What’s riding on the Election 2016 swing set?

DAYS BEFORE Election Day, Hillary Clinton's once formidable lead in the national opinion polls had shrunk to a few percentage points, spreading panic among liberals and Clinton supporters.

If the polls are correct--and given their pro-Republican slant in 2012, that is a big "if"--Clinton is in danger of losing as many as four "swing states" that Barack Obama won in two successive presidential elections: New Hampshire, Florida, Ohio and Iowa.

With the exception of Florida, all of these states have a higher percentage of non-Latino whites in their electorates, compared to the national average. All of them except New Hampshire have a lower rate of voters with a college education than the U.S. average. (See the U.S. Census data here.)

Why focus on race and education level? Because the national media and many social scientists dub "non-college educated whites" as the "white working class," which is supposedly the core of Donald Trump's base. So if Trump can cut into Clinton's support in these states, the media will explain it as the revenge of the "white working class."

Of course, that doesn't explain Florida, where non-Latino whites are 63 percent of the electorate, compared to 69 percent of the national electorate.

Nor New Hampshire, a mostly white state where 35 percent of the electorate holds a college degree, compared to the national average of 31 percent, and where 32 percent has an annual household income of more than $100,000, compared to the national average of 25 percent.

The mainstream media image of the stereotypical Trump voter is of a non-college educated older white male worker whose livelihood is under pressure from immigration and trade. But a large-scale analysis of Trump supporters by the Gallup organization, challenged many of these assumptions, as Charlie Post noted:

A systematic review of Gallup polling data demonstrates...that most Trump supporters are part of the traditional middle class (self-employed) and those sectors of the new middle class (supervisors) who do not require college degrees. They tend to live in "white enclaves" isolated from immigrants and other people of color, have worse health than the average U.S. resident, and are experiencing low rates of intergenerational mobility. While not directly affected either by the decline of industry in the Midwest or by immigration, these sectors have experiencing declining living standards and are fearful about their children's prospects of remaining in the middle class.

Other analysis has shown that Trump voters are, by and large, wealthier and better educated than the average American.

If Trump does end up winning Ohio, Iowa and New Hampshire, will that signify that the majority of voters in those states are unreconstructed racists? That's a hard conclusion to sustain given that a majority in all of those states voted twice for Obama.

But if a more Trump-friendly electorate shows up in these much-watched swing states, it might indicate something else: that eight years of having a Democrat in the White House--when most real incomes failed to recover from the 2007-08 recession--led large numbers of voters to reject the status quo.

There's two ways for former Obama voters to register that discontent: vote for someone other than Clinton this time, or not vote at all.

One other point on swing states that leans in the other direction: Trump's immigrant- and Mexico-bashing may bring out more Latinos to the polls to vote against him than 2012.

On the weekend before the election, election experts in Nevada proclaimed that Latino early voting in that state may have already defeated Trump before Election Day even arrived. The Latino surge will probably deliver Colorado--and maybe even Florida--to Clinton.

That's yet another reminder that the U.S. working class is more racially and ethnically diverse--and tolerant--then the national media ever recognize.