This is a part of a series of posts analyzing the 2010 midterm elections. This post will analyze the Pennsylvania Senate election, in which Republican Pat Toomey won a narrow victory over Democrat Joe Sestak in a Democratic-leaning state.
Pennsylvania’s Political Structure
This map, modified from the New York Times website, provides a very useful visualization of the election. Democratic strength in Pennsylvania is very concentrated. The black vote helps Democrats win Philadelphia (by an enormous margin) and Pittsburgh (by a lesser one). Working-class whites in places like Erie, Scranton (which is the blue dot at the top-right corner of the map), and southwest Pennsylvania also generally vote Democratic. Or they used to, at any rate. Finally, wealthy whites in the suburbs of Philadelphia and the LeHigh Valley are also voting increasingly Democratic.
Republicans, on the other hand, generally win everywhere else. They are dominant in rural, conservative central Pennsylvania and the exurbs of the Philadelphia metropolis.
A strong Democrat will win all the areas of the Democratic base and then expand to win areas of the Republican coalition. Here is President Barack Obama, for instance:

Mr. Obama doesn’t just win the Democratic base, he does quite strongly in the exurbs of Philadelphia. Notice how much better he does in the Republican stronghold of Lancaster County (the biggest red circle in the first map) than Mr. Sestak does.
A strong Republican candidate, on the other hand, will win all the areas of the Republican base and then expand to win areas of the Democratic coalition. Republican Governor Tom Corbett, for instance, actually won Allegheny County, which Pittsburgh is located in.
Republican Senator Pat Toomey didn’t do so well. He won the Republican parts of Pennsylvania, but lost the Democratic parts of Pennsylvania. In normal elections, when this happens the result looks something like this:

This is the 2004 presidential election, in which Senator John Kerry barely won Pennsylvania. He did this without making any gains into Republican Pennsylvania. The Democratic parts of Pennsylvania just barely outnumber the Republican parts of Pennsylvania, which is why Pennsylvania is a Democratic-leaning state.
In 2010, however, Mr. Toomey – riding on a strong Republican wave – was able to overwhelm the Democratic parts of Pennsylvania. Mr. Toomey was able to squeeze enough blood out of the Republican exurbs and rural counties to win.
This is a fascinating result because it doesn’t happen that often. More often the result looks like 2004. The 2010 Pennsylvania Senate election thus constitutes a model of a Republican overwhelming Philadelphia and Pittsburgh without making many gains into Democratic territory.
Comparisons
Let’s compare Mr. Toomey’s performance with Mr. Obama’s performance:
As this image shows, there was a very uniform shift rightwards from 2008 to 2010; almost every county moved Republican by double-digits.
There are some interesting subtleties here. The Republican exurbs of Philadelphia, where Mr. Obama did so well, snapped back very strongly rightwards. On the other hand, Mr. Sestak actually did better in parts of southwest Pennsylvania – a Republican-trending region which was particularly uninspired by Mr. Obama.
There is an economic dimension to this. Republican Pat Toomey ran a campaign based on themes, such as free trade, which appealed more to well-off voters. Democrat Joe Sestak, on the other hand, ran a campaign based on more populist themes. We thus see Mr. Toomey doing particularly well in the rich parts of Pennsylvania, such as the LeHigh Valley or Lancaster County. Conversely, he actually did a bit worse than Senator John McCain in the poorest parts of the state: the Appalachian southwest and the city of Philadelphia.
Conclusions
Throughout the entire campaign, Democratic candidate Joe Sestak polled considerably behind Republican Pat Toomey. It was only at the end that he started catching up, as Pennsylvania’s Democratic nature asserted itself. However, Mr. Sestak couldn’t quite make it all the way; the Republican wave in 2010 was just too strong.
All in all, these results were very “normal.” This is in the sense that both candidates built very normal coalitions; neither did well in places Republicans or Democrats don’t usually do well in. The state itself shifted fairly uniformly from 2008. No one place behaved like an outlier (unlike the case with other states).
The 2010 Senate election thus constitutes a perfect example of just what a narrow Republican victory in Pennsylvania looks like.
–inoljt





30 Comments

Electronic voting machine results don’t amount to a hill of beans but they make a colorful map.
You can’t justify ,without diedold, the republicans over came a 4 to 3 registration disadvantage to completely control the government.
What ever you’re describing isn’t an honest vote counting system and to spout and believe change is coming through the black boxes is delusional .
Well, I think this is a good political science analysis. Just one thing. How much did the vote totals go down and where? I bet those Philadelphia exurbs you talked about, as well as Pittsburgh, had dramatically lower voter turnout than they did in 2008.
And the reason for that was disillusionment with Barack Obama. Plain and simple.
From the same source (NYT) as the bubble map:
2008
Obama 3,192,316
McCain 2,586,492
2010
Toomey 2,028,945
Sestak 1,948,716
The Republican total dropped 22.6%, about half a million votes. Democratic total dropped 39%, about one and a quarter million.
“…the Republican wave in 2010 was just too strong.”
This is just incorrect. You can’t claim a Republican “wave” when votes for Rs drop by 22%.
THANK YOU! Republican wave, my ass. The Democratic Party must be destroyed. All we have to do is not vote for the corporatist bastards.
Exactly, the Dem base stayed home. The Republicans didn’t win; the Democrats *lost*. There’s a subtle, but important distinction to make.
Another important fact was left out of this race: Arlen Specter.
If Specter beat Sestak in the primary, he probably would have won the seat.
Specter was an incumbent (the “anti-incumbent” electorate mood was a myth too), who switched from D to R. He could have picked up substantial R support; he was also better funded (he had union support, NAACP support, etc. He was the establishment D candidate, believe it or not: http://www.youtube.com/watch?v=hs0cvwDLjow)
I don’t care for Specter, but primaring him was pretty dumb.
“…who switched from D to R.”
Specter was a longtime REPUBLICAN who switched from R to D to avoid a primary loss to Toomey.
The “establishment” that favored Specter was mostly Harry Reid, BO, and Bill Clinton, as a reward for switching when the Senate Democrats were shy of 60. Among Democrats, however, Specter was known for “always being there when you didn’t need him.”
Sestak lost a very close race partly because he voted with BO 98% of the time.
“The Democratic parts of Pennsylvania just barely outnumber the Republican parts of Pennsylvania”
Registration in PA is about 4 million D, 3 million R, and 1 million other. Is a one million, one out of eight edge “just barely”?
“Lost” is absolutely correct.
Examine the numbers for Philadelphia:
2004: Kerry 542,205
W 130,099
2008: Obama 574,930
McCain 113,260
2010: Toomey 66,617
Sestak 350,805
Sestak lost the state by about 80,000 votes, but he ran weaker than Obama’s 2008 total in the Philadelphia Democratic stronghold by 224,000, even after heavy campaigning by America’s first black President, Bill Clinton.
This is more evidence – as if we don’t have enough already – that Obama has very weak support in the Democratic Party.
It is also evidence that you can’t analyze an election by using bubble maps. Much of the story is in the numbers.
I highly doubt that there was voter fraud in this election to any extent that it affected it.
It takes a tremendous amount of effort to get voter fraud right. Given the enormous amount of data – to the extent that we can drill down to how groups of hundreds of people voted – it’s almost impossible to commit fraud without looking suspicious. If you have the racial and income data for entire state of Pennsylvania for every group of several hundred people – and we DO have that data – those groups will vote in very predictable patterns. For instance, I would be highly suspicious of any precinct more than 20% black which voted against the Democrat. Anything that doesn’t match will automatically be extremely suspicious. We would easily know if there was fraud.
Well…every midterm is going to have lower turn-out. I don’t put too much emphasis on that.
Democratic registration doesn’t matter. Kentucky and North Carolina have more registered Democrats than Republicans (maybe even Mississippi too). Doesn’t mean that they’re Democratic states.
Meh…I doubt that Specter would have won. The flip-flop narrative would have killed him.
No, you wouldn’t know easily and this rot goes to the roots at the county level.
Why did Germany outlaw these corruptible black boxes ? Because an eighth grader couldn’t duplicate the results. and if that’s not possible then it’s not an honest vote.
How do they vote, Diebold ?
Optical scans are no better but believe the black boxes over exit polls , come on .
Specter was for the public option:
http://www.huffingtonpost.com/2009/10/22/arlen-specter-i-unequivoc_n_330959.html
Specter is a soulless politician who will change his morals on a dime to get re-elected. He actually might have been a decent Dem if pushed hard enough, though I understand the skepticism. It’s more than I could say for Obama.
what explains the difference in turnout?
If Obama used Fracking as an issue to scare homeowners and farmers he could win the rural areas…if only he had a Spine.
ES&S and Diebold both have issues with integrity (Thanks John Kerry and Democrats)
However, another point is also relevant. In 2006 Carl Romanelli was the Green Party candidate for US Senate in Pennsylvania. Both Carl and Ralph Nader were actively dissuaded to not be allowed on the ballot. By legal and extra-legal means. Carl, an ordinary working class guy faced tens of thousands of dollars in legal fees and fines. Simply because he was looking for the right to run for office.
So trying to blame this on the evil Republicans does not fit. Many people in this state recognize that there is no democracy there.
Just like the rest of the country.
Congrats on getting front paged Inoljt:) I have a feeling your election diaries will be useful and more in demand soon.
Change you can believe in…clean underwear.
The two party system is a scam.
James Carville famously described Pennsylvania as Philadelphia and Pittsburgh with Alabama in between.
Look. some nice work here but I can’t ignore:
“the rich parts of Pennsylvania, such as the LeHigh Valley or Lancaster County”
Neither are rich areas, I don’t know what you are thinking. The Lehigh Valley is dominated by Allentown, the eastern end of the rust belt. Lancaster is known as the home of the Amish. The “rich” areas of PA (it’s relative) are the burbs of Philadelphia and to a lesser extent Pittsburgh.
Diedold, the friendly, dependable fall-back for every election, is the only explanation of how these GOP clowns can be so arrogant, sexist, racist and plain stupid all the way through election night without ever sweating. A friendly county clerk doesn’t hurt either.
Don’t worry. He’ll come to frackland in Pennsylvania — after a friendly visit to Wisconsin — if anyone would just send him a comfortable pair of walking shoes.
After the 2000 election: Although the senate is 50/50, Republicans can pass legislation because Cheney, as VP, can cast the tie-breaking vote.
After the 2008 election: Although the Democrats have 59 votes in the senate, they still need 60 to pass their legislation – to overcome the filibuster.
Democrats, the media & Republicans all adopted these two narratives.
The first thing Obama did after the 2008 elections was to dismantle the new army of motivated volunteers – which was his only leverage if he intended any Change.
Every move president Obama made served the elite interests and demotivated the new Democratic base.
You could see the relief in Pelosi’s face when Republicans took control of the House in 2010 – everything back to normal – Democrats could once again vote for good bills that don’t pass and vote against bad bills that become law.
The Democratic machine in PA & Obama could have gotten the vote out, but they did not want to win.
Jim Hightower said that Rick Perry won Texas with only 19% of the voting age population. For this to happen the Texas Democratic party must be incompetent or trying to loose.
http://www.counterpunch.org/2010/10/29/our-third-karpian-moment/
Inoljt,
Since your are looking at elections past, any comments on this analysis of the 2004 presidential election:
Urban Legend, Election 2004 vs. the White Ghosts of NYC, by Michael Collins
http://www.globalresearch.ca/index.php?context=va&aid=5998
http://www.scoop.co.nz/stories/HL0708/S00284.htm
Bush went from 2.3 million big city votes in 2000 to 5.4 million in 2004, a 153% increase. Bush’s rural white base abandoned him and urban whites won the election for Bush.
Consider:
1. Rural citizens gave Bush 2.5 million fewer votes in 2004 than they contributed in 2000.
2. The NEP showed that black, Latino, and Jewish city voters went for Kerry at rates of 95%, 66%, and 80%. Also, whites in big cities are the most liberal group of whites in the nation.
3. The new-voter findings nationwide showed a 3 to 2 advantage for Kerry and that 40% of the voters in the big cities were new-voters.
4. Bush did not campaign or advertise in big cities and doubled his big city vote count.
NBC Makes Unprecedented Downward Correction in Latino Support for Bush
http://www.hispanicbusiness.com/news/2004/12/3/nbc_makes_unprecedented_downward_correction_in.htm
Corporate media told us the shift of Hispanic voters toward parity was one of the most important pillars of the Bush “victory” in 2004. But Latino voters supported democratic presidential candidate John Kerry over President George W. Bush by a margin of 65.4% to 33%.
It appears there is much to reconcile in the 2004 election and the corporate media and the Democrats are silent.
Test “Santorum”test
Test Santorum