The fate of Republican Donald Trump's White House ambitions could hinge on a simple question: Can the unconventional candidate remake the 2016 presidential electorate?
As much as polls have moved in recent weeks one factor seems pretty solid at this point, Trump is having a hard time winning college educated voters. The 2012 presidential exit polls show why that's a major problem for him.
You can break the 2012 electorate roughly into thirds using race and educational attainment numbers:
- Minority voters made up about 28% of the 2012 electorate, according to the exit polls.
- White voters with a college education made up about 36% of the electorate.
- And white voters without a bachelor's degree made up the other 36%.
So minority voters plus college-educated white voters equaled about two-thirds of the 2012 electorate, 64%. Currently most polls show Trump losing badly among minorities and also losing among college-educated whites - in some cases by double-digits.
That seems to leave the Trump campaign with two options. Win whites without a college degree by astounding margins or rely more on increasing that group as a part of the electorate. The latter option seems more possible, but exactly how possible goes to question of what the voter pool will look like?
Both Hillary Clinton and Donald Trump have remarkably high negative ratings from voters. More than half of registered voters, 53%, had a negative view of Clinton in the last NBC News/Wall Street Journal poll, while 61% had a negative view of Trump.
Those numbers don't indicate an excited electorate.
But Trump is not a normal candidate and he seems to have inspired a segment of non-traditional voters. Thus the Trump hope: What if he can bring out a larger-than-usual group of his non-college educated white base, while, at the same time, a large group of college-educated whites and minority voters stay home because they are not strongly behind Clinton.
In the game of measuring "likely voters" in 2016, some polls seem to be leaning toward this theory of the electorate.
For instance, the CNN/ORC poll released Tuesday showed Trump with a two-point lead over Clinton among likely voters, 45% to 43%. That was despite the fact that, in that same poll, Clinton led among college educated white voters by 13 points (49% to 36%) and among minority voters by 53 points (71% to 18%).
In the CNN poll, Trump held an enormous lead (44 points) among white voters without a college education, 68% to 24%. That was a much bigger margin than Mitt Romney's 26-point lead with that group in the 2012 exit polls. But Romney won college educated white voters by 14 points. And in the CNN poll Trump is losing by 13 points among college-educated white voters.
Considering those numbers, how can Trump win? It's all about the composition of the electorate.
If you apply the 2012 exit poll turnout numbers to the CNN poll numbers for minority voters (28%), college-educated whites (36%) and non-college educated whites (36%), Clinton wins by about four points: 46% to 42%.
But if Trump can pump up the share of non-college white voters in the electorate to about 44% (an 8-point increase from 2012) he could win, even with those deficits in the rest of the electorate.
That wouldn't be easy. For Trump to change the electorate in such a dramatic way he will need to pull off something truly remarkable - even historic.
Minorities grow as a percentage of the electorate in every election as the nation's Hispanic and Asian populations in particular increase. Going into 2016 most demographic analysts predicted the white vote would be 70% or less of the total vote for the first time in U.S. history.
On top of that, the percentage of voters with a college education grows every cycle as well. In 1990 about 20% of Americans over the age of 25 had at least a bachelor's degree. Today it is about 30%.
Barring a major shake-up in the campaign fundamentals those numbers tell the story of the final stretch of the 2016 campaign.
Trump's best chance to win looks as though it is tied to changing the composition model of the 2016 electorate - and change it in a way that works against the dominant demographic trends in the country.
The odds look long, but so were the odds that the untested, volatile New York businessman would become the GOP nominee.