Friday, August 08, 2008

The Electoral Map Is Likely to Change A Lot...Yet Still Not That Much

For months now, we've been hearing about the Obama campaign's wishes to redraw the playing field and mix up the red state-blue state divide of the last few election cycles. At first I was skeptical that much terrain outside of the usual litany of "battleground states" would ultimately be in play, but I've seen enough state polls to where I'm convinced that we are indeed looking at an entirely different political map in 2008. As someone who has been obsessed with political demography since I was 11 years old, that premise has me fascinated with checking out the elections results map on the morning of November 5, so much so that I was actually dreaming last night about studying said maps. Nonetheless, I can't help but expect the county-by-county results maps are gonna look far more different from 2000 and 2004 than will the state-by-state maps.

There is a disconnect between the tight national polls and the media perception that little has changed since 2004. Scores of state polls provide overwhelming imperial evidence that the map ALREADY HAS changed in a substantial manner, only in a way that produces little net benefit for Obama on a national level. Obama has closed 20-point gaps to statistical dead heats in states like Indiana, North Dakota, and Montana. That is absolutely huge in terms of realignment. And even if Obama doesn't win these three states, as I suspect he will not, closing a chasm of that stature is no small feat and will give Obama license to declare success in his efforts to expand the political map, even in narrow defeat.

So with Obama tying McCain in states that Bush won by 20 points, how is he no better off (or barely better off) than John Kerry was four years ago in national polls? The answer to that question can likely be answered with one word: Appalachia. The gains Obama has made in Montana, Indiana, and North Carolina would appear to be nearly completely offset by his underperforming of past losing Democrats in states like West Virginia, Kentucky, and Arkansas. And I would be incredibly surprised if this gap narrows in the next three months, and would actually bet on it getting worse. If Bush won West Virginia by 13 points, expect McCain will win it by 23 points. Bush won Kentucky by 20 points...and McCain will likely win it by 30 points. Bush won Tennessee by 14 points...McCain will likely win it by 24 points. My bet is that there are only about seven or eight states, all of which went Bush in 2004 with the possible exception of Michigan which is perhaps very close, that are single-handedly dragging Obama's national numbers down to the near-parity with Republicans that we saw in 2000 and 2004 races.

I suspect this Obama-centric realignment will be abundantly clear on the national county map. Instead of only four of North Dakota's 53 counties going blue as was the case in 2004, Obama may win 20 in 2008. Instead of 27 blue counties in Wisconsin, Obama might be able to pull off 40. Instead of only 15 blue counties in Illinois, Obama might win 50 or more. With that in mind, the 2008 national county map could look considerably better than in the last two national elections, but with little net gain for Obama nationally. For instance, Obama could win Grand Forks County, North Dakota, but is probably just as likely to lose the Kerry County of Beaver County, Pennsylvania. More specifically, I suspect Kerry's nine-county bounty from West Virginia in 2004 will dwindle to zero counties in 2008. Kentucky's 12 counties for Kerry will probably plummet to one county for Obama. Tennessee's 18 Kerry counties will probably become four Obama counties.

Even within traditional battleground states, I suspect there will be an internal realignment in 2008 compared to 2000 and 2004. In Michigan, we're probably like to see more Wisconsin-esque counties on the Upper Peninsula go blue while the suburbs of Detroit, influenced by the metro area's racial polarization and fear of Democratic-supported increases in CAFE standards for the auto industry, trend towards McCain. In Pennsylvania, Obama's victory is likely to be scored based on a handful of affluent counties in suburban Philadelphia running up supersized Obama margins, even as Allegheny and Erie Counties are the only two counties in the entire western two-thirds of Pennsylvania to follow suit for Obama. As for Ohio, an Obama victory (which I see as a longshot) is likely to require huge Democratic gains in affluent suburban counties near Cleveland (Lake County), Columbus (Delaware County), and smaller margins of defeat in metropolitan Cincinnati, given the likelihood that Obama will underperform Kerry by double digits in the Appalachain counties of southern Ohio.

Put in a broader context, I thoroughly applaud Obama's efforts to expand the map by targeting the seven solid red states that he has. Of the seven, only Georgia strikes me a miscalculation of resource investment. Most of the states involve cheap media markets (Alaska, Montana, North Dakota) that it would be stupid for a campaign as well-funded as Obama's to NOT invest resources there even if polls showed him behind by 15 points led alone tied. Of course, his campaign's already getting cynical feedback from the likes of Real Clear Politics and others about how the campaign will have much to answer for if it loses traditional battlegrounds like Michigan, Ohio, and Florida as a result of "wasting resources" in crimson red states that would never have gone for Obama in the first place.

In other words, long live the 18-state strategy. That logic strikes me as flawed at many levels. There comes a point of diminishing returns as it pertains to campaign advertising. If $20 million worth of advertising in Ohio isn't enough to swing Ohio to Obama, then why should anyone believe $25 million would be? Again, redirecting some of that money to dirt-cheap North Dakota seems painfully obvious even if McCain was ahead there by 15 points, not 2-3 points. Ultimately, I'm doubtful that any of the targeted red states will go Obama beyond Virginia, but I'm not prepared to make that a self-fulfilling prophesy by ignoring them the way Al Gore and John Kerry did with several winnable states, and I'm glad Obama isn't willing to accept it either.
.

8 Comments:

Blogger Mr. Phips said...

Mark, what one county do you think Obama will carry in Kentucky? Im almost certain he will carry Jefferson(Louisville) and will likely do it by a good six or seven points. Are you thinking that Obama loses some of the Eastern coal counties like Elliott and Floyd that went for John Kerry by 20+ points and even went for McGovern against Nixon?

Are you sure he cant carry Ohio? All he really needs to do is increase black turnout and run even in Hamilton county(Cincinatti), while turning narrow Kerry deficits into victories in counties like Lake.

6:44 PM  
Blogger Mark said...

Mr. Phips, in Kentucky I suspect Obama will carry Jefferson County and no place else. McGovern may have won the eastern Kentucky coal counties by 20 points, but I'm highly skeptical that a black man will. This is the most racist part of the country. I read a news article during primary season where even the Pike County Democratic Chair said he didn't think he would be able to vote for Obama if he was the nominee. Why? "Race. I don't want an African-American to be President."

In Floyd County, Obama got all of 5% in the May primary. The other 90+% didn't vote for Hillary because they love her so much. They voted against Obama. I suspect every eastern Kentucky county will go for McCain by double digits.

Your formula for Ohio is Obama's exceedingly narrow path to victory in the state. Black turnout in Ohio was already robust in 2004, so further increasing it will be a stretch. But winning Hamilton County, along with exurbs in Lake County and sections of Delaware County near Columbus, and narrowing the gap in Cincinnati's exurbs will all be required for an Obama victory in Ohio, softening the blow from Obama's larger margins of defeat in southern Ohio.

My prediction is that Athens County, OH, will be the only county in all of Appalachia that will go for Obama (unless you count Allegheny County, PA, as Appalachia.

8:43 PM  
Blogger Mr. Phips said...

Would you call Belmont, Monroe and Jefferson Appalachia? If he is losing the sixth Congressional district by ten or more points, which he will if he only carries Athens, he will need to make it up in the fourteenth district which includes the Cleveland exurbs.

9:34 PM  
Blogger Mark said...

Yes, I would consider Jefferson, Belmont, and Monroe Counties in Ohio to be Appalachia, but I think Obama will lose all of them. Obama will probably lose OH-06 by about eight points or so...and is also likely to get smashed by about 20 points in neighboring OH-18. It's not just OH-14 that he will have to make up the ground in, but also bolster numbers in places like OH-01, OH-02, OH-03, OH-15, and OH-16. It's a tall order, but not out of the question.

9:59 PM  
Blogger Mr. Phips said...

Yes, he has to win OH-01, OH-12, OH-15, and either win or come damn close in OH-14 and OH-16. You think that would do it?

10:28 PM  
Blogger Mr. Phips said...

Mark, here is what I think Obama has to do in all 18 districts to win the state.

OH-01: 52/48
OH-02: 40/60
OH-03: 49/51
OH-04: 35/65
OH-05: 39/61
OH-06: 46/54
OH-07: 45/55
OH-08: 38/62
OH-09: 58/42
OH-10: 55/45
OH-11: 80/20
OH-12: 52/48
OH-13: 55/45
OH-14: 50/50
OH-15: 55/45
OH-16: 50/50
OH-17: 60/40
OH-18: 40/60

If he can do this, he will win in a squeaker.

10:36 PM  
Blogger Mark said...

Mr. Phips, not a bad calculus....although I think Obama should be doing better than 10-point margins in OH-10 and OH-13....but question whether OH-12 and OH-15 will be quite as sweeping of Obama victories as you imply. Otherwise, your numbers look realistic....and could conceivably be good enough to score an Obama win. Still, it's odds-against...the proverbial "inside straight".

9:39 PM  
Blogger Mr. Phips said...

Mark, OH-10 and OH-13 are both districts with a good number of white, working class voters. Kerry won OH-13 56%-44% but Gore won it by just 53%-44% and it barely went for Dukakis back in 1988. Kerry won OH-10 58%-41%, but Gore won it by 53%-42% and I believe it actually narrowly favored Bush over Dukakis in 1988. Obama also lost badly to Clinton in these areas in the primary.

10:39 PM  

Post a Comment

<< Home