? Obviously some people are.

And by the way, those grandmas and grandpas who go to bed at 8pm? They wake up in the morning early enough to go vote. Students? You're lucky if they wake up *before* 8pm and remember why Tuesday is supposed to be important (Obama Girl et al).

(Just kidding. The 18 to 25 year old demographic is well known for their dependability).



Ohhh okay Ciccina, First Lady is JUST like Vice-President!!!


"the old people wake up to vote" -

Apparently some have alzheimers and forgot to vote in 30 states. LOL!!!Ooops!



I still find it amusing that so many women are voting for Clinton simply because she is a woman - not even a strong woman at that! Does a strong woman cry when things get tough? Does a strong woman have Bill do her dirty work? Does a strong woman stay with a chronic adulterer? No, no, and NO!



THAT IS FRESH - Team Clinton claiming
dirty tactics! Please, what is next - saying Obama is too feminine?? What a joke!

At this point we know that:

1) Obama will end the contest with the most pledged delegates,

2) Obama will likely end the contest with the popular vote tally,

3) Obama will end the contest with the most money and greatest fundraising potential,

4) Obama will end the contest with the most states,

5) Obama will end the contest with the best poll numbers against McCain, and

6) Obama will end the contest with the most primary state victories and caucus state victories.

So what's left for Team Clinton? She has to convince a majority of the super delegates to cast their vote for her, so how does she get those supers to ignore all of the above Obama advantages in order to cast their ballot for the candidate who is losing?

Apparently, it's a two-pronged strategy.

The first is what we've been seeing this week -- tear down a candidate who has inspired and given hope to millions by appealing to white resentment and turning him into the "black candidate". It's ugly and revolting, but the Clinton campaign is banking on it scaring people away from Obama. And by "people", I mean "super delegates".

Remember, Clinton can't win based on the math. The rules -- the "process" -- are her enemy. The only way she can win is by having the super delegates ignore all of Obama's clear advantages -- a coup by super delegate. And the way that coup is by tearing Obama down and discrediting the process that gave Obama those advantages.

But here's the rub -- the "process that gave Obama those advantages" includes latte drinkers, and black people, and young people, and red state Democrats, and small state Democrats, and blue states that voted for Obama.

So it's a sort of Catch-22 -- she needs the super delegates to abandon the winner for her loser campaign, but the way she's trying to win them over is by insulting their very states and constituencies.

Clinton is in a bad place. She is behind in every metric that matters, and has been relegated to trashing our likely nominee and entire Democratic Party constituencies and states in order to make the case that she's somehow "more electable" despite all evidence to the contrary. Unfortunately for her, the super delegates aren't all cloistered in New York or in DC.

They represent the United States of America. And outside of Clinton's Blue bastions, her insults aren't winning any new converts.



Hmmmm, WRONG AGAIN Ciccina -

Gore was brought down by Clinton and his adultery - AND YOU KNOW IT!!

Rove and the dirty tactics played a part in beating McCain in 00 but mostly Kerry in 04.

So stop with the BS, allright?


Gloat while you can, KOS. Because you know what happens after Obama wins the nomination?

1. The GOP puts real muscle into a below the radar campaign to paint Obama as an extremist, using quotes and video of his lovely Pastor.

2. The GOP paints Obama as an effete, snobby intellectual who looks down on regular working class voters.

3. The GOP points out that Obama has no experience whatsoever with foreign policy or the military and failed to have a single hearing in his subcommittee (which covers relations with Europe, which means it covers relations with NATO and Afghanistan) because he was too busy promoting his own ambition.

4. The GOP paints Obama as too "soft" (yes, "feminine" to McCain's "manly man") to handle security issues. The poll numbers already show he is the weakest candidate on this issue. Suddenly you start hearing him called Bambi a LOT.

5. The GOP pulls out every African American who ever lived who said something unfortunate and connects him or her to Obama, and then asks him to reject and repudiate. Remember the "bimbo eruptions"? This will be the "radical eruptions." Instead of Lewinsky, it'll be Alinsky.

6. Suddenly there's a lot more talk about the evils of identity politics, liberal guilt, and political correctness run wild.

7. The GOP pulls out Rezko and the Iraqi businessman he's connected with and calls Obama underhanded and dirty. Exelon too.

8. Remember the moment in the second to last debate when Obama said he thinks all American schoolchildren should have to learn a second language? Hello, "Obama doesn't want English to be our official language, and wants to force our kids to learn Spanish." The Pat Buchanan crowd loses what little scraps of mind they have left.

9. Obama realizes its tough to run as "the guy who isn't Clinton" when there's no Clinton in the race. Somehow "she doesn't have any military experience either" sounds pretty weak. Voters ask him what he's accomplished and hear crickets.

and finally,

10. "Liberal" columnists start talking about how its "just too soon" for Obama, how its all Clinton's fault for weakening Obama, and isn't that too bad, damn her, but this just isn't the right time for him to be president. Its not like you'll be voting against him, they'll say - its just that you'll be voting FOR him later.

KOS, you people keep acting like the general is going to be some kind of cakewalk. The gloating is making you soft. But enjoy it while you can. One way or another, its not going to last.


John - Spokane, WA:

this preaching to the choir stuff is so amusing - its like being a spectator at an Obama backslapping contest. Speaking of Preaching.... Obama's admired Reverend Wright has made every racial slur in the book along with stating "God Damn America" and after the 911 attack he said "Well the chickens came home to roost" as well as a full endorsement of Louis Farrakahn and Obama looked upon him as (In his own words) "An old uncle" and "An inspiration to his book entitled The Audacity Of Hope". Yes, Mr. & Mrs. Obama supported the ideas and listened intently to the sermons of Rev Wright for more than TWO DECADES and they were FINE with it and accepted his ideas. NOW, Obama Rejects, Denounces, disclaims his relationship with Rev Wright. Its clear to see where Obama picked up some of his speaking abilities and now its Clear to look into his 20 year education by Rev Wright and determine what type of Mindset attracts Mr. Obama. This isnt Rhetoric - The proof is in the puddin folks and he's losing supporters in a steady stream over this and his other skeletons.


Survey USA is the only survey showing Clinton beating McCain in Ohio in the general, and they show 2 digit leads. Why is that? It casts suspicion on all their other work, in my mind.


Mark Lindeman:

Any survivor of TruthIsAll's Election Model has to shudder a bit at something like this. But of course it isn't intended as a prediction.

I think it's somewhat interesting for exploring the relationship between the popular vote and the electoral vote for these particular candidates. Actually, I wish they would take it a bit farther in that direction, by somehow setting Clinton's and Obama's popular vote shares equal... but damned if I can figure out what the results would mean if they did.

I like fivethirtyeight.com's approach, in principle, since it incorporates more information per state -- although that has its own pitfalls (house effects, differences in survey timing).



A few comments:

() An excellent analysis based on far better data than has been available so far. The competing simulations I've seen are based on aggregate, rather than individual, level data and, as such, are inherently problematic.

() The complaints that the analysis doesn't take into account the greater commitment of "young" and "new" voters to Obama are both purely hypothetical and, at least according to exit polls, wrong.

In fact, voters under 30 years of age have consistently averaged between 14% and 16% of voters in almost every primary. (The one exception for a primary was New Hampshire where they constituted 18% of the Democratic primary vote. And NH, in case anyone has forgotten was won by Clinton.)

In short, "young" voters are an overhyped phenomenon (at least so far) in the 2008 race. Many more have voted in 2008, but not disproportionately. They are still outnumbered significantly by voters 60 and over.

Frankly, as an Obama supporter I'm tired of hearing about the effect of "young people" when their effect is primarily on internet chat boards and not in voting booths.


just pixels:

Let's see: After 7 years of Bush, the economy is in the toilet, dollar in decline, gas prices at record high. The Dow is lower than 8 years ago. Iraq was a monumental blunder. Almost everyone (76%) wants a new direction.

How can the Democrats lose? Watch and see.



just pixels - the democrats are skillful losers.



Gore lost Tenn. because of Clinton? That is funny!! Gore did not even allow Clinton to campaign with him



Obama camoaign implodes with the WRIGHT and REZKO disclosures


John - Spokane, WA:

Obama, in all of his smooth techniques is trying desperately to downplay the expose' of his personal extremist Mentor, Rev Wright - This is NOT GOING AWAY NOr should it. Today, Obama said in Indiana that we have to move on - Too bad he coudlnt of taken his own advice 20 years ago when he instead chose to embrace this Hate Mongering-Anti American Racist Wright. Have ya noticed, Obama NEVER wears a USA Flag Lapel Pin ? I guess he never thought of that either. I thought it was interesting today that Obama put a bunch of White people behind him while he spoke at the podium instead of the usual AA's.




And your point is....?
Those attacks will work the same way - and dare I say - even better against Senator Clinton. After all, Ms Lewinsky is not alone. And who knows what the ClintonS have done since leaving office with not much money, to the point of loaning Senator Clinton's campaign $5 million (which is 3-4 times Senator Obama's net worth, by the way.) Uranium-mining deals in undemocratic countries, anyone?

Lifetime of experience - will Senator Clinton stand up to Senator McCain on that? War hero, favored guest at the Hanoi Hilton... against a "strong" woman who stood meekly by her philandering, lying husband. Yes, that's a very favorable comparison for Senator Clinton.

You may think all the Clinton scandals have come out - but there will be plenty of voters, particularly young ones, who haven't heard about them much, and there will be more scandals coming out. Enough to provide a few-tenths of a percent swing to Senator McCain, the (much as you might feel otherwise) honorable war hero, enough to allow a GOP victory this November. [And that's not even counting any Democratic voters who get disgusted with a Clinton candidacy - aided by the Party Establishment - in the event that Senator Obama ends up with more pledged delegates.]

Anyway, back to the guest pollsters. As they admit themselves, the data they use - the SurveyUSA 50-state poll - at best represents a snapshot of the electorate at this point in time. You can bitch and moan about their methodology all you want, but hey, it's one way to skin the cat. They have done it in a methodical, transparent manner, and it's a definite improvement on the SurveyUSA electoral college tallies.

Presumably, Professors Erikson and Sigman can update their model with future polling data. Should be interesting!


John - Spokane, WA:

@ RS,

Its always interesting to see that you try to defend Mr. Obama's affiliation with an Anti-American Racial Monger and compare it to anything related to the Clintons is a weak knee'd attempt to totally ignore & address the facts that Obama has Patronized, Admired, envied and been Inspired by this Anti-American who makes no bones abouting Hating his Country and the White people in it ! The fact that Obama Idlelized this man for more than two decades automatically disqualifies him to lead this Great Nation !



@John - Spokane, WA:

I am pretty sure that thus far I, personally, have not commented on the Rev. Wright controversy, either defending or attacking Senator Obama.

Unfortunately, that's the only part of your rambling post that I understood.



@John - Spokane, WA:

On second thoughts, the last line of my comment @7:32 PM was uncalled for. I apologize.


John - Spokane, WA:

@ RS,

Thank you, the purpose of my comments are soley to point out what I (and appearantly many others) beleive to be imparative to the Identification of Candidate Obama. I'll go on to say, I dont think any Candidate is without fault - but to measure the sincerity of these people I think must be an optimum goal. Especially when they present themselves as a representation of unity in the future - They must prove themself to be genuine and with the recent uncoverings, I think we as citizens of this Country must have absolute knowledge of what they stand for and they must be accountable for any issue that surrounds them like this one does Obama. I know Hillary Clinton has been questioned on her Business dealings AND shes been grilled over it - and if someone can bring something to the table that incriminates her than she should be held accountable for that. I have Never seen anything in her Campaign that even closely relates to an extended relationship that she has or has had with an Anti-American and racist such as Rev Wright. It makes some of these other things that we discounted earlier as just being rhetoric on Obama come more so to light now.


Here's a thought:

What's a "Big State"?

Let's go with the Top 10 by population:

New York
North Carolina

Of these, 2 are embroiled in controversy and two haven't voted for sure. Texas appears to be a tie, since Clinton won the primary and Obama the caucus (and Obama got more delegates overall).

That means that Clinton is ahead 3-2-1 in "Big States".

What's the "big" fuss about? Seems awfully close to me. Oh, and if you really want to include New Jersey, #11, you really should explain why Virginia, #12, shouldn't count, as it's only 2 EV away.



Frankly as an Obama supporter myself, I think you are misinterpreting the exit poll data you have concerning the youth vote. Here's what you said above:

The complaints that the analysis doesn't take into account the greater commitment of "young" and "new" voters to Obama are both purely hypothetical and, at least according to exit polls, wrong.

In fact, voters under 30 years of age have consistently averaged between 14% and 16% of voters in almost every primary. (The one exception for a primary was New Hampshire where they constituted 18% of the Democratic primary vote. And NH, in case anyone has forgotten was won by Clinton.)

In short, "young" voters are an overhyped phenomenon (at least so far) in the 2008 race. Many more have voted in 2008, but not disproportionately. They are still outnumbered significantly by voters 60 and over.

Frankly, as an Obama supporter I'm tired of hearing about the effect of "young people" when their effect is primarily on internet chat boards and not in voting booths

you are wrong on your analysis on many counts.
First, let me acknowledge as you said that the youth vote(17/18-29 years old) has been on average around 16%- though I haven't verify this assumption, my random sample told me you may be very close to the mark!

What you are missing is the high increase in turnout among young people this election cycle compared to other primaries. You can't just compared them to older voters who have consistently make up a higher share of the electorate but even doing that will make your point moot. Let's look at the relative increase in turnout between the two groups.
Let's look at some data

Youth vote % for Iowa

2000: 9%
2004: 17%
2008: 22%

youth vote % for Mississippi

2000: 7%
2004: 7%
2008: 14%

youth vote % for Ohio

2000: 8%
2004: 9%
2008: 16%

youth vote % for Texas

2000: 9%
2004: 10%
2008: 16%

Note how the youth vote has increased dramatically for every single state above. Over a 4 years peiod: In Iowa: an increase of 29%, in Mississippi: an increase of 100%, in Ohio: an increase of 77%, and in Texas: an increase of 60%

To make your argument looks more unconvincing by the hour, let's see how voters over 60+ have done over the same period and under the same circumstances.

voters 60 years or older for Iowa

2000: 39%
2004: 27%
2008: 22%

Voters 60 years or older for Mississippi

2000: 46%
2004: 31%
2008: 29%

voters 60 years or older for Ohio

2000: 30%
2004: 34%
2008: 23%

voters 60 years or older for Texas

2000: 34%
2004: 31%
2008: 22%

As everyone can see, the share of the 60+ has decreased considerably while the youth vote has spiked over the same period. More astounding, the youth vote in Iowa matched the turnout of the 60+ turnout- a surprising result by any standard.

Over a 4 years period: In Iowa: 60+ has decreased by 20%, decreased by 7% in Mississippi, decreased by 32% in Ohio, and decreased by 29% in Texas.

Conclusion: Before you come here yapping about the youth vote turnout this cycle, please consider to look closely at the facts.

Ps: I would gladly like to have a debate with you about this topic anytime/anywhere.

The people above have a point about the importance of the youth vote in the general election where a larger pool of voters will be in play. The youth will be critical for both Obama and the congressional races when a lot of races might be decided by just 1-2% points.


Gary Kilbride:

Erik, you can probably define the "Big States" as the ones with double digit electoral votes which are considered swing states in a general election. I would use the measure that they vote within 5 or 6 percent of the national average. Off the top for '08, that would isolate Florida, Ohio, Pennsylvania, Michigan, Missouri, Virginia, Wisconsin, Minnesota and Washington. The later two are blue states, but not overwhelmingly. States like North Carolina are out of reach for a Democrat unless the popular vote is a comparative landslide, meaning they are pile-on irrelevant.

Of course, in '08 the big states can also include foundational building blocks, lesser populated states like Colorado, Iowa and Arkansas. If Democrats try to forge a map minus Ohio and Florida it requires a near sweep of the mid level swing states. Extremely flimsy footwork, IMO.

In KOS's long comment, I noticed he tried to slip in something that isn't absolute, among the know-knows. No one called him on it so I will:

"5) Obama will end the contest with the best poll numbers against McCain"

First of all, that's hardly certain. The margins are very close, and I found it amusing he posted an assertion like that in a numerical-based thread with Obama projected to 51.5% vs. McCain, and Hillary at 50.7%. Wow. A modern day version of Landslide Lyndon.

Besides, "ending the contest" is irrelevant toward November, in terms of poll numbers. Every other point KOS made focused on the final primary as conclusion. Valid. But it's perfectly reasonable for analysts, and yes super delegates, to eye November in terms of where polarizing Hillary might slot against McCain, as opposed to largely untested yo yo Obama. Yo yo as in either way in a general election vs. GOP lobs. Higher upside and downside than Hillary.

Sorry, but this campaign has been comparative WWE. And '04 in Illinois wasn't a test. You had to be anti-war to win that Democratic primary, given the climate. That's one of the unspoken truths about '08. Obama is receiving credit for a position he had to take. Kristen Breitweiser has detailed Obama's absence during the decisive moments of debate. And Alan Keyes was obviously a farce.

There was a logic to super delegates in the first place. And we can't pretend primary voters at the outskirts of each party who dictate the nominee don't get it decisively wrong on occasion, in terms of electability. Fellow Nevadans and I have suffered through that for more than a year, after Dina Titus trounced Jim Gibson in the Democratic primary, despite polls and demographic common sense screaming Gibson had more of a chance than Titus vs. GOP scandal-plagued goof Jim Gibbons.

It's interesting that Obama backers front and center math in terms of delegate relationship to Hillary, but have no problem ignoring electoral math when it comes to reaching 270 electoral votes amidst apparent demographic vulnerability in Ohio and Florida, the two most critical swing states. Obama has run a terrific campaign but it can't be understated how fortunate he has been. Florida will dismiss him regardless of primary date or method. Dodging that defeat with an excuse earned a benefit of a doubt where it does not belong. Just imagine the state of debate right now if lopsided losses in Florida and Ohio were front and center on his resume. I'm admittedly nervous as hell trying to log 270 electoral votes with apparent weakness in those two states. They are much closer to our camp, from a historical perspective, than places like Colorado or Virginia.

Also, I note the Obama backers seem to be frantically defensive, an ongoing system shock, now that the media tilt is absent. They coddled that as a rightful norm, and even denied its existence. Like reading a golf putt as dead straight, when the real world sees a 10 foot break.

I think Hillary would make the superior president. Weird rationale, I concede. I'm hardly as paranoid about the Iraq vote as fixing the economy, and it's anything but a push button certainty any Democrat can succeed there. Give me the '90s team with a demonstrated record. Lousy handicappers prefer the sudden risers, like the college football player who was nowhere on the charts entering his senior year, then fooled the masses into top pick status. Those guys are inevitably less than they appear.



there are a LOT of ways how Obama can reach 270 EV without winning Florida and Ohio.
Here is one of them:

NOT Delaware
NOT Florida
New Hampshire
NEw Jersey
New Mexico
New York
NOT Ohio
Rhode Island
--> 281 EV, if I am right.
Or take Pennsylvania away and give him Delaware, Missouri, the Dakotas, Virginia
--->291 EV


Gary Kilbride:

My comment last night extended to a marathon so I sliced a major portion, but let me re-instate it now.

When I defined the major states as double digits electorally and relatively close on the partisan index chart, it wasn't lost that the measure favored Obama in terms of primary outcomes. I mentioned that in a comment that was spit out the other night, in response to Mark Mellman's analysis of primary results in relation to general election.

No one can rightfully assert Hillary is demonstrably more electable. But IMO it's equally foolish and thin to insist otherwise, like the KOS claim here and the poblano diary which has been making the rounds, heralded on pro-Obama sites, embracing laughable variance between Hillary and Obama vs. McCain in specific states. There is no way two Democrats will fare that differently against the same Republican. I've mentioned several times that the partisan index will feature more deviation than we're accustomed to, due to factors like first black or first woman. But that equates to a handful of points, not the massive gap implied by early and sporadic state polling, on theoretical matchups. It will be a natural tightening, not unlike Obama making up ground in states he supposedly trailed Hillary by 20 or 25.

Frankly, the netroots gets too full of itself and fails to recognize regulators of a changing playing field. It's much simpler to make up open lengths in a primary with everyone more or less thinking the same way, as opposed to a general election with 40-45% locked against you. Outcomes like Lamont overtaking Lieberman in the primary and Obama running down Hillary lead to false and fanciful projections going forward, a stupid belief that your guy is superman and can flick aside long term partisan realities. I remember late '06 claims that Lamont's general election poll deficit was a mirage, because his ground game would make up at least 3 points. On site after site I was attacked for dismissing that theory as ignorant desperation. I supported Lamont but I bet on Lieberman. Similarly, while Obama's great youthful organization has dominated caucus after caucus, that's remarkably irrelevant come November. You win general elections via preference, not turnout. We're not going to overwhelm the GOP via registration drives, the late folly of '04. Obama may be our best ticket but in a pro-Democratic year I'd be more comfortable trying to win a squeaker with a known quantity like Hillary.

Studies like the one in this thread, an accumulation pointing to national popular vote estimation, are more valid than accepting state by state projections. The interesting aspect is something Mark Lindeman asked about, and I've mentioned on a few sites the past few weeks. Is even really even, in terms of Obama vs. McCain and Hillary vs. McCain? What happens in the electoral college if Hillary and Obama are estimated at the same popular vote level? Obama's apparent vulnerability in Florida and Ohio led me to speculate Obama needs to lead McCain by greater margin than Hillary, merely to draw even in electoral likelihood. I got carried away and estimated 2-3 points on a couple of sites. That was partisan nonsense, brutal exaggeration. When I finally ran it through my Excel model it balanced at .4 to .5. Admittedly, I'm a sports guy who dabbles in political analysis so while my sports models can hold up to the ones I've seen on the internet and elsewhere, the political version even after 12 years is somewhat basic. But I'm confident it defeats subjectivity.

Otherwise, this is what I don't get about Obama supporters, whining about recent turn of events. If Obama is truly a phenom who inspires millions, backed by political foundation, he'll be immune to surface charges and supposedly illegitimate scrutiny. Otherwise, he probably isn't the candidate you think he is, nor become the caliber of president you expect and Democrats need. The last thing we can afford is a weak presidency a la Jimmy Carter. I'm old enough to remember mocking condemnation of Democratic politics in general, based on Carter's reputation post '80. I've seen younger analysts on progressive sites assail Democratic strategy and performance in the '80s and '90s, pathetically unaware how far Carter set us back. In '88 specifically, while I wanted to accept the early polling allowing Dukakis the lead, my base instinct kept saying, "the country is content and it's too soon after Jimmy Carter."

As much as I want to believe in Obama, and will support him since unquestionably he's a huge favorite to be the nominee, when I watch old clips I saved from '92 and '96, Obama is not Bill Clinton. It's startling, side by side. Bill Clinton could combine the courage to change hoopla with devastating specific breakdown of GOP and Democratic priorities and the real world consequences. Obama is a glorified pep rally.




Thanks for the analysis of the proportionate Youth vote increase. I'd say that what we are seeing is a massive increase in the youth vote numbers that may be far above what you are suggesting. This is especially so since the OVERALL turnouts for the Democrats are so large. What may be occurring is that Over-60 voters are coming out at their usual (or slightly higher than usual) numbers...but the Younger demographic voters are turning out in NUMBERS that are not just double, but triple or quadruple the numbers that they have before.

In addition the SURVEY USA poll did not incorporate in differential PARTY enthusiasm.
Again, granted that their poll is just a "snapshot" of current "likely voter" attitudes. As such it cannot easily incorporate the differentials in turnout support b/w McCain and either Obama/Clinton in the General Election unless in future polls they weigh some "voted in the primary" factor into their counts. But given that Electoral Turnouts have been about 2:1 Democrat vs. Republican in almost all the States (even strong "Red States") that could play a factor in Red States.

In Blue States, as I've pointed out, this will simply reinforce a Democratic candidate win (whether Clinton or Obama). But it will be critical in these battleground and potential cross-over States...especially in obtaining the Congressional seats needed to govern effectively.


John - Spokane, WA:

@ Gary,

I agree with the jest of your thought, that is Hillary is in the Better position to serve the Country. Before you can even consider Obama's mathmatical abilities at this point - They must first overcome all of these Anti-American questions that have now come into to play. As much as Obama supporters would love to just see this go away - ITS NOT. The first and formost qualification of your President must be UNCONDITIONAL AND DEDICATION TO YOUR COUNTRY AND THE FLAG OF THE U.S. He has too closely involved himself with an organization (Calling itself a Church) that has "God Damned" our Nation and made sympathetice remarks about those who have attacked our Nation over the years and Blames it all on White People ! There is NO WAY that this man (WHO PATRONIZED THIS CONGREGATION) for 20 years and welcomed Rev Wright into his Family as if he were an adopted Uncle and inspiration in his life - NO WAY HE CAN WIN THE NOMINATION NOW let alone the GE.



This SUSA megapoll is irrelevant now that Obama has now fallen behind (or reduced his advantage over) McCain in several polls.

The poll was taken in late February. It's mid March.


Post a comment

Please be patient while your comment posts - sometimes it takes a minute or two. To check your comment, please wait 60 seconds and click your browser's refresh button. Note that comments with three or more hyperlinks will be held for approval.