Articles and Analysis


SurveyUSA's 50 States with Sampling Error

Topics: 2008 , Barack Obama , Gallup , Hillary Clinton , IVR , IVR Polls , John McCain , Pollster , Rasmussen , SurveyUSA

There has been a considerable buzz over the last two days about the surveys released yesterday by SurveyUSA that test both McCain-Obama and McCain-Clinton trial-heat questions in all 50-states. Putting aside the concerns some have about SurveyUSA's automated methodology and the other usual caveats about horse race polling at this stage in the campaign, I tend to agree with the critique from Matt Yglesias (via Sullivan):

Each of these polls has a sample size of 600, so the margin of error will come into play. What's more, there are 100 separate polls being aggregate here, so the odds are that several of these are just bad samples.

True on both counts. SurveyUSA colors in states on their maps even if a candidate leads by a point or two, margins that are not close to achieving statistical significance. However, since SurveyUSA says they did 600 interviews in each state, we can take their analysis a step further, applying statistical sampling error to the candidates' margins in each state.

Professor Franklin and I have done just that, classifying each state based on the statistical significance of the candidate's lead. We call a state "strong" for the candidate if they lead by a margin that is statistically significant at a 95% level of confidence, the level typically used to calculate the "margin of error" attached to most surveys. We label as "lean" any state where a candidate leads by more than one standard deviation, which amounts to a 68% confidence level. We label all other states as toss-ups.

Note also that these significance tests assume "simple random sampling," which produces a smaller error margin than we would get if we could take into account that SurveyUSA, like virtually all pollsters, weights its data. We would need access to the raw data and weights in order to do truly correct significance testing.

The tables and maps appear below, followed by some discussion. First, here are the results and a map showing an Obama vs. McCain match-up (you can click on any of the images for a larger size version):


And here are the results and a map showing an Clinton vs. McCain match-up:

If you would prefer, you can also download the spreadsheet that we used to create the tables.

Now that you have all of the data before you, let's consider the merits of the project and a few caveats about the data. First, this sort of project -- which involved 30,000 interviews completed in 50 states over a three-day period (February 26-28) -- would not have been feasible with live interviewers.

On the other hand, the automated methodology is controversial with traditional survey researchers. I wrote about the arguments for and against IVR (interactive voice response) surveys Public Opinion Quarterly, and I have blogged often on the subject often, both here at Pollster and on its forerunner MysteryPollster. Readers are obviously welcome to share their opinions about the IVR methodology in the comments.

The other caveats noted by SurveyUSA are worth repeating: They surveyed all self-reported registered voters, and did not attempt to screen for "likely voters" (although many national pollsters do the same at this stage, feeling that we are too early in the process to attempt to predict what voters will actually cast ballots). McCain would likely do slightly better in both match-ups under a "likely voter" screen. Also, we are obviously still eight months from the election. Much can and will change in terms of voter perceptions and preferences.

Let us also keep in mind the limitations of random sampling error. It tells us only about the variability that comes from calling a sample of households rather than dialing every working phone number in every state. As with any survey, it tells us nothing about the potential for error based on the wording of the questions, the selection of respondents within the household and the voters missed because they lack land-line phones or do not participate in the survey. Be careful about using the misnomer "statistical tie" to describe states in the toss-up category. One candidate would likely show a "significant" lead if we could increase the sample size -- we just lack the statistical power to know which candidate that would be.

Finally, keep in mind that since we are looking at 100 tests (2 each in 50 states), these results probably misclassify five states by chance alone (as opposed to the way we would classify them if SurveyUSA had called every working telephone in the 50 states).

With all the caveats out of the way, what does all this data tell us? Consider this summary of the electoral vote totals**:

03-07 electoral vote summary.png

These data are less useful in forecasting the ultimate result than they are in gauging the relative strength of both Clinton and Obama as of last week (February 26 to February 28). Those dates are important, since both the Gallup Daily and Rasmussen Reports automated tracking have shown Clinton gaining ground on Obama nationally over the last week.

Nonetheless, as of last week, Hillary Clinton led in states that add up to a slightly greater electoral vote total counting the leaners (250 for Clinton vs. 244 for Obama. Still Obama appeared to put more states into play (138 pure toss-up for an Obama-McCain race vs. a Clinton-McCain race). So Obama's initial electoral vote advantage is greater.

The most interesting aspect of these surveys is the states that explain those differences. Let's consider first the states where Obama does better than Clinton:

  • Obama moves three states from lean McCain to strong Obama: Colorado, Iowa and Oregon
  • Obama moves two states from strong McCain to lean Obama: Nevada and North Carolina
  • Obama leads in two states that are toss-ups in a Clinton-McCain race: New Mexico (lean) and Washington (strong)
  • Obama moves four states from strong McCain (against Clinton) to toss-up: Nebraska, New Hampshire, North Carolina and Virginia

On the other hand, Clinton does better than Obama in a smaller number of states:

  • Clinton moves one state from strong McCain to strong Clinton: Arkansas
  • Clinton moves one state from strong McCain to lean Clinton: West Virginia
  • Clinton leads in the two states that are toss-ups in an Obama-McCain race: Florida (strong) and New Jersey (lean)
  • Clinton moves one state from strong McCain to undecided: Tennessee
  • Clinton moves one state from lean McCain to undecided: Pennsylvania

Here is another table that makes it easier to see these comparisons (again, click on the image to see a full size version):

So, Pollster readers, what do you think?

**And yes, after putting these tables together I see that SurveyUSA split the Nebraska electoral votes based on on the vote totals, something I did not do.

Update: Nick Beaudrot (via Yglesias) creates thematic maps based on the same data keyed to the size of the candidate margin.



The title of your Clinton Vs McCain chart reads as "Obama Vs McCain".May be you should correct it to avoid confusion ?



Mark B:
Awesome work as usual!
More later, but pointing out a typo:
"Obama moves two states from strong McCain to lean Obama: Nevada and North Carolina"
- That should be Nevada and North Dakota.


Bob Evans:

This shows that Billary cannot win even if she cries all day and night before the election! Better ask for a refund Billary suckers!




Interesting question Mark. On one hand numbers look good for Obama because he seems to have more so called "solid" states in his column. But in I think it's important to note that many of the tossup states in Obama's side are unlikely to vote Democrat. On Hillary's side, some states that show tossup or lean McCain were actually the same states that were leaning Bush in pre-election polls in 2004, but ended up solid Democrat, Pennsylvania, Oregon, Michigan and Washington. It's a difficult call to make, but from my perspective, Hillary looks to have more states on her side that can actually deliver.



What do I think?

Besides clearly showing that Obama is a better general election candidate, it also shows that Clinton's argument that she should get the nomination because she wins "big states" vs Obama is bunk.


John London:

Hillary getting the Dems nomination will destroy the inspired youth vote and Prospects of racial equality or in more blunt terms piss-off the youth and the AA vote. Great way to start a run for president.


Andrew S. in California:

As objectively as I can say this I think this shows that all of us democrats need to stop talking about electability since both candidates clearly are very equal in their electability vs. McCain.


Fantastic, Mark! I love spreadsheets like these. Its sick, I know, but true.

I think the SurveyUSA 50 state chart is a cool toy that will become increasingly useful as the campaign progresses, but for all the reasons you state, and at 8 months from the finish line, past performance is a much better predictor. IMHO.

So I'm going to re-post a comment I posted to a previous thread because I think its more applicable here. Please pardon the redundancy.

Isn't the simplest way to think about November this:

Based on the dynamics of this election, which are more favorable to Democrats than in 2004, it's reasonable to expect that whoever the nominee is will win the Kerry states, except perhaps NH (4 EVs). (I don't believe for one second that Oregon would go red in November. Nor do I believe ND would go blue. Nope. Not this cycle.)

I don't mean to be insulting to Kerry people, but if he could win it, any Dem can win it this time around given the improved environment. Kerry didn't even carry New Mexico and Iowa, which Gore did. And neither won Ohio. Have there been any demographic shifts in the last four years big enough to flip a Kerry state red? Would a Clinton candidacy drive away the unaffiliated / swing belligerant white male vote any more than the Gore candidacy did? Gore was (wrongly) tagged with all the Clinton "scandals" and was criticized in much the same way as Hillary (cold, calculating, corrupt etc). Would Hillary really perform that much worse than Gore?

Kerry won 252 EVs, 18 short of 270. Back out New Hampshire, which I think was kind of flukey, and the deficit is 22 EVs.

So the nominee would need to flip the steady blues plus

(1) Florida - 27 EVs, OR

(2) Ohio + NM for 20 + 5 EVs, OR

(3) neither Ohio nor Florida, but a combo like CO (9) + VA (13) for 22 EVs, OR

(4) some constellation of CO (9), VA (13), NM (5), AR (6), NV (5), NH (4), IA (7) or some other state that I missed.

But certainly the simplest scenario is to win the Kerry states except NH, but pick up Florida, or win the Kerry states except NH, but pickup Ohio and any one small state.

Related to the SurveyUSA numbers showing Obama possibly flipping Virginia or another southern red state, the question I have is this - I assume that overall voter turnout will be higher for the general than the combined turnout of both primaries. But since Af-Am voters are overwhelmingly Democratic, and given the enthusiasm, I wonder if that segment performed near full capacity in the Dem primary.

I know that Obama the nominee would bring even more enthusiasm, but is it feasible to drive up the Af Am turnout by the same percentage that the non-Af Am vote will increase (especially that belligerant white male vote that the Dems keep losing)? Can the Af-Am vote realistically get high enough to change the playing field in a state like Virginia?

I do wish people would spare everyone the useless anecdotes like "everyone I know hates Hillary." "Everyone you know" probably loved Bradley and hated Gore, but Gore still pulled basically the same scenario as Kerry).



This fails to take the most important thing into consideration: If Hillary gets the nomination, she will ask Obama to be her running mate. She's as good as said so. And given his age and aspirations, he'd surely take it. And their "dream ticket" will bring everyone who supports Obama to their side. But if Obama gets the nomination, he likely either won't ask her to be his running mate (unless forced to by the DNC) and even if he does, she's very unlikely to accept it. (She's a 60 year old senior senator and former first lady; that would be a demotion for her). So that would turn many Hillary supporters who don't think Obama has enough experience (like me and most of my friends) away from Obama's ticket and they'd vote for McCain (because he's a very moderate "not scary" Republican). So in that sense, she's the much stronger candidate.


Wow, talk about redundancy. Sorry for the multiple posts, people.


Andrew S. in California:

You know Patrick that Obama could never accept that nomination to VP to Hillary. He's run on a ticket of not doing politics of usual in Washington. People have voted for him for this reason. I don't know if they'd vote against him if he broke his word or not but it could be a possibility.



I think things have changed in the last week and that one potential nominee is much more vulnerable to attack and has a weaker base and that in NOV these numbers will be very different.



John London,

I voted for Obama and still think he would be the better candidate, and possibly the better President. However, Clinton clearly is supported by more "traditional" Democrats than Obama is. Women, particularly those over 60, would be as angry about Clinton losing as the African Americans would be if he won. Polls have shown much greater support for her among lower-income men, gays, and Hispanics as well. In a national Democratic primary (no Independents or Republicans) she would be considerably ahead.


richard pollara:

These are interesting but given the volatility of the race in the last 10 days not predictive. The two national polls which bookend The Survey USA poll (ABC/Washington Post and Rassmussen) show a thirteen point swing from Obama to Clinton shortly after these numbers were compiled. More to the point are the Ohio exit polls which continue to show weakness for Obama amongst white males, Catholics, those making under 50,000, those with only a high school degree, older voters and women. It is a gaping hole in the Obama coaliton which has brilliantly brought together high income liberal white voters with African Americans. Unless Obama can find a way to attract those Reagan Democrats as well as Latinos I believe he has an uphill fight in the general election.


From A Canadian observer point of view, it's interesting to notice that none of the comments include anything about how the rest of the Globe will view this future "NEW" leader of the free world.
Also it looks to me like there is no doubt which of the candidates is the "Leader" that has caused all the excitement and set "The Tone" of this race. (HRC would make a good "Manager")
By the way Obama lead his 25 states by over 11.3%, Hillary averages +8.1% and the committed delegate count, as listed by the two camps themselves shows an insurmountable spread of 184.5 and 188 resp.



Whoops. Sorry for the double-post. And now I won't try to re-post this, even if I get a code error (can't remember exactly what it was).



To pre-empt flames - sorry if that last statement of mine ("Obama essential to Clinton, but Clinton not so much to Obama") came off as stand-offish or arrogant. That was purely the result of an intellectual exercise.
FWIW, I think the fairest resolution would be a joint ticket - I'd just prefer the person with the most pledged delegates to be the Presidential nominee, who ever that is.


Brad K:

Hillary, her staff, and some of her supporters here are stuck in the same old dead-end thinking that doomed Gore and Kerry: write off 45% of the electoral vote as unwinnable, concentrate on a dozen or so "big states" of which you need to win virtually every one with little or no margin for error, never make the Republicans defend their own goal, and lose the electoral vote by a whisker when the Republicans concentrate their resources on your weakest links and pick off one or two of the states you need. It's a disaster.

That's precisely why the Democrats elected Howard Dean DNC chair, because Dean saw the folly of this strategy---which by the way also writes off a ton of winnable congressional districts, governorships, and state legislative seats that control future redistricting, and dooms the Democrats to minority party status.

So far Dean's 50-state strategy is working pretty well. It played a huge role in helping the Democrats win back the House and Senate, and they're doing pretty well in governorships and state legislatures, too. But listen carefully to Hillary's "big state" argument for her electability, echoed by some of the posts here. It's a call to return to Terry McAuliffe's failed 50% + 1, no margin for error "big state" approach to presidential campaigning. Democrats should have the wind at their backs this time, but it's still a terribly risky strategy.

Obama won't have a 50-state strategy for the general ection, of course, but he's demonstrated at the ballot box that by bringing in independents and moderate Republicans and expanding the electorate by exciting young people in record numbers and energizing the African American base, he can compete in places like Colorado and Virginia that Gore and Kerry wrote off, and that Clinton is still writing off---and teeing off, I might add, with her campaign's constant patronizing dismissals of their electoral significance. These early Survey USA results confirm that pattern holds in the general election, and potentially extends to places like North Carolina and Nebraska, and even (gulp!) Texas. Yes, it's early and the dynamic of the race can change, and the particulars of the poll numbers certainly will change. But these results, as you've broken them down, show at this point a clear electability advantage for Obama, a less polarizing and more broadly appealing figure who dramatically broadens the electoral playing field. That goes away only if you close your eyes, bury your head in the sand, and convince yourself that Obama's appeal must be illusory because you don't want to believe it, based on your loyalties to Clinton or the past performance of candidates who did not share Obama's breadth of support. There's one campaign that's deluding itself here. And it's not Obama's.



Wouldn't a joint ticket just neutralize many of Obama's strengths? Speaking anecdotally, the Republican friends I have, many of whom are actually supporting Obama this year, have an "anybody but Hillary" mentality and would vote McCain if she was anywhere on the ticket. I'd love an Obama/Clinton ticket, but he already has my vote.



I keep looking at these maps and my mind is blown by the fact that Texas is a toss up state. That's impressive.



I surprised that people are trying to massage the data with their "instinctive" impressions. It's clear that something antagonistic to Clinton occurs in the Pacific Northwest as both Washington and Oregon demonstrate the same pattern, and almost by the same percentages. Similarly a pattern emerges in the Rocky Mountains and Great Plains. Obama does significantly better along the Mid-Atlantic States.

These numbers don't include issues like turnout potential...if Clinton draws out more Republicans antagonistic to her, or Obama enthuses his base greater than McCain things obviously could change. But as a "base-line" the numbers are interesting and do show that Obama can win "bigger" in so many more ways. He only needs to win 36 of 138 delegates in play. Clinton needs to win 20 of the 65.

Now why is this important. Isn't a win a win? That relates to the impact of the candidate helping win down-ticket Senate and House seats, and perhaps governorships. Because Clinton's strengths are concentrated in already Blue States it's unlikely that she will have much impact on the down ticket races. Obama picks up far more Red States and puts far more into play. That necessarily means more Senators and House seats that can be won.

Any Democratic President will need to win a substantial number of Congressional seats in order to establish a new direction.

If Survey USA does a similar poll in the next month that shows similar effects I think it will something that needs to be taken seriously. If I were a Democratic Superdelegate I'd be already deciding which nominee would benefit the PARTY the most this November.



A winning electoral strategy is to try and make lots of people vote for you. Then you can win the election by 5 points and the electoral votes will take care of themselves.


richard pollara:

The difference between polls and elections is the same as that of the preseason and the regular season in baseball or football. You might be hitting .350 in spring training, but if you can't hit a major league curve on opening day you are not going to make it. In the regular season, Obama has not proven that he can hit a curve ball. In all of the major primaries (Wisconsin being the exception) he has had no luck cracking the core Democratic constituencies. The curve ball in this election are the downscale ethnic white voters and Latinos. Without the Reagan Democrats and Hispanics a Democrat cannot win in the general election. Obama's pledged delegate lead rests on two things: Upscale whites in small red states (he holds a 100 delegate margin in 10 states whose total population is roughly equal to Pa.) and AA vote in southern states. In South Carolina for example nearly 75% of his total vote came from AA. Despite all the hoopala about a broad coalition to date he has failed to attract those voters who will determine the election in November. I cast my very first vote for George McGovern in 1972. I still think he was one of the finest people the party has ever put forth. Even on election eve I thought we were going to win. But I learned that a candidacy that can not appeal to the middle is a recipe for electoral disaster. Hope is a great thing, but some times you have to get real.



Obama partisans are an emabrassment! Instead of a cool headed analysis of the data presented, people go about the whole crap of 50 state strategy and what not!. Hill is actually a better candiate than Obama because she is more likely to win Michigan (17), Pen (21) and wash (11) than McCain. These wins will take her to 299. Obama on the other hand is likely to win Mich (17), NJ (15) and NH (4) taking his tally to 280.

Since both candidates are likely to win and Obama is likely to be leading the nomination in Nov, the best way out for the Dems is for Obama to be the nominee for this cycle and him supporting Clinton as the nominee in 2012.





Rasmussen is historically better than SUSA in GE polls.



Its amazing how the focus is always on OBAMA having to carry whites, the traditional reagan democrats, Clinton also must carry blacks... if all the black people in Illinois, Texas, Maryland,Ohio stayed at home in protest she may lose. She has the Hispanics, he has the blacks big deal. So this argument goes both ways. I guess I'm tired of seeing the innuendo on these postings that its a presumption that another group of core democrats the AA's will just fall in line... they may not. Im not saying they will vote McCain, that would put a lot of people into cardiac arrest but they may boycott



This is the WORST post I�ve ever seen here except those strange post that Clinton will win PA 80-20 because of God�s support.

Hill is actually a better candiate than Obama because she is more likely to win Michigan (17), Pen (21) and wash (11) than McCain.
Obama and her are doing same against McCain in MI and Obama is doing BETTER in WA state.

These wins will take her to 299. Obama on the other hand is likely to win Mich (17), NJ (15) and NH (4) taking his tally to 280.

He is likely to win NJ? Maybe more than Clinton?
This is a bad joke, it seems he could LOSE New Jersey.

First I thought you changed the names, but it wouldn�t make more sense then.

Since both candidates are likely to win and Obama is likely to be leading the nomination in Nov
So when you think Obama will be the next POTUS, why should he drop out then in 2012 and support Clinton?


Well, to get back on the topic, it is interesting that in the states that voted yet,
the primary winner is doing better against McCain 27 times, the opposite is true 10 times.

I wrote a delegate counter similar to that of CNN (that counter does not work for me)

and I ran 4 different models (with statewide delegate allocation, would be too complicated to do this for all districts that are still to vote in all states):

1) My predictions:

WY: 40-60
MS: 44-56
PN: 58-42
Guam: 67-33
Indiana: 47-53
NC: 46-54
WV: 64-36
KY: 63-37
OR: 42-58
Puerto Rico: 69-31
Montana: 45-55
South Dakota. 39-61
Superdelegates: 39-61

I don�t know why, but with that I get the following results:
Obama 2025 (!!!)
Clinton 1899

There is no error in my calculations, is that because of the states that lost 1/2 of their delegates or what?

2) For the SUSA�s 50 states- poll seemed to get along very well with the primary results, I took the margin of the Obama-McCain and Clinton-McCain margin *2 and took that number for the margin of victory in the primary:
That worked surprisingly good, for one exception: Clinton would win Mississippi.

When she does, my model is perfect ...

WY: 36-54
MS: 54-46
PN: 56-44
Guam: 67-33
Indiana: 42-58
NC: 44-56
WV: 72-28
KY: 62-38
OR: 37-63
Puerto Rico: 69-31
Montana: 38-62
South Dakota. 42-58

She�d net out 20 PD�s from all the primaries here, so with the superdelegates 50-50 it would be:
Obama 2003
Clinton: 1921

When the superdelegates split 60-40 for Obama, what is more realistic, I think, we get:

Obama 2030
Clinton 1894

Just 5 delegates more for Obama than in my prediction, this system seems to work!

I have to go, I will calculate best cases Obama and Clinton later today.



This survey is completely pointless.

At this point with emotions running high in the race and with things not decided, people are so likely to say "I'll vote for X if he is the nominee and otherwise vote for McCain". I want to see in November how many Obama supporters or Clinton supporters follow up on the threats.

I do not believe that the identity of the candidate is enough to so significantly swing states like that, especially in states that typically go Republican anyway.

Also, I don't see in the tables (maybe in the spreadsheet, haven't looked at it yet) how many people are going to vote in each case and how many will abstain.

Anyway, this is a strong weapon for the Obama camp.



@illinoisindie: AAs traditionally vote for the democratic nominee (Kerry, Gore and Clinton were all fairly white and got the majority of AA vote. We lost FL because some AAs were not counted, etc.).

It's only because this race became an "AA vs. the oppressing white power" thing that AAs might get insulted and vote for McCain, in which case they deserve another 8 years of Republican economy with no healthcare.

The funny thing is that Hillary is considered "old establishment", considering that she is a woman, a (majority) that is even more oppressed than AAs. Just look at the way people talk about her here. Would Bill get the same treatment if he was running after she was president rather than vice versa? If Black women weren't playing the race card first, they'll be flocking to Hillary and for real economic solutions.

And just a clarification, I do not mean this post to be racist in any way. The fact is that this election is about race. You don't get 80%+ of your vote from your racial minority if race doesn't play a role, no matter how good you are.



State Clinton% Obama%

Wyoming (12) 30 70
Mississippi (33) 35 65
Pennsylvania (158) 53 47
Guam (4) 45 55
Indiana (72) 42 58
North Carolina (115) 44 56
West Virginia (28) 56 44
Kentucky (51) 58 42
Oregon (52) 37 63
Puerto Rico (55) 60 40
Montana (16) 38 62
South Dakota (15) 42 58
Superdelegates 275* 33 67

* on 8th March

Barack Obama 2074
Hillary Clinton 1850
[best case scenario Obama I see]

State Clinton% Obama%

Wyoming (12) 42 58
Mississippi (33) 43 57
Pennsylvania (158) 62 38
Guam (4) 70 30
Indiana (72) 53 47
North Carolina (115) 52 48
West Virginia (28) 72 28
Kentucky (51) 63 37
Oregon (52) 43 57
Puerto Rico (55) 68 32
Montana (16) 43 57
South Dakota (15) 42 58
Superdelegates 275* 66 34

* on 8th March

Barack Obama 1931
Hillary Clinton 1993

[best case scenario Clinton that is POSSIBLE, but I don´t see PA breaking for her 62-38, neither the mid-west-caucuses just with a 10 point margin for Obama. But it´s still possible. Note that even with that you she wouldn´t come close to 2024/2025 delegates]



"Still Obama appeared to put more states into play (138 pure toss-up for an Obama-McCain race vs. a Clinton-McCain race)."

This is key. This is a huge point that both sides need to understand...

With a massive advantage in the money, having more toss-ups will help Obama a lot. He will force the GOP to spend money defending states they normally count as easy wins. And they simply don't have the money.


john grillo:

Clinton will not win NC...plus a 2 to 1 superdelegate victory for Clinton after Obama wins the majority of votes, states and pledged delegates...while I admire your imagination, this is nothing more than wishful thinking.

Superdelegates at best will split even Clinton/Obama...that is the best she can wish for. The Math isn't there and a scorched-earth strategy by Clinton only increasing McCain's chances of winning.



Hmm... 50 separate estimates? Seems like you should probably do a Bonferroni adjustment in order to account for the multiple inference problem. Right?


Chris G:

Mark- it's great to see someone in your position acknowledging that MOE understated because of the assumptions going into the poll sampling. unfortunately, if a likely-voter screen were used one expects MOE to go up even more because of all the assumptions there, especially given Obama's virtually unprecedented ground operation.

the best way to look at a 50-state outcome this far out is to explicitly model the uncertainty in each state using more than the standard MOE formula. you guys are asking what the chances of a particular outcome in the future, and should consider all the things that could happen between now and election day. if someone's 15-20 pts up right now, it would take Dukakis-proportion screw ups to blow it in that state. but if they're 5 pts up, even if its outside "MOE" chances are much higher they could lose the state



Actually ahm it's 100 separate estimates, 50 for each matchup.

The Bonferroni-adjusted significance level would be .0005, which with a sample size of 600 would give you an effective margin of error of about 7%. That's for the "strong" classification.

For the "lean" classification (68% confidence level), the effective margin of error would be about 6% with the Bonferroni adjustment

Feel free to correct me if I screwed up the math somewhere along the way


"These numbers don't include issues like turnout potential...if Clinton draws out more Republicans ... or Obama enthuses his base ..."

Uh, yeah...

It's the turnout, stupid.

Why do you think McCain's going after Obama with everything he's got?

Hillary: "a one-woman solution to the Republicans' problems"

Hillary will turn out Republicans in droves. Dobson and Limbaugh said they'd stay home of McCain is the nominee. You can be damn sure they won't stay home if Hillary's the nominee.

Survey USA: "These are not surveys of likely voters, these are surveys of registered voters."

The Economist says it plain and simple, yet again, in their article on McCain:

"If Democrats were to deprive Mr McCain of the chance of running against Hillary Clinton, that would be the cruellest blow. Mrs Clinton would be a one-woman solution to the Republicans' problems, a guarantee that money will flow into the party's coffers and that true-red voters will troop to the polls."

They said it months ago, in their article on the Clintons:

"If what should be a cakewalk in November turns into a rout, the Democrats will know who to blame."




Hey guys -- has anyone done polling comparing Clinton/Obama v. McCain and Obama/Clinton v McCain? I'm seeing this race as headed to a brokered convention. Unless both names are on the ticket, I think there will be major defections on both sides - or just "stay at homes." Anyway, would be interesting to see this data - although I am very skeptical on polling data for November. These polls just show how people think about the candidates today.



@URI: The numbers dont lie, you dont have to be racist to point out that this is identity politics at its finest given the trends in the different voter blocks. But once again if the AAs boycott a Clinton ticket and therefore dont deserve Universal Healthcare, neither do the McCain defectors in the case of an Obama ticket. I guess my point is that the conversation here always trends towards (at least in the case of Clinton supporters) well if he cant carry Ohio/PA the white Reagan democrats he cant win the GE. I was simply pointing out that there are many democratic voting blocks which include the AAs that will be tantamount to a democratic favorable outcome in November. While I agree that AA democrats will vote republican on a cold day in hell, their turnout for a Clinton nomination matters. For example in a state like maybe Michigan that we have solidly placed in the Clinton corner according to the above analysis, if the AAs are mad, upset and disenfranchised, (I can hear Jesse Jackson already) Clinton could have a problem.



A just-released Newsweek poll has Obama up by 1% against McCain; while Clinton has a 2% edge over the Republican.



I'm surprised people are fighting over the Survey USA numbers. I thought we'd all agree they are meaningless. When it gets closer in, if Survey USA starts applying a likely voter model, then you can argue, but now - I thought this was obvious - they're just beta-testing.

Uri, the African American vote is usually split 80-20 or some extreme like that in favor of the Democratic party. Its nothing special attributable to Obama - It'll go that way in the general no matter what. The only question is how high can Obama jack up the AA turnout in the states that actually matter - like Ohio - or did that population perform at capacity in the primary. At least, that's the only question I have. (Clinton, by contrast, would be working the suburban white woman swing voter population, the same group that chose Gore over Bush by 10 points in 2000. And there is a virtually unlimited supply of those voters. Advantage Clinton).

So much of this hissing back and forth about the general is pointless. The fact is, Hillary will have high negatives, and so will Obama. Just as high as hers when the GOP gets done with him. And the press will go along because they love McCain and are easily bullied by the right wing. Period.

So the only question is can either nominee do better than Kerry, and where.

Honestly - those Survey USA numbers - Oregon will go to McCain on the same day I grow a long bushy tail.



"The funny thing is that Hillary is considered "old establishment", considering that she is a woman, a (majority) that is even more oppressed than AAs."

Key point - she's no "ordinary" woman, but she's a CLINTON. Arguably the most successful Democratic politician-couple in the last 40 years. I say couple, because Senator Clinton takes credit for all the goodness of Clinton-42.

If she was President first and Bill was running now - it depends, doesn't it? Would she have as many personal scandals as President Clinton has? Make out with the White House pool-boy, perhaps? Pardon a gazillion people on the last day of her term, including some widely-derided by Republicans and Democrats alike (see Rich, Marc)? Would Bill have abused all Republicans as part of a vast-right-wing conspiracy? Would Hillary have lied under oath about said (hypothetical) pool-boy?

I could go on and on, but hopefully you get the picture. One doesn't need to be an MCP to be unhappy over Senator Clinton's bid for the White House, the same as one need not be racist to support Senator Clinton over Senator Obama.



RS - well put, as usual, but the problem with your analysis is that you don't see Mrs. Clinton as an individual, just an adjunct of her husband. What she takes credit for is her particular contributions - Beijing, for example - and for being part of a team that may have flubbed some details but got the big things right. Other people - the press, for example - are the ones dubbing her co-president, etc. There was a pretty detailed AP article today about this very thing.

Uri, the African American vote is usually split 80-20 or some extreme like that in favor of the Democratic party. Its nothing special attributable to Obama - It'll go that way in the general no matter what. The only question is how high can Obama jack up the AA turnout in the states that actually matter - like Ohio - or did that population perform near capacity in the primary. At least, that's the only question I have. (Clinton, by contrast, would be working the suburban white woman swing voter population, the same group that chose Gore over Bush by 10 points in 2000. And there is a virtually unlimited supply of those voters. Advantage Clinton, in my opinion).

So much of this hissing back and forth about the general is pointless. The fact is, Hillary will have high negatives, and so will Obama. Just as high as hers when the GOP gets done with him. And the press will go along because they love McCain and are easily bullied by the right wing. Period. They did it Gore, they did it Kerry, they will do it to him too.

So the only question is can either nominee do better than Kerry, and where.

Honestly - those Survey USA numbers - Oregon will go to McCain on the same day I grow a long bushy tail.


Dave in DC:

What was flukey about Kerry winning NH? Gore only lost it by 7,000 votes. Clinton won it twice. In 2006, it elected its first Democratic legislature in over 100 years, flipped both house seats from red to blue, and reelected a Democratic governor. This year, more people voted in the Dem primary than the Repug primary by a significant margin, with both of them hotly contested. Shaheen is likely to take out Sununu in the Senate race. I'm not predicting a landslide by any means, but this is definitely a purple state, and getting purpler all the time.




Those Oregon numbers are backed up by a Rasmussen poll from last month as well.



Andrew S.in California:

People need to understand how "different" Washington State and Oregon are in their liberal characterizations. They have high moderate independents that lean left and are highly educated. Republicans are making inroads that are similar to John McCain like Dino Rossi the guy that almost broke the streak of Democratic governors in Washington State. The area is just doesn't fit into a Hillary equation. People hate universal health care there, they want the government out of their lives in general, and they want a fresh face. There is no working class industrial workers that identify with Hillary over here or high Hispanic populations. McCain can easily take both of these states while Obama represents something different thus widely appealing to the educated young populations of these states. Being an ex-Washingtonian and part-time Oregonian I can really see Hillary losing these previously democratic stronghold states for elections to come.


Mark Lindeman:

On the Bonferroni thang, that is generally indicated for multiple tests on the same data -- not lots of tests on lots of related but (at least in theory) independent data.

The data might have correlated errors, but Bonferroni won't directly help with that. The data might be almost meaningless, but Bonferroni won't directly help with that, either. If it helps people to avoid taking this survey literally, OK.

Oh, for comparisons between Obama to Clinton within each state, it could be argued that the MoEs of the differences are theoretically smaller than they appear, because the samples aren't independent. It also could be argued that question order could matter -- if, as it appears, SUSA always asked about Clinton before Obama. I'll leave all that to others.


Leaving aside all the nasty and pointless comments about various candidates:

It's clear that this is the Democrats year, no matter what. Obama appears stronger, but somewhere right around the margin of error.

The more interesting question is, "What do you do if you are McCain?" I think he has three things to do to close this gap:

1) Re-introduce himself - the standard "War Hero" and all that other soft junk, in nationwide ads.
2) Find some way to look tough, such as meeting with the NY Times Editorial Board or telling Limbaugh to do something obscene. He has to show he's in charge.
3) Work the Midwest, where the Dems appear to have the most vulnerability in their base, starting with Michigan.

The Dems have to work things out on their own (and as a Democrat, I think this is just ridiculous and I can't believe people are letting it happen, but whatever). McCain has some time right now, and he needs a bit of a headstart to get to the finish line according to this poll.

How would YOU make use of this free head start?




I fail to see your point.

>Key point - she's no "ordinary" woman, but >she's a CLINTON.

Where's Barrack Obama is your average AA. Oh wait, he's born to a rich family, white mother, lived in different places, went back home, went to good schools, etc. And he gets the same cold shoulder from the media as every AA politician before him.

>Arguably the most successful Democratic >politician-couple in the last 40 years.
I love how everyone hates the Clintons. Bill Clinton was the only democratic president to get reelected in my lifetime. Actually, he was the only Democratic president elected since the early 80s.

>I say couple, because Senator Clinton takes >credit for all the goodness of Clinton-42.

>If she was President first and Bill was >running now - it depends, doesn't it? Would >she have as many personal scandals as >President Clinton has? Make out with the >White House pool-boy, perhaps
She'd be getting it a lot worse... Especially in a country that has a double standard about a woman's sexuality. What was that movie where the VP candidate is female and someone releases a rumor about her having been in an orgy in college? Same story.
Nevertheless, I never understood why Americans were so upset with the whole Lewinsky affair, except for the poor taste he's shown in using his office for it. They were consenting adults, and infidelity, though sad, is not that bad a thing. I'd rather get a president that f**cks everything that moves than one that regularly f**cks the constituency. An going with a full investigation about that.... Good thing there wasn't a shotgun wedding, it was just so ridiculous.

>Pardon a gazillion people on the last day of >her term, including some widely-derided by
How many voters even remember that about Clinton? Most remember the BJ. Besides, Bush pardoned his owned croneys, even worse.

>One doesn't need to be an MCP to be unhappy >over Senator Clinton's bid for the White >House the same as one need not be racist to >support Senator Clinton over Senator Obama.
I'm not talking about unhappiness. I'm talking about basic respect, and the fact that there is a lot of disrespect for her. Look at the number of times on TV where it is "Sen. Obama" vs. "Mrs. Clinton". She is a Senator. Nobody in the press should be using "Mrs."



Apparently, neither of my two previous attempts last night was successful...

Ciccina - I consider the Clintons as equal partners, and pretty much (at this point!) inseparable from each other. My favoring Senator Obama over Senator Clinton has more to do with governing philosophy (Clinton favors attrition) and openness (Clinton's lack thereof). You can look up my blog for two Feb post - "my personal take on the US Presidential elections," and "the battle over health care reform."

I also have a few quotes I found directly or apparently attributing the "two for one" or "buy one, get one free" quotes to President/Governor Clinton. That's a March post "Two for the price of one."

Erik Hare:
Apparently, Karl Rove thinks this extended Democratic fight is *bad* for Senator McCain, as it keeps the spotlight on the Democrats... At least, that's what I think I heard on the morning talk shows.

I was trying to separate opposition to Senator Clinton (Obama) from opposition to Women (Blacks). There may be some latent sexism and racism where it matters - among voters, forget TV talking heads - but hopefully not so much.

As for Americans being upset over the Monica Lewinsky thing - maybe it's the purported puritanical history of America, the rise of the Christian Right in recent times, and the fact that President Clinton LIED under oath. If he had just admitted to having sexual relations with "that woman," I think the repercussions would have been less (not zero.) [You should search my blog - click my name - for "monica lewinsky."]

Believe me, if Senator Clinton's the nominee, all that baggage - pardons, Presidential library donations, everything - is coming out. If not directly by Senator McCain, then by the 527s.


RS - I'm happy to check out your blog (I like it very much) but ever since my dog launched her jihad against every cable in my house, my internet access has been limited.

My point of disagreement with you is your choice of verb tense. Its not "all that baggage - pardons, Presidential library donations, everything - is coming out." It already has come out. Everyone already knows about it. We had 8 years of the press bashing us over the head with it. The only people who might be unaware of it are the youngest voters, but then Obama's camp has already convinced them that she's a lying, fake, corrupt corporate stooge who never worked a day in her life (except for Walmart) and spent 8 years jet-setting around the world to have tea parties in exotic locales.

Its already out there, and those who care, care; those who don't, don't. I don't care at all. My boyfriend, on the other hand, cannot accept that she didn't do something shady with Whitewater and the futures trading. Neverthless, he wholehearted supports her candidacy for reasons he feels are more important. Bringing all that up again isn't going to change his mind.

This isn't to say that she might not make fresh mistakes which would drive her negatives higher. Plus we can count on the press to just make stuff up (i.e. the racism charges).

Obama, on the other hand, is a different story. He's running almost as a saint, one who honestly admits he made one "boneheaded mistake") but is a fundamentally different kind of person than McCain and Clinton.

I don't believe it, but even if it were true, that is not how he's going to be portrayed. The GOP and the media created a completely false picture of Gore. As I was reminded the other day, this decent, honorable, dedicated man was portrayed as someone who would stoop to anything to get elected, someone who "would lick the bathroom floor" to become president. I could go on but you know what happened then. But if you doubt this, go back and read some of Maureen Dowd's pre-election columns. It was disgusting.

This WILL happen to Obama, and his negatives will go up. Even if he is the most honorable, decent, patriotic guy on earth. Even if he doesn't make a single mistake. This will happen. They made a decorated war veteran into a traitorous coward; and they'll do something similar to Obama.

I think this dynamic gives a certain advantage to Hillary; but I could just as well argue that Barack's negatives probably won't go any higher than the other Dem nominees.

Someone should do a comparative analysis of Gore and Kerry's negatives, where they started and where they ended up. If I can fend off the dog long enough, I'll try to do it myself. As it is, I can only type with my right hand; the left is used to keep her at bay.


Damn! I double posted again. Sorry.

Biscuit - those Oregon numbers are pure, unadulterated horse crap. A snapshot of the state eight months out from election day is worthless. Remember how Clinton, for a long time, was really far ahead in certain primaries, but Obama closed the gap in the last week or so? Same thing. The snapshot of that primary taken eight months out would have been just as worthless as the Survey USA numbers are now.

Honestly, I thought everyone would realize this. I just don't understand why people are discussing these numbers as if they matter.

Uri, I agree with you about the overt sexism displayed in the media, and think its important because it sets a pernicious norm for political discussions. At the same time, the media has tip-toed around race, which sends a clear signal to (white) people that racism is not socially acceptable.

But I would bet that quite few voters will end up saying Obama is more "uppity" than audacious; that running in his first Senate term shows he doesn't know his "place"; that he thinks he deserves "special treatment"; and other racist variations. And they'll insist, while they're saying it, that of course they are not racist. Just like the people who criticize Clinton for being "ambitious" - ambition being a positive for a man but pejorative for a woman.

Personally, I think McCain and allies will go after working class white ethnic voters in areas hard hit by job losses (the "Reagan Democrat" type) with that sort of message. I've heard this kind of talk (not directed at Obama; rather, generalized against Af Americans) from certain dreadful members of my own extended family, living in the heart of the blue zone, so its easy for me to imagine.

In past elections the GOP ran directly at the key strength of the Democratic nominee. Kerry's outstanding strength was his record as a war hero; they set about to demolish it. Obama's greatest strength (well, one of them) is his desire to 'bring people together' - watch for the GOP to portray him as someone who is divisive, who supports "special rules" or "special treatment" - messages that have a very racist component. They'll say he associates with black radicals or separatists, they'll probably trump up some stupid thing(s) Sharpton said and demand that Obama disavow him, etc. And people with a prejudice / fear of Af-Americans will go for it (even though Bush associates with Dominionists and other Christian extremists and he's never been asked to disavow them).

Really, its just that right now the sexism is overt, the racism covert. What we're seeing in the media is the overt/covert dynamic in action.


Biscuit - re: Survey USA showing McCain beating Clinton in Oregon -

Oregon has a Democratic governor, one Dem Senator, and four of their five House members are Dem. Oregon supported the Democratic candidate in the last five presidential elections. And the political environment for Dems is more favorable now than in 2004.

Barring a major terrorist attack, Oregon is not going to go red.


Biscuit - re: Oregon

4 of the 5 House members are Dems. The Governor is a Dem (as well as one Senator). Oregon went for the Dem in the last five presidential elections.

Barring a major unforeseen event (terrorist explosion; candidate implosion; third party candidate), Oregon is not going red, not if its Hillary v. McCain, nor if its Obama v. McCain.

Like I said - long bushy tail.


Brad K:

I don't put much stock in the particulars of state-by-state poll numbers this far out from a general election, but what's interesting is the broader pattern, which is broadly consistent with primary results so far: Clinton is stronger in the Northeast and Florida, while Obama is stronger in the Upper Midwest (Wisconsin, Minnesota, Iowa, the Dakotas, Nebraska), the Southeast (Virginia, North and South Carolina), the Mountain West (Colorado, Nevada), and the Pacific Northwest (Washington, Oregon).

Clinton's "electability" argument is that she can carry New Jersey, Pennsylvania, Ohio, Michigan, and Florida due to her strength with seniors and blue-collar white voters. The initial SurveyUSA results do indeed show Clinton doing better in New Jersey, Pennsylvania, and Florida, but they have Obama and Clinton running about even in Michigan and Ohio. This is consistent with Ohio exit polls that showed about as many Democratic primary voters saying they'd be "satisfied" if Obama is the nominee (67%) as Clinton (69%), suggesting the Ohio primary vote wasn't so much an anti-Obama vote as a choice between two attractive candidates, either one of whom can win there in the fall.

Obama's "electability" argument is that New Jersey, Pennsylvania, Ohio, Michigan, and Florida all remain very much "in play" for him in a general election, but he is also competitive in bunches of states where Clinton is not, giving him a much broader electoral playing field, a higher margin for error, and the ability to force McCain to play "defense" in states that in recent elections the Democrats have conceded to the Republicans. That argument is broadly supported by these initial polls, by the primary results to date, and by the kind of coalition Obama has built, with support from large numbers of Independents, some moderate Republicans, unprecedented youth turnout, and a highly energized African-American base.

Both are plausible arguments, but to a lot of Superdelegates, Obama's is going to look more attractive because it suggests a capacity to grow the Democratic Party and to help down-ticket candidates in a lot of marginal and Republican-leaning districts. Yesterday's upset win by Obama-backed Democratic congressional candidate Bill Foster in former Speaker Dennis Hastert's old district--a district Republicans had held for 35 years and which Bush carried with 55% of the vote in 2004---adds further luster to Obama's claim.



Well, first I think that the democrats are doing better in gubernatiorial and senatorial races as in the presidential elections.

Second, a nomination of Hillary Clinon IS a major unforeseen event, and when Hillary gets the nomination, what is not going to happen, I am looking forward to see you with a long bushy tail...
I do not think that CLinton would lose the general election by a landslide, but there are blue states she would probably lose- including Oregon, I think.



Well, first I think that the democrats are doing better in gubernatiorial and senatorial races as in the presidential elections.

Second, a nomination of Hillary Clinon IS a major unforeseen event, and when Hillary gets the nomination, what is not going to happen, I am looking forward to see you with a long bushy tail...
I do not think that CLinton would lose the general election by a landslide, but there are blue states she would probably lose- including Oregon, I think.



Sorry about the (shameless) promotion of my blog, but that was an attempt to limit my post-lengths here...

However, I disagree that just because the scandals have occurred and been publicized, they won't be brought up again. One important rule is that public memory is short; once those days are brought up again, not sure how many independents/swing voters will want the Clintons back in the White House. The campaign Senator Clinton will likely wage is the old game of tenths-of-a-percent, and any ABC feelings can swing things Red, especially since the GOP has Senator McCain, not Governor Huckabee.

At the same time, much has happened since 2000; the NYT article about President Clinton's visit to Kazakhstan with a uranium mining company was just the start. I have the impression that when the Clintons left the White House, they didn't have much money, but have made millions since. Writing books and lectures is fine, but that may not explain everything. Not when even the Gray Lady puts out the uranium mining and other stories.

Anyhoo - I still like my football analogy. A McCain-Clinton match-up will have the Dems playing D; a McCain-Obama match-up will have the GOP playing D.

I look forward to seeing your bushy tail :-)



I really enjoy these polls but I do believe from all the people that I have been talking to which are mostly women, we will and I mean seriously will begin supporting John McCain if Barrack Obama becomes the candidate. So come November I think there will be quite a surprise when John McCain walks away the winner over Barrack Obama



I really enjoy these polls but I do believe from all the people that I have been talking to which are mostly women, we will and I mean seriously will begin supporting John McCain if Barrack Obama becomes the candidate. So come November I think there will be quite a surprise when John McCain walks away the winner over Barrack Obama



Take a look at what this site is doing with those SurveyUSA numbers... http://www.fivethirtyeight.com/



I think more than anything else these surveys suggest that Clinton and Obama would have to achieve 270 EV's in different ways. I would be interested in hearing who the group thinks would be a better candidate for down ballot races? My instincts are to lean Obama but am willing to listen to the groups' logic.




Same here. If Hillary wins, me a group of DEMS in my area will also vote McCain.


Linda, I think the frustration of Hillary supporters is finally catching up with Obama. We've been listening to Obama supporters bark on about how she's horrible, a beast, the same as Bush, blah blah blah, and how they won't vote for her no matter what. And its finally worn a lot of people out - myself, and an increasing number of women I know.

I think the prospect of Hillary losing to Obama touches a real nerve with a certain demographic - the thought that someone as qualified and accomplished as Hillary would lose to a showboat who hasn't even completed one term in federal office is too grating. Not to mention that when a lot of people hear the actual words he's been saying (rather than descriptions of the words) they get freaked out. Its like listening to a cult leader - all this quasi-spiritual language. Frankly, among my friends, mother, mother's friends, friends' mothers, I've only ever seen one of two reactions to him. Either they get creeped out by the whole messiah thing, or they bust out laughing saying "*that's* what everyone is getting so excited about?"

One aspect of the prospects for November I haven't seen talked about it what will happen to the gender gap. Remember, the big advantage Gore had in 2000 all but evaporated for Kerry because of the security issue. The threat of terrorism and worry about Iraq were still too raw. If independent (white) women don't feel confident that Obama can handle national security, there's a good chance they'll walk. I think what we saw in 2004 was that yes, women voters care the most about the issues Democrats do best with - economy, health care - but its predicated on their feeling of safety. Its safety first, in other words.

To me this is a real concern; there is nothing Obama can do between now and November to shore up his foreign policy/security cred. The fact is he was never involved in foreign affairs at all before he joined the Senate. He can criticize McCain for supporting the war, but the fact is that at the time we invaded Iraq, no one would have let Obama get anywhere near a decision-making role. McCain can at least say that he knows what it is like to take responsibility, to make hard decisions and have to live with the results.

Obama has huge gaping vulnerabilities on the issues vs. McCain, and unlike Bill Clinton, who was really likable in 91-92, Obama comes off as smug and arrogant.

Professional women, mothers, women of a certain age want someone qualified, some one who can handle crises, someone who knows what s/he's doing in the White House. Obama is not that guy, and no amount of speechmaking is going to change that.

RS - I'm not saying the scandals won't come up again. I'm saying they are already out there right now. No one has forgotten the impeachment / Lewinsky, and people like Chris Matthews and Maureen Dowd will continue to beat the drum as loud as they can. But after the 10th, 20th, 100th time the message is delivered and the candidate is still well-regarded by a large group of voters, you have to admit that the message just isn't going to work with those people.



I hardly see this as "incontrovertible proof" of Obama's superior electibility.

1. When you take out the tossups, Clinton does better than Obama against McCain, winning 249 to 244 EV's (obviously that margin is so close that it is much more accurate to say that both candidates are about even).

2. Where the candidates differ is in the variability of prospects. With Obama, an electoral sweep where McCain only wins a few states is probably possible. However, a McCain sweep is also possible. Obama's unknowns are high, and his electoral coalition is new and untested.

3. The Democratic party organization in the states Obama will need to win is comparatively weaker than those for Clinton. Obama will need to win in some combination of Virginia, Texas and South Carolina (as well as others) - states where the Dems are weak historically. For Clinton, by contrast, Tennessee is the only historically red state that is a toss-up (and it is not a must-win for Clinton).

On that count I would say that Clinton is the more electable and safe candidate, although she is not likely to sweep 40 states, whereas there is a chance Obama will.



You still assume a long-lived public memory, whereas I think otherwise; I believe conventional wisdom is on my side. Even if that chance is small - only a few tenths of one percent of voters have to be reminded of the bad aspects of Clinton-42 - that might be enough to swing the election to Senator McCain. Remember, Senator McCain is no typical Republican, but almost a Democrat.

The way opinions are presented here on this board, it appears as if irrespective of whether Senator Clinton or Senator Obama is the nominee, each is going to lose. Women will walk away if a man with 20+ years of varied life experience (successful and smart, not a legacy candidate like Bush-43) is nominated, while Obamacans, ABCs and some independents will vote McCain if Senator Clinton wins.

When Senator Kerry was elected the nominee in 2004, I was very unhappy - the Dems kept going for the purported "electable" candidate (I supported Senator Edwards then). That's a hock IMHO - the only popular (= re-elected) Democratic candidate elected in the last 40 years (or, to my knowledge, any full-term, re-elected President since FDR, barring Nixon's shenanigans) has been someone who could connect with *all* voters - men and women, Black and White and Brown. That can be said of only one Clinton.

Sorry, but that's the truth. Nothing against Senator Clinton's capabilities, but Governor Richardson, Senators Biden, Dodd and Gephardt, and a bunch of other very fine, experienced Democrats are just as capable, were directly involved longer in elected and/or accountable positions, and yet were soundly rejected by the electorate, in 2004 and 2008.

Here's a thought experiment - IF Senator Clinton was not running this year, who would the Democrats have voted for? Would we have seen this same "oh, a suave, smart, young, relatively untainted, dynamic person is winning over an experienced person," so we should not vote for him dynamic?

You can keep thinking that voters are choosing Senator Clinton for her experience, but all those fine, experienced, multiply-elected Democrats who dropped out after getting less than 5% of the vote in Iowa and NH would beg to differ.



Virginia is turning blue as Northern Virginia continues to grow. Next year both senators will be Democrats. The Governor is a Democrat and now the state senate is controlled by Democrats. And it has 13 electoral votes.

As to the point about who is better in a general election, none of us really know. It would seem from these surveys that Obama puts more states in play. But it is just one survey and the election is 8 months away. There seems to be some weakness for Clinton in the pacific northwest? If she lost WA and OR she would have to win Florida. Just playing around with the EV totals it looks like Obama would have more combinations than Clinton in getting to 270.



We�ve been talking alot about how Obama or CLinton could get to 270 ELectoral Votes.
That�s fine.
But I don�t see, how McCain would get to 270 EV�s.
Sure, if he wins all the states the Dem candidate loses in your scenarios, he gets the 270 EV�s, but I mean a REALISTIC scenario.
I still do not see CLinton winning Florida, on this point it�s the same with me as with Ciccina and Oregon. BUT I also do not see McCain New Jersey, Michigan, even Missouri...

Another thought for the Obama scenario ( I like it to give thoughts without having to work them out :-) ):
If he won Texas, he could compensate Florida with that and the polls show a toss-up there...

I am german, interested in american politics, but I know nothing about the political environment in Texas except what I read here and on other blogs in the days before the primary.
Do you think an Obama win there is possible?


Yancey Ward:

Let me first say that this site sucks for software. There are way too many double posts and posts that never see the light of day. Fix it! I am getting tired of rewriting comments.

On the the issue at hand.

Any survey that claims Nebraska, North Dakota, and Texas are tossups is worthless, and, even worse, completely laughable.

As to which candidate for the Democrats is more electable in November, I think it is becoming more clear by the day that the conventional wisdom is completely wrong. The elephant in the room that no one seems willing to notice is race. Look to where Clinton is strongest and you will see the racial divide in the vote most clearly. The Bradley Effect is live and well, and any survey of Obama vs McCain will have to be taken with a grain of salt. I think it very conservative to subtract 5% from Obama's polling numbers vs McCain.

Now, it may very well be that either Democrat will win in November- I certainly wouldn't give McCain even odds, but if McCain loses it will be due to the dispirited Republican voters. However, it seems clear to me Clinton will actually perform better than Obama in the general election.


Bill Clintoni:


That had to be one of the most ludicrous posts I have ever seen.

Are you with the Clinton campaign? If so, please go to hillaryclinton.com and post there. Just saying "that is wrong" makes no sense. Do you have any evidence or are you just praying it is so?


Andrew S. in California:

The whole "Bradley Effect" is garbage IMO. What's to prevent Obama with a "Dinkins Effect"? This is laughable statistical science. The conclusions from these effects are uncharacterized and are not isolated in statistical significance beyond race. Just to find a couple instances like Bradley and Wilder then come to the conclusion that some sort of pattern exists is bad science and to come here and post that a similar case will happen with Obama is statistical propaganda.


RS - did you really mean to write "Remember, Senator McCain is no typical Republican, but almost a Democrat"? Perhaps you meant "McCain is perceived as being more aligned with the Democratic agenda" or somesuch?

Because McCain is a hardline conservative but for tripping up in two areas with the wingers. One, campaign finance reform, and two, appearing to be disloyal to the president. I don't know his stand on immigration or with whom he is aligned, so maybe that puts them at odds too. (And possibly they hold against him the fact that he dumped his first wife after she had a disfiguring accident so he could marry a much younger heiress whose father bankrolled his political career. Its a possibility). But beyond those troubles, he is a hard core, straight up conservative.

In short, "almost a Democrat"? arghehhhehhasrhhrhgg! Are you trying to make my head explode?

Ben, I see your point re: Virginia. But then again Virginia had a Democratic Governor (Wilder) and one Democratic Sen. (Robb) in 1992 but still stayed red in the presidential (and every presidential since then). And they loved that dreadful George Allen.

I'd love to see a demographic breakdown of the changes to Virginia since 2004. I think the military vote will continue to be very important there, esp. with a former POW on the ballot. I think a persuasive case can be made for McCain understanding the needs of soldiers, veterans, and their families at a deeper level than Obama AND/OR Clinton. (Not saying I agree, or that I don't think counter-arguments could be made; just that its potentially a powerful message).



First, admitting a mistake - I meant the notion of "the most electable candidate" is a CRock, not a Hock. :-)

Perception is reality in politics and all that :-)
There are quite a few areas where Senator McCain is at odds with the GOP. As you note, campaign finance reform; though I think now he's loyal to a fault (Iraq, tax-cuts etc.) There are other areas:
immigration - McCain-Kennedy allows a path to legalization of illegal immigrants. Something that ought to endear him to Hispanics in particular. Senator McCain took a lot of hits during the GOP debates on "amnesty."
climate change and the environment - McCain-Lieberman Stewardship Act (apparently weak, but anything's a huge improvement on the typical GOP stance) - helps with the Greens.

And it also helps that Senator McCain is not overtly religious/evangelical. After all, Senator Kerry did ask Senator McCain to be his VP!

I think there was an article that said "conservatives think Senator McCain's telling the truth on tax-cuts, etc. while liberals think he's just lying to appease the GOP base." Or some such - e.g. his hobnobbing with the Religious Right this election cycle, compared to his comments in 2000. Everybody (publicly) loves Senator McCain, and Democrats aren't gonna swift-boat a VietNam vet/POW over any dumping of said disabled first wife. That's something the GOP would do.

Virginia - am I glad you see the potential. My impression is that northern Virginia is filling up with White/affluent DC transplants, which might help the Dems...


John - Spokane, WA:

These tables are totally speculative, these guys are struggling to get the primary polls right - they better concentrate on that for now. This reminds me of some accountant trying to dazzle boardmembers at a corporate meeting, usually the accountants just end up downsizing these groups right of business. Alot of hype and no bearing. Sorry






3/9/08 338 LV 37 54(Obama)



Glad to see Darrell Huff's 'How to Lie with Statistics' is still well read. How may examples can you find?


RS - Virginia. On the one hand you have the Dems making gains, on the other you have the fact that Bush beat Kerry there by 8 points (he beat Gore by 8 points as well). Things are changing, and the environment is different, but is it enough to close an 8 point gap? I still doubt it.

I certainly agree about perception. The media tells us McCain is a moderate; many voters believe he's moderate; therefore he's a moderate. I believe I'm paraphrasing a line by Anthony Burgess when I say that its funny how some things only seem really real when you view them on a screen.


David O:


Virginia is going blue for Obama in November. Sorry, it is true. Obama is no Kerry. Just check out poll snapshots of him. Have you ever in your life seen a candidate skyrocket like that? No. And it is just the beginning. Wait until he campaigns in smaller settings with extended time. McCain is a decent guy but no one is going to vote for an ugly, handicapped dinosaur.

Obama changes the map in way that no one else in politics could have (and most people STILL haven't heard a lot about him). That is why poor Hillary has no chance.

All the Dems know that with her conservatives will come out of the woodwork to vote against her, but with Obama conservatives will vote FOR him. (How do you think Bill Foster just won Hastert's seat in a heavily Republican area?)

If they were smart, they ALL would have endorsed him some time ago. But who ever said politicians were smart *cough Spitzer* ?

Mississippi -

65 - Obama
35 - Clinton

By the way, he is going to win Pennsylvania - you heard it here first folks.


Comm Director:

David O -- I didn't see one single piece of data-backed point in your argument. Sure, you mentioned key examples (Hastert, Spitzer, Kerry), but your comment does absolutely nothing to counter the fact that Ciccina et al have been talking. It makes your last comment about Obama winning Pennsylvania sound bogus and full of false air.

FYI, I am a dyed-in-the-wool Obama supporter.

This is more complicated than simple "hope" vs. "experience." Lets not boil it down to "new, fresh guy" excites more people vs. the "other candidate."



Congratulations on an excellent job of refining the Survey USA data and making it a whole lot more meaningful. Your work clearly shows that Clinton is a more polarizing candidate than Obama and that Obama potentially has many more states that he could add to his column in a race against John McCain. However, I would just add the cautionary note that this reflects voter sentiment as of March 2008 and can't predict what it will be in November. By that time and if Obama is the Democratic nominee, the American public may be scared to death of Obama's lack of political experience and may swing towards McCain. However, I had a lot of fun looking at the data, and my hat is off to you for the work that you put in.



Time marches on and with it there are new polls. Here is a website that, like Mark's, starts out with the SurveyUSA results and continually updates them as new polling information becomes available. As of today it shows a much tighter race than does Mark's analysis as it gives all the states in the "toss-up" category in a race between Obama and McCain to McCain except for New Jersey. At the same time it awards all the states in the "toss-up" category between Clinton and McCain to Clinton except for Washington and Tennessee. I don't know which of these states have had more recent polling data than did Survey USA in early March. However, the MyDD analysis did award some states to McCain that SurveyUSA awarded to Obama (New Hampshire, Michigan, and Virginia) so I assume that is based on more recent polling.



Sorry, I forgot to post the website. http://www.mydd.com/



I would like to make one adjustment to the MyDD data. They have McCain leading Obama in electoral votes by 273 to 265, but they have Pennsylvania and Ohio in Obama's tally. The latest polls according to RealClearPolitics show that these two states would likely go to McCain if he were running against Obama. Therefore subtract 41 electoral votes from Obama's total and give them to McCain. The result is 314 electoral votes for McCain to 224 for Obama. MyDD's total for a McCain vs. Clinton contest don't change as MyDD has both Pennsylvania and Ohio going for Clinton which is what the RealClearPolitics poll of polls also shows. This results in 276 electoral votes going to McCain and 262 to Clinton. Therefore, from this one can conclude that although McCain is ahead against both Obama and Clinton that the race for electoral votes is much closer between McCain and Clinton.


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.