Statistikhengst
Diamond Member
- Banned
- #1
Charlie Cook and David Wasserman just put out an interesting analysis of the 2016 election in terms of demographics, and the concensus is that it looks very, very ugly for the GOP, much uglier than the GOP wants to admit.
So, here we go:
2016 Voters by the Numbers - NationalJournal.com
That's it, in a nutshell.
To back up the stats, here are the exit polls from 1976 through 2012:
How Groups Voted in 1976 - Roper Center
How Groups Voted in 1980 - Roper Center
How Groups Voted in 1984 - Roper Center
How Groups Voted in 1988 - Roper Center
How Groups Voted in 1992 - Roper Center
How Groups Voted in 1996 - Roper Center
How Groups Voted in 2000 - Roper Center
How Groups Voted in 2004 - Roper Center
How Groups Voted in 2008 - Roper Center
How Groups Voted in 2012 - Roper Center
Yes, the statistic from 1980 is correct but at the same time somewhat misleading. Ronald Reagan only took 56% of the White vote in 1980 and still won with a +9.74% landslide over Democratic incumbent Jimmy Carter because it was a three man race and John Anderson (Independent) was also in the mix.
In 1984, the second largest popular vote landslide since 1964 and the largest EV landslide since 1936, Ronald Reagan took 66% of the White Vote, which is the high water mark in this category since Roper has been polling for voter demographics.
The main message is that Romney came close to 60% of the White Vote in 2012 and still lost to Obama by 3.86 points. The reason is obvious: a shrinking White electorate, and a elderly White electorate that is shrinking even more quickly.
So, in 2016, even if for some reason, a GOP nominee were to get 66% of the White Vote as did Ronald Reagan in 1984, since the White Vote is likely to go from 72% of the electorate in 2012 to 70% of the electorate in 2016, that would probably not be enough to get a GOPer over the top in the EC, but maybe in the PV. The problem with that is that no GOPer is likely to get to 66% of the White Vote, because the female vote is already showing a major tendency toward Hillary Clinton (D), also among White Women.
As the data shows, had Obama won with the same groups, but with the projected 2016 demographics, his margin would have been +5.4 over Romney instead of +3.9. That's an automatic 1.5 point difference, just based on expected shifts in the demographics for the next presidential cycle.
The math makes it clear: without inroads into the minority vote, there is no real path for a GOP victory at the national level, plain and simple.
And Charlie Cook's assumption that the White Vote will sink again in 2016 is exactly in line with electoral history. In fact, instead of 70%, I would not be surprised if the White Vote clocks in at 69% in 2016.
Here the exact numbers for the White Vote, vis-a-vis electorate and GOP take of the White vote:
YEAR / % of electorate / Republican take
1976: 89% / 52%
1980: 88% / 56% (three-man-race)
1984: 86% / 66%
1988: 85% / 60%
1992: 87% / 41% (three-man-race)
1996: 83% / 46% (three-man-race)
2000: 81% / 55%
2004: 77% / 58%
2008: 74% / 55%
2012: 72% / 59%
Actually, I think that Charlie Cook's comparison of Reagan to Romney was false. It is much more logical to compare Bush 41 to Romney. In 1988, Bush 41 took 60% of the White Vote and won the election by +7.73%. In 2012, Romney won the White Vote by 59% (he did better than Bush 43 from both 2000 and 2004!), but lost by -3.86%. That's an 11.59% difference, in spite of the fact that both gentlemen took essentially the same amount of the White Vote.
This is how much of a difference the electorate shows when it goes from being 85% White (1988) to 72% White (2012).
Let's look at the shifts as well. In every cycle EXCEPT 1992, the White Vote shrunk over the previous cycle. And since 1996, the White Vote has shrunk by no less than 2% over the previous cycle, so a prediction of 70% for the White Vote in 2016 is absolutely in-line with electoral history.
Also, let's think about the Women's vote:
Obama won the women's vote by +13 in 2008 and by +11 in 2012. Hillary is polling between +15 and +21 in the women's vote and has been the entire time - and I am being VERY conservative with this estimate here: against Ted Cruz, it's +29. Let's take a mean of +18, 5 points higher than Obama's margin from 2012. Assuming 90% of the Black Vote for Hillary, 70% of the Asian Vote and at least 70% of the Latino vote, this can only mean a certain amount of that rise in the women's vote must come from White women voters, which leads to the logical assumption that it is going to be very hard for a Republican to get above 59% of the White Vote overall, much less even hold at 59%.
So, at a time when Donald Trump is holding a mega-anti-immigration rally in a state well known for it's antipathy toward Latinos, namely, Arizona, the demographics are pointing clearly to a need for the Republican nominee to make inroads into the Latino vote.
It's that simple. And it is also the reason for the title of the OP: "The Divide" - because there is a huge divide between that which seems to attract the ultra-conservative majority of the Republican party in a primary election - and that which will attract General Election voters in November 2016.
In other words, the phenomenon of the "red-meat" that Mitt Romney threw to the Right in 2012, which then ended up being a poison-pill for him in the General, appears to be even more prominent looking toward 2016.
George W. Bush (43) managed to get to -9 in the Latino vote in 2004 and he took 58% of the White Vote, and that barely got him over 270 EV (286, to be exact). With a reduced White electorate looming in 2016, a GOPer cannot afford anything worse than -10 in the Latino vote, and it looks like the cleft might be as large as -60. That is a math that they cannot win with.
Yepp, the demographic numbers look brutal for the GOP in 2016.
So, here we go:
2016 Voters by the Numbers - NationalJournal.com
That's it, in a nutshell.
To back up the stats, here are the exit polls from 1976 through 2012:
How Groups Voted in 1976 - Roper Center
How Groups Voted in 1980 - Roper Center
How Groups Voted in 1984 - Roper Center
How Groups Voted in 1988 - Roper Center
How Groups Voted in 1992 - Roper Center
How Groups Voted in 1996 - Roper Center
How Groups Voted in 2000 - Roper Center
How Groups Voted in 2004 - Roper Center
How Groups Voted in 2008 - Roper Center
How Groups Voted in 2012 - Roper Center
Yes, the statistic from 1980 is correct but at the same time somewhat misleading. Ronald Reagan only took 56% of the White vote in 1980 and still won with a +9.74% landslide over Democratic incumbent Jimmy Carter because it was a three man race and John Anderson (Independent) was also in the mix.
In 1984, the second largest popular vote landslide since 1964 and the largest EV landslide since 1936, Ronald Reagan took 66% of the White Vote, which is the high water mark in this category since Roper has been polling for voter demographics.
The main message is that Romney came close to 60% of the White Vote in 2012 and still lost to Obama by 3.86 points. The reason is obvious: a shrinking White electorate, and a elderly White electorate that is shrinking even more quickly.
So, in 2016, even if for some reason, a GOP nominee were to get 66% of the White Vote as did Ronald Reagan in 1984, since the White Vote is likely to go from 72% of the electorate in 2012 to 70% of the electorate in 2016, that would probably not be enough to get a GOPer over the top in the EC, but maybe in the PV. The problem with that is that no GOPer is likely to get to 66% of the White Vote, because the female vote is already showing a major tendency toward Hillary Clinton (D), also among White Women.
As the data shows, had Obama won with the same groups, but with the projected 2016 demographics, his margin would have been +5.4 over Romney instead of +3.9. That's an automatic 1.5 point difference, just based on expected shifts in the demographics for the next presidential cycle.
The math makes it clear: without inroads into the minority vote, there is no real path for a GOP victory at the national level, plain and simple.
And Charlie Cook's assumption that the White Vote will sink again in 2016 is exactly in line with electoral history. In fact, instead of 70%, I would not be surprised if the White Vote clocks in at 69% in 2016.
Here the exact numbers for the White Vote, vis-a-vis electorate and GOP take of the White vote:
YEAR / % of electorate / Republican take
1976: 89% / 52%
1980: 88% / 56% (three-man-race)
1984: 86% / 66%
1988: 85% / 60%
1992: 87% / 41% (three-man-race)
1996: 83% / 46% (three-man-race)
2000: 81% / 55%
2004: 77% / 58%
2008: 74% / 55%
2012: 72% / 59%
Actually, I think that Charlie Cook's comparison of Reagan to Romney was false. It is much more logical to compare Bush 41 to Romney. In 1988, Bush 41 took 60% of the White Vote and won the election by +7.73%. In 2012, Romney won the White Vote by 59% (he did better than Bush 43 from both 2000 and 2004!), but lost by -3.86%. That's an 11.59% difference, in spite of the fact that both gentlemen took essentially the same amount of the White Vote.
This is how much of a difference the electorate shows when it goes from being 85% White (1988) to 72% White (2012).
Let's look at the shifts as well. In every cycle EXCEPT 1992, the White Vote shrunk over the previous cycle. And since 1996, the White Vote has shrunk by no less than 2% over the previous cycle, so a prediction of 70% for the White Vote in 2016 is absolutely in-line with electoral history.
Also, let's think about the Women's vote:
Obama won the women's vote by +13 in 2008 and by +11 in 2012. Hillary is polling between +15 and +21 in the women's vote and has been the entire time - and I am being VERY conservative with this estimate here: against Ted Cruz, it's +29. Let's take a mean of +18, 5 points higher than Obama's margin from 2012. Assuming 90% of the Black Vote for Hillary, 70% of the Asian Vote and at least 70% of the Latino vote, this can only mean a certain amount of that rise in the women's vote must come from White women voters, which leads to the logical assumption that it is going to be very hard for a Republican to get above 59% of the White Vote overall, much less even hold at 59%.
So, at a time when Donald Trump is holding a mega-anti-immigration rally in a state well known for it's antipathy toward Latinos, namely, Arizona, the demographics are pointing clearly to a need for the Republican nominee to make inroads into the Latino vote.
It's that simple. And it is also the reason for the title of the OP: "The Divide" - because there is a huge divide between that which seems to attract the ultra-conservative majority of the Republican party in a primary election - and that which will attract General Election voters in November 2016.
In other words, the phenomenon of the "red-meat" that Mitt Romney threw to the Right in 2012, which then ended up being a poison-pill for him in the General, appears to be even more prominent looking toward 2016.
George W. Bush (43) managed to get to -9 in the Latino vote in 2004 and he took 58% of the White Vote, and that barely got him over 270 EV (286, to be exact). With a reduced White electorate looming in 2016, a GOPer cannot afford anything worse than -10 in the Latino vote, and it looks like the cleft might be as large as -60. That is a math that they cannot win with.
Yepp, the demographic numbers look brutal for the GOP in 2016.
Last edited:
