Interesting exit poll trend data

[JA]

Difference between Democratic margin of the vote among male voters and the national popular vote (Election margin/male margin)

1972 | -02% (-23%/-25%)
1976 | 00% (+02%/+02%)
1980 | -09% (-10%/-19%)
1984 | -07% (-18%/-25%)
1988 | -09% (-07%/-16%)
1992 | -03% (+06%/+03%)
1996 | -09% (+08%/-01%)
2000 | -11% (00%/-11%)
2004 | -08% (-03%/-11%)
2008 | -06% (+07%/+01%)
2012 | -11% (+04%/-07%)
2016 | -14% (+02%/-12%)

Takeaway: The gender-gap is most prenounced when a new Republican is running (1980, 1988, 1996, 2000, 2012, 2016) with the exception of 2008. 2016 also saw the largest gender gap and it corresponded with the Democrats nominating the first female Presidential nominee of a major party.

Difference between Democratic margin of the vote among female voters and the national popular vote (Election margin/female margin)

1972 | +01% (-23%/-22%)
1976 | 00% (+02%/+02%)
1980 | +08% (-10%/-02%)
1984 | +06% (-18%/-12%)
1988 | +06% (-07%/-01%)
1992 | +02% (+06%/+08%)
1996 | +08% (+08%/+16%)
2000 | +11% (00%/+11%)
2004 | +06% (-03%/+03%)
2008 | +06% (+07%/+13%)
2012 | +07% (+04%/+11%)
2016 | +10% (+02%/+12%)

Takeaway: Democratic and Republican incumbents tend to improve on their numbers among females in their re-election contests. The election with the largest deviation of female support for the Democrat was actually 2000, not 2016.

Difference between Democratic margin of the vote among White voters and the national popular vote (Election margin/White margin)

1972 | -11% (-23%/-34%)
1976 | -07% (+02%/-05%)
1980 | -11% (-10%/-21%)
1984 | -11% (-18%/-29%)
1988 | -12% (-07%/-19%)
1992 | -07% (+06%/-01%)
1996 | -11% (+08%/-03%)
2000 | -12% (00%/-12%)
2004 | -14% (-03%/-17%)
2008 | -19% (+07%/-12%)
2012 | -24% (+04%/-20%)
2016 | -23% (+02%/-21%)

Takeaway: The deviation between the Democrat's margin of the vote (ex: Obama in 2012 scoring 51% to 47% for Romney, meaning a margin of +4%) had remained fairly stable between 1972 and 2000/2004. It was not until Obama was the nominee that the White vote deviated substantially from the national popular vote, which increased in 2012 and did not decline in 2016. It is hard to imagine this not being related to Obama's racial identity.

Difference between Democratic margin of the vote among Black voters and the national popular vote (Election margin/Black margin)

1972 | +87% (-23%/+64%)
1976 | +65% (+02%/+67%)
1980 | +84% (-10%/+74%)
1984 | +99% (-18%/+81%)
1988 | +81% (-07%/+74%)
1992 | +67% (+06%/+73%)
1996 | +64% (+08%/+72%)
2000 | +81% (00%/+81%)
2004 | +80% (-03%/+77%)
2008 | +84% (+07%/+91%)
2012 | +83% (+04%/+87%)
2016 | +78% (+02%/+80%)

Takeaway: Black voters strongly disliked Reagan, Bush Sr., Bush Jr., and Trump, although Trump seems to have the smallest deviation for a successful Republican among Black voters.

Difference between Democratic margin of the vote among Hispanic/Latino voters and the national popular vote (Election margin/Hispanic margin)

1972 | +53% (-23%/+30%)
1976 | +50% (+02%/+52%)
1980 | +28% (-10%/+18%)
1984 | +43% (-18%/+25%)
1988 | +46% (-07%/+39%)
1992 | +30% (+06%/+36%)
1996 | +43% (+08%/+51%)
2000 | +27% (00%/+27%)
2004 | +12% (-03%/+09%)
2008 | +29% (+07%/+36%)
2012 | +40% (+04%/+44%)
2016 | +34% (+02%/+36%)

Takeaway: Republicans were make considerable inroads among Hispanic/Latino voters under President Bush, but have since returned to their more typical substantial losses.

Difference between Democratic margin of the vote among White colleged graduate voters and the national popular vote (Election margin/White college graduate margin)

1980 | -14% (-10%/-24%)
1984 | -07% (-18%/-25%)
1988 | -14% (-07%/-21%)
1992 | -07% (+06%/-01%)
1996 | -12% (+08%/-04%)
2000 | -08% (00%/-08%)
2004 | -08% (-03%/-11%)
2008 | -11% (+07%/-04%)
2012 | -18% (+04%/-14%)
2016 | -06% (+02%/-04%)

Takeaway: White college graduates have been a consistent Republican voting bloc. There was an enormous swing in their margins between their largest in 2012 and their smallest in 2016. There is no indication that Democrats are positioned to win this demographic in the foreseeable future.

Difference between Democratic margin of the vote among White without a college degree voters and the national popular vote (Election margin/White without a college degree margin)

1980 | -11% (-10%/-21%)
1984 | -14% (-18%/-32%)
1988 | -12% (-07%/-19%)
1992 | -05% (+06%/+01%)
1996 | -07% (+08%/+01%)
2000 | -17% (00%/-17%)
2004 | -20% (-03%/-23%)
2008 | -25% (+07%/-18%)
2012 | -29% (+04%/-25%)
2016 | -41% (+02%/-39%)

Takeaway: This group has been consistently moving hard away from the Democrats since 2000; a movement that shows no signs of ending. In fact, historically, Republican incumbents have improved upon this deviation in their re-election bid.

Difference between Democratic margin of the vote among Democratic party affiliated voters and the national popular vote (Election margin/Democratic party affiliated margin)

1972 | +47% (-23%/+24%)
1976 | +53% (+02%/+55%)
1980 | +51% (-10%/+41%)
1984 | +67% (-18%/+49%)
1988 | +72% (-07%/+65%)
1992 | +61% (+06%/+67%)
1996 | +66% (+08%/+74%)
2000 | +75% (00%/+75%)
2004 | +81% (-03%/+78%)
2008 | +72% (+07%/+79%)
2012 | +81% (+04%/+85%)
2016 | +78% (+02%/+80%)

Takeaway: Democratic affiliated voters became increasingly partisan under Reagan - a process that was only amplified during all subsequent elections, save the ones with Bill Clinton.

Difference between Democratic margin of the vote among Republican party affiliated voters and the national popular vote (Election margin/Republican party affiliated margin)

1972 | -62% (-23%/-85%)
1976 | -83% (+02%/-81%)
1980 | -67% (-10%/-77%)
1984 | -67% (-18%/-85%)
1988 | -76% (-07%/-83%)
1992 | -69% (+06%/-63%)
1996 | -75% (+08%/-67%)
2000 | -83% (00%/-83%)
2004 | -84% (-03%/-87%)
2008 | -88% (+07%/-81%)
2012 | -91% (+04%/-87%)
2016 | -85% (+02%/-83%)

Takeaway: Republicans have been almost consistently more partisan in their voting habits than Democrats. This partisanship has been increasing since the Reagan administration.

Difference between Democratic margin of the vote among Independent voters and the national popular vote (Election margin/Independent margin)

1972 | -03% (-23%/-26%)
1976 | -13% (+02%/-11%)
1980 | -15% (-10%/-25%)
1984 | -09% (-18%/-27%)
1988 | -05% (-07%/-12%)
1992 | 00% (+06%/+06%)
1996 | 00% (+08%/+08%)
2000 | -02% (00%/-02%)
2004 | +04% (-03%/+01%)
2008 | +01% (+07%/+08%)
2012 | -09% (+04%/-05%)
2016 | -08% (+02%/-06%)

Takeaway: Independents have been rather consistently inclined towards the Republicans, save for 1992, 1996, 2004, and 2008 when it was either tied or they were inclined to vote Democratic. This collapsed in 2012 when their deviation fell from +01% in 2008 to -09% in 2012.

Difference between Democratic margin of the vote among Protestant voters and the national popular vote (Election margin/Protestant margin)

1972 | -19% (-23%/-42%)
1976 | -13% (+02%/-11%)
1980 | -14% (-10%/-24%)
1984 | -15% (-18%/-33%)
1988 | -12% (-07%/-19%)
1992 | -13% (+06%/-07%)
1996 | -13% (+08%/-05%)
2000 | -14% (00%/-14%)
2004 | -16% (-03%/-19%)
2008 | -16% (+07%/-09%)
2012 | -19% (+04%/-15%)
2016 | -21% (+02%/-19%)

Takeaway: The Protestant vote has been rather steady and favors Republicans. However, it fell to McGovern levels for Obama in 2012 and to an even lower level for Clinton in 2016.

Difference between Democratic margin of the vote among Catholic voters and the national popular vote (Election margin/Catholic margin)

1972 | +12% (-23%/-11%)
1976 | +08% (+02%/+10%)
1980 | +02% (-10%/-08%)
1984 | +09% (-18%/-09%)
1988 | +02% (-07%/-05%)
1992 | +03% (+06%/+09%)
1996 | +08% (+08%/+16%)
2000 | +02% (00%/+02%)
2004 | -02% (-03%/-05%)
2008 | +02% (+07%/+09%)
2012 | -02% (+04%/+02%)
2016 | -09% (+02%/-07%)

Takeaway: Republicans have been making notable inroads among Catholics for many years now, with 2008 being an aberration. Trump outperformed all previous Republicans among this group.

Difference between Democratic margin of the vote among non-religious voters and the national popular vote (Election margin/non-religious margin)

1972 | +45% (-23%/+22%)
1976 | +09% (+02%/+11%)
1980 | +15% (-10%/+05%)
1984 | +37% (-18%/+19%)
1988 | +33% (-07%/+26%)
1992 | +38% (+06%/+44%)
1996 | +28% (+08%/+36%)
2000 | +29% (00%/+29%)
2004 | +39% (-03%/+36%)
2008 | +45% (+07%/+52%)
2012 | +40% (+04%/+44%)
2016 | +40% (+02%/+42%)

Takeaway: Save for 1976 and 1980, non-religious voters have been a solidly Democratic voting bloc. It was not until Obama in 2008 that Democrats reached McGovern level support among non-religious voters again. They have been at deviation +39% or higher for 4 consecutive elections now.

Difference between Democratic margin of the vote among married voters and the national popular vote (Election margin/married margin)

1984 | -06% (-18%/-24%)
1988 | -08% (-07%/-15%)
1992 | -07% (+06%/-01%)
1996 | -10% (+08%/-02%)
2000 | -09% (00%/-09%)
2004 | -12% (-03%/-15%)
2008 | -12% (+07%/-05%)
2012 | -18% (+04%/-14%)
2016 | -12% (+02%/-10%)

Takeaway: Married voters have been slightly trending away from Democrats for years now; this process has remained stable since 2004, save for the record deviation of -18% for Obama's reelection.

Difference between Democratic margin of the vote among White evangelical voters and the national popular vote (Election margin/White evangelical margin)

2004 | -54% (-03%/-57%)
2008 | -57% (+07%/-50%)
2012 | -61% (+04%/-57%)
2016 | -67% (+02%/-65%)

Takeaway: White evangelical voters have been quickly moving away from the Democrats since at least 2004, when exit polling of this group started.

Difference between Democratic margin of the vote among non-White evangelical voters and the national popular vote (Election margin/non-White evangelical margin)

2004 | +05% (-03%/+02%)
2008 | +03% (+07%/+10%)
2012 | -03% (+04%/+01%)
2016 | +22% (+02%/+24%)

Takeaway: In 2016, non-White evangelicals split decisively from White evangelicals by surging in their support for Clinton to a record deviation of +22%, while White evangelicals reached a record deviation of -67%. There is a very strong division that emerged in 2016 between these two camps.

[JA]

Difference between Democratic margin of the vote among under $30,000 voters and the national popular vote (Election margin/under $30,000 margin)

2004 | +23% (-03%/+20%)
2008 | +26% (+07%/+33%)
2012 | +24% (+04%/+28%)
2016 | +10% (+02%/+12%)

Takeaway: Clinton experienced a significant drop in support among the traditionally Democratic under $30,000 income category in 2016.

Difference between Democratic margin of the vote among $30,000-49,999 and the national popular vote (Election margin/$30,000-49,999 margin)

2004 | +04% (-03%/+01%)
2008 | +05% (+07%/+12%)
2012 | +11% (+04%/+15%)
2016 | +07% (+02%/+09%)

Takeaway: Obama seems to have surged among this group in 2012, both his margin and his deviation, despite having decreased in overall support from 2008. Clinton did not entirely hold these gains, but still did better than Obama in 2008 in deviation and Kerry in 2004 in margin.

Difference between Democratic margin of the vote among $50,000-99,999 voters and the national popular vote (Election margin/$50,000-99,999 margin)

2004 | -09% (-03%/-12%)
2008 | -07% (+07%/00%)
2012 | -10% (+04%/-06%)
2016 | -06% (+02%/-04%)

Takeaway: Clinton improved upon Obama's 2012 margin with this group and still performed better than Kerry. In terms of deviation, Clinton did better than Obama and Kerry.

Difference between Democratic margin of the vote among $100,000-199,999 voters and the national popular vote (Election margin/$100,000-199,999 margin)

2004 | -12% (-03%/-15%)
2008 | -10% (+07%/-03%)
2012 | -14% (+04%/-10%)
2016 | -03% (+02%/-01%)

Takeaway: Clinton made substantial inroads among this high income category while simultaneously plummeting among the lowest income category.

Difference between Democratic margin of the vote among $200,000+ voters and the national popular vote (Election margin/$200,000+ margin)

2004 | -25% (-03%/-28%)
2008 | -01% (+07%/+06%)
2012 | -14% (+04%/-10%)
2016 | -04% (+02%/-02%)

Takeaway: Obama (2008) and Clinton (2016) made significant inroads among this income category, but it shows itself to be incredibly elastic and very favorable to a Republican candidate who is seen as pro-business/tax cuts/deregulation.

New York Times

Is there anything in this data that stands out to you? Or that you came to a different conclusion than me?

[JA]

Is there anything in this data that stands out to you? Or that you came to a different conclusion than me?

Honestly, it's hard to determine what trends will actually stick beyond this election. This election does seem to have accelerated some existing trends, and it is notable that there were such huge swings despite the close margin. For instance, Obama did well with 200k+ voters in 2008, but he also won by over 7 points. Clinton got just a smidgen over 2 points and nearly matched Obama '08 among that group.

If I had to guess the long-term trends from this election, I would say it sped up the movement of college educated whites and high-income voters to the Democratic Party, but that Democrats may never have a true stranglehold on high-income voters due to their policy agenda. I wonder just how stable these party coalitions will be in the future, as the platforms of each party don't seem to naturally align with their bases.

I came to a similar conclusion. It is almost like the parties have maxed out their possible trends; I cannot see the current system holding under such contradictory conditions. Party policy and ideology are too much at odds with core constituencies that something will have to give, essentially an unwraveling. Could that mean a realignment is imminent? It would appear so.

[JA]

People always talk about 1980 being the main election in the Reagan revolution but I would argue that 1984 and 1988 were more important as those were the ones what validated the Reagan years. Even if a Bernie/Warren/Sherrod wins in 2020 he still has to be as successful as Reagan was in their time in office to ideologically realign the country.


It wasnt 1980 what ideologically realigned the country it was the Reagan presidency.

This was also true of Lincoln and FDR.

Of course it was , and I don't think the far left will be successful. In my opinion a far left candiate wins in 2020 and then after being a failure populism dies and we go back to having center left and center right politicians

People always think that'll happen with extremist Presidents. As I said, Lincoln, FDR (at least when he ran for reelection), and Reagan were all seen as extremist Presidents by almost half the country. Lincoln's election was the tipping point for half the country to secede. FDR instituted the most left wing economic policy in our country's history and won reelection by a 20+ point landslide. Reagan was seen as such an extremist kook that even his own future V.P. in his own primary referred to his economic theory as "voo doo".

I don't think in these political and economic times with our polarization, desperation, and anger that somehow we'll shift back to normal after one unsuccessful administration.

Among what might be called more "comfortable" segments of the population, populist politics is viewed from the perspective of an outsider. They do not experience the sort of exasperation and despair with the system and ordinary life that a very large, perhaps even a majority of the population are experiencing. So, they see populist revolts as temporary aberrations before a return to the status quo. But there is no returning to a status quo for tens of millions of Americans; that is simply intolerable at this point.

Trump won riding a populist wave. In many respects, Cruz voters could even be seen as part of that as well. Sanders voters were obviously the left wing manifestation of that. Trump declined among the wealthy and increased substantially among the lowest income group because the former is comfortable and the latter is in despair. Trump may be able to feed off of that populist rhetoric again in 2020, but he will be lucky to hold his gains among the lowest income when they start to realize he is governing (rhetoric and unsophistication aside) like a conventional Republican. The wealthy will return to him in 2020 for that exact reason.

People are craving substantial, FDR/Reagan/Lincoln level change. They were even willing to vote in a crass vulgarian to achieve that. He made obviously deceptive promises in his campaign that coopted aspects of leftwing populist politics; none of which he has remotely signaled he intends to pursue. His chickens will come home to roost; perhaps in 2020, but more likely for the next Republican candidate in 2024.