Sep 19 2008

Print this Post

Polling and Electoral College Prospects

Share to Google Plus

The debut of Obama post-partisanship earlier this year brought with it the promise of an end to the Red/Blue divide in America, and the possibility of a 50 state strategy for the Democrats.  However, September can be a cold shower for hope in an election year, and the broad Red/Blue divide remains, even though upsets remain possible.

Considering outlier outcomes, attention remains strong on Obama’s potential out west, with certain polling showing Democratic strength and trends in Colorado, Nevada, New Mexico and even Montana as potential keys to unlocking the Electoral College. But historically, this is less newsworthy than it seems.

In 1992 Clinton carried Montana and Colorado. In 1996, he carried Arizona. He carried Nevada in both elections, and New Mexico went Democratic three times between Clinton and Gore. Victories here for Obama would be significant but not groundbreaking. There is Democratic support out west for the right candidate.

The real story of upset, however, is actually brewing in Obama’s electoral back yard. While Ohio has been a traditional ground zero for presidential politics, both Michigan and Pennsylvania have been an essential part of the Democratic electoral coalition for 20 years. All three however, are showing unaccustomed competitiveness in a year when at least Michigan and Pennsylvania would have long since been put away.

To understand why, you need to review the results of the Democratic primaries.

Consider the exit poll results from the three states, showing Obama’s margin versus Clinton among key demographic groups:

Category Ohio
v. Clinton
v. Clinton
Michigan 1
v. Clinton
Male – 2 +2 –  8
Female -18 -18 -24
White -30 -27 -35
Black +74 +80 +68
> 50k -14 -8 -14
>100k -10 -10 -10
No degree -18 -16 -24
Catholic -30 -40 N/A

With the exception of African American voters, Obama lost to Clinton by double digit margins in almost every category, with particular emphasis on women, whites, working class voters and Catholics.

The loss is instructive when you remember that the Ohio and Pennsylvania came toward the middle and end of the Democratic primaries, at a time when Obama was seen as a likely winner, and still he was not able to capitalize on his momentum. This demonstrated not just a lack of support for Obama, but potential resistance to his candidacy.

Using this data as a base, we can imperfectly assess the impact of Obama’s primary performance by comparing his showing to that of Senator Kerry in the three states during the general election in 2004.

Category Ohio
Obama v. Kerry
v. Kerry
Michigan 2
Obama v.
Male +1 + 3 -5
Female -9 -13 -17
White -10 –  8 -13
Black +3 + 6 -7
> 50k -16 –  8 -17
>100k –  7 –  7 -9
No degree -10 -12 -18
Catholic –  8 -21 -21

Again, the data shows that Obama is under performing Senator Kerry by double digits in demographic groups essential to victory.

Comparing Obama’s performance to the actual two-party results of those three states in 2004 demonstrates key Democratic weaknesses and areas where Republicans can significantly build on past support and potentially win each of the states.

Category OH
Obama v. Kerry
v. Kerry
NET Oppty PA
v. Kerry

v. Kerry
NET Oppty MI
Obama v.
v. Kerry
NET Oppty
Male + 1 +  5 +4 + 3 +  3 0 -5 +  2 +  7
Female –  9 0 +9 -13 –  8 + 5 -17 –  7 +10
White -10 +10 +20 –  8 +10 +18 -13 +10 +23
Black + 3 -68 +71 + 6 -78 +84 -7 -79 +86
> 50k -16 -16 0 –  8 –  9 –  1 -17 -19 –  2
>100k –  7 –  3 +4 –  7 –  5 +  2 -9 –  7 +  2
No degree -10 0 +10 -12 –  8 +  4 -18 -0- +18
Catholic –  8 +10 +18 -21 –  2 +19 -21 –  1 +20

While the comparison is a snapshot in time, it still carries relevance given the razor’s edge competition between Obama and McCain in historically Democratic demographics. By that measure, the comparison offers a rough roadmap for Republicans to capitalize on potential weaknesses in the Democratic ticket, particularly, with women, white, working class voters and Catholics.

While it is true that a large portion of the Democratic base has moved to Obama since he has secured the nomination, the comparison matters in all three states because Hillary Clinton is remains a factor in the election. Consider the favorability ratings for the major candidates in comparison to Mrs. Clinton.

Candidate Ohio Michigan Pennsylvania
Obama 56 56 56
McCain 56 56 55
Palin 50 52 46
Biden 47 49 46
Clinton 59 58 55 3

Three facts jump out from the poll. First, in two out of the three states, Mrs. Clinton is more popular than the other four candidates running, and trails by a point in the third. Second, Mrs. Clinton is holds a double digit favorability edge over Democratic vice presidential nominee Joe Biden in all three states. And finally, while Obama and McCain have virtually identical ratings in all three states, Governor Palin is more popular than Senator Biden in two states and tied for support in a third.

This would suggest that the Democratic ticket would have been stronger with Mrs. Clinton on it, that these states would be locked down now, and further that in at least some fraction of the electorate in these states, the addition of Governor Palin is helping Republicans in the absence of Mrs. Clinton.

But ultimately, why is any of this important?

Simply put, these three states represent a huge McCain opportunity. With memories of the primaries and Mrs. Clinton still firmly entrenched, support for Obama among core Democratic constituencies remains soft.

By the numbers, nailing down these states could assure McCain victory.

Consider that if McCain were to win these three states (and hold Florida as most polls indicate), the Republicans could lose Nevada, Colorado, New Mexico, Montana, Virginia and New Hampshire and McCain will still win, 289-249 in the Electoral College.

Realclearpolitics.com’s composite state poll averages (including the Battleground poll) has Obama leading in PA and MI, but only by half the margin that Kerry won by in 2004 and well within the margin or error of the polls.  McCain leads in Ohio by 6/10s of a point less than Bush won by in 2004.

If McCain can lock advantage in two out of the three states, he is in the pole position for the election.

If he pulls off a hat trick, he can’t lose.

High commands, take note.

1. Barack Obama’s name did not appear on the Michigan primary ballot and results are assessed from those that voted “Uncommitted”.

2. Barack Obama’s name did not appear on the Michigan primary ballot and results are assessed from those that voted “uncommitted” and are imperfect as a result.

3. Big Ten Battleground Poll results September 18, 2008.


1 comment

  1. Geraldine

    Good job mankig it appear easy.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

Couldn't connect to server: php_network_getaddresses: getaddrinfo failed: Name or service not known (0)