|Some Counter-Arguments Countered.
(1) This is just a measure of name recognition, nothing more. By the end of the campaign whoever runs will have close to 100% name recognition.
Counter-argument 1: Name recognition surely plays a role, but is hardly determinative. Edwards outperforms Pennsylvania governor Ed Rendell in his home state, and Rendell has virtually no noticeable advantage in next-door Ohio.
Counter-argument 2: Name recognition comes along with favorable and unfavorable impressions and is something that politicians work hard to earn. The fact that Edwards has higher name recognition and is a more formidable candidate is not something that we can simply assume a less-well-known candidate could make up for-particularly if they do not fit into the analysis summarized above.
Counter-argument 3: Name recognition at the end of the process is irrelevant. What matters is what the candidate brings to the ticket, and name recognition plus strong polling as a VP is certainly a definite advantage. Someone without that advantage should have compelling arguments to make as to why they would be better in the long run-and the fact that they will eventually have higher name recognition is not such an argument.
(2) Hillary should be on the ticket. She earned it.
Counter-argument 1: There is no precedent for this. Candidates have chosen to take the runner-up as their running mate sometimes, and have passed them over other times. This, the only precedent is that it is the candidate's choice.
Counter-argument 2: Clinton clashes with the logic of Chris's realignment argument above. She is an expert insider politician, and she campaigned as an expert insider politician. This directly contradicts the logic of the realignment that Obama is facilitating.
Counter-argument 3: Clinton's conduct and attacks on Obama's lack of experience, coupled with her praise for McCain would make her a particularly discordent VP candidate, the exact opposite of a reinforcing one.
Counter-argument 4: The strength of Edwards as a VP candidate shows that Obama does not need Clinton, despite hurt feelings that remain among many of her loyalists. The voters as a whole have moved on.
Counter-argument 5: As will be seen below, the Obama/Edwards ticket does just fine with female voters.
Counter-argument 6: Clinton would better serve the party as Senate Majority Leader.
(3) The poll questions put unrealistic and disproportionate weight on the VP slot.
Counter-argument 1: It's a standard practice to ask people questions about possible running mates. If candidate preferences are firm, then little variation will be found. The fact that wide variation is found can only be explained if two things are true: (1) Support for the presidential candidates is still significantly fluid. (2) At least some of the VP candidates have significant impact. These are logical necessities.
(4) This poll is unrealistic, because in the long run, the importance of the VP will fade.
Counter-argument: There is certainly some truth in this, as other issues are bound to emerge over the course of the campaign. However
Counter-Argument 1: The choice of a running mate is a key indicator of judgment that will have enduring importance and will not go away over the course of the campaign.
Counter-Argument 2: Short-term advantage can help determine the course of the election early on-and in this case the advantage gained from Edwards is decisive in key swing states.
Counter-Argument 3: The argument laid out at the beginning of this post concerns the identity/issue structure of the party. Even as the importance of the VP as a person may fade, these other factors will not. Thus, a VP choice who is in synch with the argument stands a high chance of not having their impact fade, even though it is not their personality per se that is responsible for the impact.
(5) You are claiming that Edwards is the only one who can deliver these states, and that's just not realistic.
Counter-Argument 1: No, I am not claiming that. I am only claiming that he is the only one who can reliably deliver them now, and thus provide us with the momentum to keep the GOP playing defense. Since the GOP is playing defense in both House and Senate races across the country, keeping them on defense at the presidential level in key swing states as well is a powerful means for keeping them off-balance and error-prone throughout the entire election cycle. It's not a long-term guarantee-there is no such thing in politics. But it's a very good way to start out.
(6) There must be something wrong with these polls. I can point to a number of results that just aren't believable.
Counter-Argument 1: Unbelievable results turn up all the time. Just because you don't believe something at first glance doesn't mean it's not true.
Counter-Argument 2: Even being extremely generous, and throwing one poll out as completely flawed, all four of the remaining polls would all reach the same conclusion: Edwards is the clearly superior Dem VP candidate. With SUSA's reputation, it simply defies reason and experience to claim that all five polls are flawed. This argument takes wishful thinking to the level of religious faith-and beyond.
There are certainly other arguments that could, and no doubt will be made. But this is a good roundup of the ones seen so far, and why they do not carry great weight.
Analysis of Cross-Tabs
And now, onto the analysis of cross-tabs. For this analysis, we will look at each state in turn, taking cross-tabs in the same order throughout. SUSA does not provide intersecting cross-tabs, so we don't have figures for white males, for example. But a good surrogate can be constructed by multiplying the figures for these two categories together, then multiplying by 2-an approximating "renormalizaiton" that accounts for the fact that males are roughly 50% of the voting population. Since we're using the same measure for everyone, and are only looking for general trends, this approximation should do just fine. We use it for white females as well.
In Pennsylvania overall, Obama has a solid 8-point lead:
Governor Ed Rendell helps Obama in every matchup except Huckbee, where he gives only a little ground, but Edwards is consistently better: Hagel and Sebelius lose ground against every GOP pairing.
Turning to voters over 65, McCain has a 9-point edge--a 17 point swing from the 8-point edge that Obama has among voters as a whole.
While Edward only gives Obama an outright edge against Pawlenty, he cuts McCain's lead with every running mate in the over-65 age group. As Rendell's performance shows, all that's really needed in this demographic is to hold McCain's margin to what it is in the two-person matchup. However, Edwards runs strongest among this demographic, too.
Obama enjoys a slight 3-point edge among independents-less than half his overall edge.
This is a rare instance where Huckabee outperforms Edwards-and does so decisively. Edwards is just barely better than Hagel here. However, Edwards continues to outperform Rendell in all other matchups, and the rest of the Dem VP showings are quite weak.
The story with moderates is dramatically different-a 23-point Obama edge.
Both Edwards and Rendell manage to improve on the large 23-point Obama edge, except for Rendell vs. Lieberman, where the loss of ground is tiny. The other candidate lose substantial ground, from 5-13 points.
Obama enjoys a comfortable 11-point edge with white females-hardly an indication of a lingering Hillary problem:
Once again, Edwards and Rendell both improve Obama's edge across the boards, with Edwards leading the way. Tellingly, Sebelius does not show any superiority to Hagel, or to her general level of performance seen so far.
Among white males, Obama manages a narrow 1-point margin:
Again, Edwards shows across-the-board appeal, strengthening Obama in all matchups. Edwards basically holds the line, making headway only against Pawlenty. Sebelius and Hagel lose badly, except for Hagel vs. Pawlenty, where the loss is more modest.
In Short: Edwards shores up potential weaknesses with white males, independents and voters over 65. At the same time, he pumps up the already substantial margins with white women and moderates. Rendell has a generally similar profile, but with a weaker effect and some holes. The only Edwards hole is against Huckabee with independents.
Turning to the ultimate swing state, we find Obama with a solid 9-point lead:
Outside his home state, Rendell is weaker overall than Sebelius. Only Edwards reliably extends Oboama's lead, doubling it against Pawlenty.
Obama is even weaker with seniors in Ohio than he was in Pennsylvania, loosing by 15 point:
Edwards holds the line across the board, reducing Obama's loss modestly in two cases, slightly in one, and drawing even against the hapless Pawlenty. The other three show little variation in their showings against the different GOP VPs though Pawlenty is weaker as usual, while Romney is surprisingly strong.
Obama is also significantly weaker among independents, at minus 10:
This is Edwards' weakest showing so far. He loses ground vs. Huckabee, holds even against Lieberman, and cuts Obama's losses against Romney. The only flat-out win is against Pawlenty. But he's solidly ahead of all the rest in every case, and overall there's no striking difference among them.
Again, with moderates is strikingly different as Obama has a 29-point edge:
Even at this high level, Edwards extends Obama's lead against every opponent except Lieberman, and scores nearly double the margins of the other VP candidates who are tightly bunched, and do not hold Obama's margins even against Pawlenty.
Obama holds another daunting advantage among white females, 15 points:
But only Edwards can improve things for Obama, as the others all lose substantial ground. Huckabee is particuarly effective. Only Sebelius manages to hold a slight lead for Obama.
Among white males, Obama has a good 6-point lead:
Again, Edward is the only one to improve on this lead, all the rest turn it into a modest or substantial loss, except against Pawlenty.
In Short: Edward is the only one to hold and improve Obama's substantial leads overall, among moderates, white females and white males. He cuts losses against all comers among seniors, and against two candidates among independents, holding even against one more. Huckabee is only one to best him in that single category.
Obama leads in Virginia by 7 points overall:
Again, Edward is the only candidate who reliable wins Virgnia with Obama, increasing his lead against all comers. The rest hold slight leads against Pawlenty, while Sebelius just squeaks past Huckabee.
Over 65 again favors McCain by a whopping 15 points:
This time, Edwards losses ground-though only slightly--against three of four candidates. He only makes significant gains against Pawlenty. The others are fairly uniformly worse, with a slight nod to Hagel.
Among independents, Obama has a modest 4-point lead:
Only Edwards maintains Obama's lead, increasing it significantly in three cases, while yeilding ground slightly against Lieberman, but still maintaining a win. The rest lose badly to Lieberman, less badly to the rest.
Again, Obama does an outstanding job among moderates, with a 31-point lead:
Edward yeilds a single point against Lieberman, but otherwise increase Obama's lead. The others yield ground significantly, especially against Huckabee and Lieberman.
Among white females, Obama is down six:
Edwards alone takes Obama into winning territory against all 4 GOP VPs, a feat none others manage even once as they double, triple, even quadruple the margin of his loss.
Amongst white males, Obama's loss is double that among women-12 points:
Edwards pulls close to even with Huckabee and Romney, cuts the loss almost in half with Lieberman, and rolls up a handsome little margin against Pawlenty. The others repeat their dismal performances among white women, but with an additional loss of five to ten points.
In Short: Edwards again is the only one to consistently increase advantages and cut losses in most cases, though even he has a harder time in Virgnia in some cases. The rest are almost uniformly miserable, if not downright disastrous.
Obama is tied with McCain head-to-head in New Mexico:
Again, the difference is clear: Obama wins with Edwards, and loses with everyone else. Even Pawlenty gives McCain solid leads of 4 points or more, while Huckabee wins by double digits.
Obama loses New Mexico Seniors by four points:
Edwards improves in every category-still losing slightly to Huckabee, but beating everyone else. The rest all get worse-modestly against Pawlenty, significantly against everyone else-though Hagel is consistently the least weak of the three.
Obama loses independents by a substantial 13-point margin.
For once, there is no significant Edwards advantage. In fact, Hagel outperforms him slightly. Edwards does cut Huckabee's margin by a hair's breadth. Hagel cuts it more, and cuts Lieberman's margin as full three points. Sebelius cuts Pawlenty's margin by two.
Obama cleans up among moderates, however, with an even stronger 19-point lead:
Here, too, Edwards re-asserts his dominance, improving Obama's massive lead against all opponents, while the rest barely hold on to a fraction of Obama's lead. Only agains Pawlenty are all three in double digits.
With a substantia non-white population, Obama loses white females by 14 points:
Edwards improves Obama's performance in every category, winning outright against everyone but Huckabee. His seven-point loss against Huckabee is half of Rendell's loss against Pawlenty-the next best performance by another DEM VP. Huckabee more than doubles McCain's margin against Hagel and Rendell.
Obama has an identical 14-point deficit with white males:
But Edwards cannot help as much this time He only wins outright against Pawlenty. But he still improves against all the rest. No other Dem VP improves against any GOP VP, though Rendell holds even against Pawlenty.
In Short: This is the roughest state so far, the first one Obama doesn't win outright. But Edwards still improves performance in every matchup, except among independents.
Obama leads by 8 in California:
Edwards improves Obama's performance against all comers, the rest lose ground, and two lose outright to Huckabee.
Obama trails by a whopping 17 among seniors:
Again, Edwards is the only one to improve matters for Obama. Sebelius holds even against Pawlenty. Everyone else losses ground.
Obama loses independents by 5 points:
Edward wins one, gains ground in one, but loses ground in two. Everyone else losses ground across the board-embarrassingly so against Huckabee and Romney both.
Obama enjoys a comfortable 16-point edge among California moderates:
Edwards improves on Obama's high standard in three cases, holding steady in one. Everyone else loses gorund modestly against Pawlenty, more seriously against the rest.
Obama carries white females by 6 points:
Edward more than triples Obama's margin against Lieberman and Pawlenty, but loses just over a point against Huckabee, and gains a similar amount against Romney. Everyone else loses the white female vote against all comers-includign Sebelius.
Obama has a dramatic 17-point margin among white males:
Edwards increases that already substantial margin across the boards, doubling it against Pawlenty. No other candidates improves the margin against any GOP VP. Leads collapse almost entirely against Huckabee and Romney.
In Short: Edwards again is far superior to any other VP candidate. He struggles occassionally against Huckabee, but routinely outperforms all others, often embarrassingly so.
Edwards is superior to all other VP candidates by margins that persist in virtually every category in almost every state. This is significant evidence that his appeal is not limited to a single demographic. More than that, he both improves the margin where Obama is already substantially ahead, and cuts the loses where Obama lags, a clear indication of the breadth of his support. While none of the other Democratic VP candidates are particularly strong, it is difficult to imagine another VP candidate who could match the breadth of Edwards' appeal.
It is certainly conceivable that Clinton might, and if she had conducted a different sort of campaign, one might even be able to argue that simply by virtue of being a woman she had fulfilled the qualifications of being a reinforcing "change" candidate, if not an outside vs. in one. But as it is, she does more than fail the test of fitting the theory of realignment that Chris laid out in late 2004-a theory that Obama's success has confirmed beyond Chris's expectations at the time. She does more than fail that test-she campaigned against its premises as an insider expert. And for that reason, one need not even consider how she would have done in the SUSA matchups. Of course, as a nerd I would love to have those numbers. But Edwards fits the theory and he delivers the needed margins of victory quite comfortably against all comers.
Forget John McCain. Kang and Kodos don't have a chance.