Talk Elections

I have a copy of How Barack Obama Won (Chuck Todd and Sheldon Gawaiser).  An excerpt from their chapter on Pennsylvania:


This analysis seems wrong in several respects.  First, Pennsylvania doesn't have anywhere close the the educational attainment levels of New York, New Jersey, New England and Maryland - so either there was a huge skew towards college graduates voting in PA in particular that year or these exit poll surveys include a fair number of people who don't actually have 4-year degrees.  Also, it seems to downplay Obama having a pretty solid WWC vote in PA.

Economic collapse muted the display of Mccain's electoral strength. All pre-Lehman polls indicated Mccain to be a stronger candidate in swing states than Romney 2012.

Romney's economic positions actually hurt him.

https://www.youtube.com/watch?v=hF9ndn-R1_I&t=126s

This is the day Lehman collapsed (so polls before this day had to be compiled for this map). Mccain was doing far better than Romney ever did in 2012 at this point.

McCain got a huge jump in the polls because of the Palin pick. Then she opened her mouth.

Palin was highly scripted. No different from Rubio, but she was a VP, all she needed to do was stick to the script. Financial crisis complicated the issue, because it was a dynamic situation for which the Republican establishment did not have an established talking point for (initially some opposed, others supported TARP, until they eventually formally endorsed TARP in unison). Rubio had similar troubles in non-scripted settings.

Even in the period following the first 2012 debate? 

Have you seen Game Change (the movie)?

Same as Rubio. Rehearsed, Scripted, did everything per handler recommendation.

What's interesting to me is this analysis suggests that a WWC-populist campaign could never work because too much like NY/NJ/MD, Philadelphia suburbs or whatever.

Yep. Romney was still losing Nevada and Ohio (two states Mccain was leading in), while being tied in Virginia, Colorado, Florida (Mccain doing a bit better in all of these states, with florida leaning for Mccain). Oregon and Washington were never even competitve for Romney (Mccain had closed to single digits), and Mccain was polling better in MN, MI, and even WI. Romney was never ahead of Obama, even at this best.

The final results of 2008 make it easy to forget what a close, intense election it was for much of the year.

It also allows people to forget how much of a polarizing figure Obama was. Hillary actually polled better than Obama swing-state wise.

https://www.theatlantic.com/politics/archive/2008/05/clintons-closing-argument-to-superdelegates/53314/

Obama wasn't picked for electability, he was picked for running a polarizing ideological campaign.

I agree with this post. Without the financial crisis McCain could easily have squeaked a victory on the back of a Palin-inspired culture war victory as Trump managed in 2016.



This is a bit exaggerated though. Really, Obama was picked for running a (deliberately non-polarising) "let's make America post-racial" campaign. Clinton polled better in the swing states because she had more support among the WWC demographic which is highly concentrated there (as we saw in 2016), not because Obama was polarising. In fact, his fav/unfav (http://news.gallup.com/poll/111838/obama-bush-contrast-popularity.aspx) during the campaign were stable in the low-60s/mid-30s. I also found this (http://www.politifact.com/truth-o-meter/statements/2008/feb/11/barack-obama/clintons-negatives-are-higher-than-obamas-/) article from the primary about Obama bringing up Clinton's relatively high unfavourability and saying she "starts off with 47% of the country against her" (sound familiar?).

Though, in retrospect, I think things would have worked out better for the Democratic party and America if Clinton had won the nomination in 08. President Hillary could have avoided the whitelash Democrats suffered under Obama and also probably would have handled healthcare with more experience, then handing the baton to Obama's unifying post-racial message in the more 'woke' America of 2016.

He was more polarizing in Middle America (and especially the South) for sure. That's how the Blue Dogs were wiped out, that event begat the Tea Party. He won by pandering to the Left, and used the anti iraq war card to make an argument to overturn the foreign policy consensus.

https://www.youtube.com/watch?v=H6f4tZFZ_-g

I still believe that in the end, the best McCain could have done was a 1968 Humphrey or 1976 Ford style narrow loss even without the financial crisis.  The fundamentals were still horrible for McCain and much worse than they were for even Humphrey and Ford.  Bush's approval ratings were in the 20s and the economy was clearly in recession territory.  Unemployment rose from around 4.4% in summer 2007 to 6.2% in late summer 2008.  No candidate has ever won with the fundamentals so much against them.

Fundamentals were pretty horrible for Truman in 1948.

One of McCain’s bigger mistakes was leaving Obama’s biggest scandal on the table until the very end. If he had hammered on Reverend Wright from the get-go I think he keeps IN/NC and puts OH into squeaker territory, even in OTL.

The economy had fantastic economic growth for most of 1948 with a huge increase in real disposable income.  The economy didn't start sliding into recession until late Fall, well too late to start having an impact on the election.

Apparently, the fundamentals were stacked against GWB, how did he do?

Someone said that had the economic collapse not happened, Obama's 2008 win would resemble his 2012 win - 4 popular vote lead, and the 2012 state margins.

That sounds about right. Palin could have been a huge boon if she was better prepped. She truly was the star of the 2008 election - even over Obama. No one cared about McCain.

Well he did lose the popular vote and it was the closest electoral college result in history. He only won Florida by the skin of his teeth (10,000 votes).

Would've been far closer than 2012. Before Lehman collapsed, not a single poll had Obama up in FL, VA, etc., in contrast, Hillary easily beat Mccain in FL and was competitive in a number of southern states like AR.

There were plenty that had Obama ahead of McCain in Virginia before 9/15.

https://www.realclearpolitics.com/epolls/2008/president/va/virginia_mccain_vs_obama-551.html#polls

Was exaggerating, but if you look at the trending of numbers before 9/15 overall Mccain looked more like Obama did in 2012 in those states than like Romney.

For sure, had Romney been the nominee in '08, he would've been wiped out.

Obama was struggling in the polls in Virginia and Florida leading up to election day 2012.

He would have separated himself in 2008 as well. Same margins as 2012. No one cared about McCain and Palin was a flop after initial excitement.

Obama would have won Virginia 3-5 points in 2008 and Florida by 0.5-1.5.

You're talking about last minute partisan surges, which happens in all normal elections, and would've happened for Mccain too, if not for the collapse. That would've put him over the top in a number of states. Like I said, Palin was no different from Rubio, she had a script, she wouldn't have needed to have done anything else other than read off it under normal conditions. Put Rubio in the same position, and he would've done the exact same things as Palin.

Less people were excited about Romney. Mccain at least had some centrist appeal.

The most interesting thing about this thread is that it doesn't really talk much about the state in question but rather about generalized situations regarding the campaigns overall in 2008.

Another mistake made in this thread is thinking of Eastern PA as a single unit. It is better to think of PA in 6 Parts (NW, SW, NE, Central, Philly Burbs and Philly itself). The NE part of the state functions more like the Western Part of the state and has similar demographics and political inclinations on issues like trade, energy and immigration). The clear difference though is coal mining in NE died in the 1950's and 1960's whereas the decline in the SW is more recent, hence why Romney built on McCain's trends in the SW but didn't get anywhere in Luzerne. War on Coal wasn't going to flip NE PA, trade and immigration would and did.

Here is the obvious answer as to why McCain lost PA. McCain is a sunbelt neocon who was running to be the third term of a sunbelt neoconservative Presidency in the midst of an economic meltdown that was being pinned on the Republicans. McCain is a great candidate for Florida, for Virginia even, and also for the Southwestern States. But he is a terrible candidate for the rust-belt.

1. The Economy. No candidate running could have erected enough distance to separate themselves from the recession and the economic collapse.If the candidates were Huckabee and Clinton, Huckabee could have tied the economic meltdown to the bills that Bill Clinton signed in the late 1990's. But Huckabee would have suffered for other reasons and probably would not have won.

2. McCain was moderate on the wrong issues to win this election. McCain was conservative in that he was pro-life and generally in favor of low taxes. But the specific issues that would have differentiated him with Obama and gave him an advantage in appealing to working class swing voters, energy, trade and immigration, he was for the most part in agreement with Obama on. He did flip flop on offshore drilling when gas prices peaked and that contributed to a substantial boost for McCain in that August surge that he experienced. But McCain was fundamentally the wrong candidate to flip PA.


3. McCain did make strides and even won the old PA-12, which was the only Kerry-McCain district, in western PA. But it was small potatoes compared to what Trump got in the counties bordering WV. If you cannot win the Philly burbs, then you need to counter them with running up the margins in all the rural counties and neutering Dem margins in places like Erie and Lackawanna, which Trump did (he even won the former).


At the end of the day, and the article in question did this as well, there was an assumption that there was a upward limit on the Republican candidate in "out-state PA". This was based on several presumptions about Republicans on issues like trade as well as about the voters themselves. These presumptions narrowed the path to victory and said you had to win all four of Chester, Bucks, Montco and Delco, to win the state as a Republican. The last Republican to do so was Arlen Specter in 2004, and no Republican has done so since to my knowledge.

Since McCain's redneck PA attempts failed to deliver the state, it served as confirmation bias. The problem was McCain is a poor fit for blue collar PA, because he couldn't connect with them on economic issues and worse still he had the Bush era economic and economic collapse weighing him down and little to no attempt was done to defend the Republican brand in terms of the economic collapse. A frequent critique of the Bush years in many ways and it is a miracle that the GOP was able to recover so quickly after 2008, and that was largely by embracing the Tea Party, though such came with a price.

Palin was different from Rubio in that she was a phenomenon. Her acceptance speech got nearly as many views as Obama's did (virtually tied). Her debate with Biden was the highest rated debate (70 million viewers) since Reagan vs. Carter.

Palin had almost as much online interest as Obama did AFTER Obama was declared the winner. To compare how incredible that is, Hillary Clinton didn't even get near Trump after the 2016 election. Remember that Hillary Clinton is also a controversial/infamous figure in American politics.

2008 was Palinmania. If the economic collapse never happened and Palin was better prepared, then perhaps McCain could have made it a close race.

But McCain was doomed due to George W. Bush.

Even without the crash McCain , McCain still loses with this map:


()

Great analysis. I was amazed by Obama's 10.3% margin win in PA in 08. It was the first double digit win in that state since Nixon 1972. Not even Reagan 1984 did as well. And Obama did better in PA than in any Midwestern state except his home state of IL.

WI, MI and MN were all more Democratic than PA.

Huge brain fart. Yeah, Obama won WI by 14, MI by 16. He won MN by 10.2.

Inspired by the recent state elections in PA, I spent a while today staring at the county results for PA in the 2012 and 2016 elections. I figured I might as well use this old thread to continue discussion on PA. Here's my take on the state in 2016:

I roughly divided the counties into three groups: major metro areas (Philly, Pittsburgh, and Pike County, which is part of the NYC metro area and is so small that it doesn't really affect results anyway), the remaining metro areas (as defined by the US Census), and all other counties.

Major metro areas (~1/2 of total vote):
Summary: Compared to 2012, Democrats actually did pretty ok here due to gains in Philly suburbs
2012: Obama 59-Romney 40
2016: Clinton 58-Trump 39

Between 2012 and 2016, Democrats made significant gains in just three PA counties: Allegheny, Chester, Montgomery; Allegheny contains the city of Pittsburgh, while the latter two counties are in the Philly metro area. One other thing that stands out about all three counties is that they're the only counties in the state where Trump received significantly fewer votes than Romney had (he also finished slightly behind Romney in Delaware County); it seems pretty likely that Romney-Clinton voters were important here.  Democrats lost a fair amount of ground in the city of Philadelphia, but ultimately it wasn't that big of a deal compared to other shifts in the state, and Democrats still won by a margin of 475k votes in the city (compared to 490k for Obama). Bucks and Delco, the other two counties in SEPA, didn't see a major movement towards either party. Overall, Democrats actually improved on their performance in the Philly metro area by just under 50k votes.

The gains in Allegheny (which has about 4x the population of the city of Pittsburgh) were overwhelmed by significant losses in every other county of the Pittsburgh metro area; Obama won 40% in three of the non-Allegheny counties in the Pittsburgh metro area, while Clinton didn't win 40% in any of them. The most important of these counties, Westmoreland, went from 38% Obama to 33% Clinton. The Republican margin in the Greater Pittsburgh area increased from 14k votes in 2012 to 58k votes in 2016, the third-largest gain either party made in any single metro area (after Scranton-Wilks-Barre and Greater Philadelphia).

Despite overall losses in the Pittsburgh metro area and a decline in the city of Philadelphia, the gains in Allegheny, Chester, and Montco left the Democratic margin in the metro areas unchanged: they won them by about 600,000 votes in both 2016 and 2012. With that said, one could make the argument that Democrats should have made more gains here, especially in Philadelphia. And while every county in the major metro areas saw an increase in turnout (http://[url=https://www.washingtonpost.com/graphics/politics/2016-election/precinct-post-analysis/), so too did every other county in the state, and as a share of Pennsylvania's vote the major metro areas declined from about 54% in 2012 to about 53% in 2016.

Other metro areas (~1/3 of total vote):
Summary: Democrats held their own ground in a few metro areas, but Republicans made major gains overall
2012: Romney 52-Obama 47
2016: Trump 55-Clinton 40

I'm going to divide these into four categories:

Places Democrats largely held ground: Democrats actually gained ground in Centre County (home to Penn State). Clinton also did pretty ok in two of the three metro areas of South Central PA (the exception being the the York-Hanover MSA), as she nearly matched Obama's 40% in Lancaster County, and the Dem share declined from 45% to 43% in the Harrisburg-Carlisle MSA. Overall, the Republican margin increased by about 10k here compared to 2012.

Moderate Republican gains: The Dem share fell from 52% to 47% in the Allentown-Bethlehem MSA (aka the Lehigh Valley), from 49% to 43% in Berks County, and from 39% to 33% in the York-Hanover MSA. The Republican margin increased by about 60k votes here.

Major Republicans gains: Obama won over 55% of the vote in both the Erie MSA and the Scranton-Wilks-Barre MSA, and Clinton declined by at least 10% in both MSAs. She also declined by 12% in Mercer County, a part of the Youngstown, OH MSA that Obama had nearly tied Romney in. Of all the PA metro areas, Scranton-Wilks-Barre saw the largest losses by any party between 2012 and 2016; Democrats went from winning it by 30k votes in 2012 to losing it by a little less than 30k votes in 2016. Overall, the Republican margin increased by about 90k votes in this category.

Smaller MSAs: Obama won 40% or less in the Johnstown MSA, the Altoona MSA, the Lebanon MSA, and the Williamsport MSA, and Clinton did significantly worse in all four. Each MSA cast less than 100k votes in 2012, but collectively the Republican margin from this category increased by about 30k votes.

One thing I noticed is that the total number of Republican voters increased in each county, while, with a few minor exceptions, all of the counties saw a decline in the total number of Democratic voters. Overall, the Republican margin of victory in these counties increased from about 100k votes in 2012 to about 300k votes in 2016. So Democrats won the metro areas (major and otherwise) by about 500k votes in 2012, but by just 300k votes in 2016.

Rest of state (~1/6 of total vote):
Summary: Democrats went from bad to worse
2012: Romney 61-Obama 37
2016: Trump 68-Clinton 28

These counties pretty much all tell the same story: Democrats went from bad to worse. The one exception is Monroe County, which went from a 56-43 Obama victory to a 49-48 Clinton victory. However, Monroe is in the eastern part of the state and is surrounded by MSAs, and (even from my own personal experience) it really seems to fit better in the "other metros" group. At least in this exercise, its voting behavior looks a lot like the neighboring Allentown-Bethlehem MSA.

Obama won between 40% and 45% of the vote in Columbia, Schuylkill, Clinton, Elk, Greene, Lawrence, and Warren counties. Not coincidentally, all of these counties are part of micropolitan statistical areas, meaning that they have a core urban area of at least 10k but less than 50k. Clinton's share of the vote declined by at least ten percent in every single one of these counties. Schuylkill, one of the three most populous counties in the non-metro group, stands out as the worst of these counties for Democrats; Obama won 43%, while Clinton won just 27%. Overall, the Republican margin in these counties increased by about 50k votes compared to 2012.

Obama won less than 30% of the vote in Juniata, Mifflin, Bedford, Fulton, Jefferson, Potter, and Somerset. Clinton declined by at least 5% in each one. Fulton was Obama's worst county (he won 21%) and Hillary's worst county (she won 13%). Even though all these counties are quite small, the Republican margin increased by 20k votes.

In all but one of the 21 remaining counties, where Obama won between 30% and 40% of the vote, Clinton's share of the vote declined by at least 5%. The one exception was Union County, where Trump actually won a smaller share of the vote than Romney did. This seems odd, especially since the biggest Democratic decline in this category happened in neighboring Northumberland County; the one thing that does stand out about Union County is that it's home to Bucknell University, a small but fairly prestigious liberal arts school. Anyway the total Republican margin in these counties increased by 85k votes.

Much like the mid-sized metros, every single county saw an increase in the total Republican vote, and all but one county (the mysterious Union County) saw a decline in the Democratic vote. So, even though Republicans were already winning all but one of these counties by fairly large margins, and proportion of the vote cast by these counties did not increase dramatically, Republicans went from winning these counties by ~200k votes to winning them by ~350k votes. Add up the 600k Democratic margin in major metro areas, the 300k Republican margin in other metro areas, and the 350k Republican margin in rural areas, and you get a 50k Republican victory (the exact Republican margin was 44,292 votes).

TLDR: Democrats lost ground in the city of Philadelphia, but more than made up for those relatively minor losses with gains in Chester County and Bucks County. However, they lost a significant amount of ground in most other metro areas, most dramatically in the Scranton-Wilks-Barre area and Greater Pittsburgh. Those losses, combined with Trump running up the score in non-metropolitan area counties, led to a Trump victory.

I took a look at Casey's 2018 victory compared to 2016. Obviously the landslide re-election of an incumbent senator in a mid-term year isn't directly comparable to a presidential election, but I thought it could shed some light on how Democrats could win PA in 2020. I figure Democrats probably have five main ways of improving in 2020:

1)Run up the score in Philadelphia
2)Win the Philadelphia suburbs more decisively
3)Improve in Greater Pittsburgh
4)Improve in the remaining metro areas
5)Limit losses in the rural areas

In 2018, Casey:

1)Actually benefited less than Clinton had from Philadelphia. However, this makes sense in a mid-term, since lower statewide turnout means that both parties benefit a little less (in terms of total margin of victory) from decisively winning counties. And Casey won 86.8% of the vote in Philadelphia, better than Obama did in 2012 or 2008.
2)Made major gains in the Philadelphia suburbs. Casey improved by his share of the vote by 5-8% of the vote in all four counties, perhaps most notably in Bucks, which went from being a rough tie between Clinton and Trump to a 56/42 Casey win.
3)Made major gains in every single county in the Pittsburgh MSA, most importantly in Allegheny County. Allegheny County went from a 15-point win to a 30-point win, Beaver went from a 20-point loss to a tie, and every other county in the region saw a 10 to 15 point Democratic gain between Casey and Clinton. Given that Obama '12 didn't even come close to the margins Casey won here, this strong of a performance is probably unsustainable, but it still points to the possibility that Democrats could bounce back a bit here.
4)Saw a lot of improvement in mid-sized metro areas, gaining 5-10 points in pretty much every county in this category. He won several of these counties, and improved by at 10 points or more in the traditionally-Democratic counties of Erie and Lackawanna, as well as in Centre County. His improvements in York and Lancaster counties also stand out in that they limited losses in two of the best counties for Republicans. Except for Lackawanna (which Obama won by 63%), Casey outran Obama in pretty much all of these counties.
5)Improved on Clinton's share of the vote by at least 5 points in all of the rural counties, but in many cases his improvement was closer to 10 or even 15 percent. Unlike Clinton, in only a handful of counties did he fail to take at least 30%. Nonetheless, he still ran behind Obama in a lot of these counties.

W Bush won Florida by 537 votes in 2000. In 04, he crushed it, winning the state by 5%.