Stop it with the Clinton coronation. Trump can still win

There's something funny about all the focus on whether Donald Trump will concede the election results gracefully. There's something odd about all the speculation on whether Trump will drag Congressional Republican candidates down with him. And there's something off about people wondering just how large Hillary Clinton's landslide victory will be.

Yes, most of the polls say Trump is behind nationally and in the key battleground states. Yes, those news event betting sites are all setting odds at about 80 percent for a Clinton victory. And polls and odds aside, you could argue a politically incorrect outsider like Trump never really had a chance anyway. But the thing is, he hasn't lost yet.

This column is not meant to make some kind of case for Trump's chances. But it's a simple statement of fact that in almost every election, there is some variety in poll results. This one is no different as the L.A. Times/USC poll calls it a tie, and Rasmussen Reports and the historically most accurate Investors Business Daily Poll both say it's just a one point race. Two other polls, Reuters/Ipsos and Economist/YouGov have it as a four-point lead for Clinton. This is hardly landslide or even comfortably certain territory. But maybe it's not the actual, "who are you voting for?" polls we should be looking at now.

First off, people often do vote or tip off their vote with their feet. And Trump rallies are still jam-packed, compared not only to Clinton's rallies, but the usual attendance we see at political rallies even this close to Election Day. It's that kind of consistent enthusiastic turnout that has Stony Brook University Professor Helmut Norpoth convinced that Trump will win the election based on a model he constructed that's mapped out every election result since 1912 correctly, (except for 1960).

Most of the pundits, including University of Virginia elections expert Larry Sabato, still believe turnout for Clinton will still be strong enough for her to win. But in every election in the past where negative campaigning has been dominant, turnout is depressed. And when was the last time a presidential election was as nasty and negative as this one?

"Trump rallies are still jam-packed, compared not only to Clinton's rallies, but the usual attendance we see at political rallies even this close to Election Day."

But what about that elephant in the Trump Tower penthouse? His personality is a problem. Most of the electorate simply does not see him as a good or even decent person. In any other election year, where the personal likability of the candidate is probably the most important factor, the public's lack of esteem for Trump would be lethal.

Yet this isn't any other election year. Trump is running against a candidate with an almost equally poor likability factor. So by definition, millions of us will be voting for someone we don't personally like or even greatly respect. And in this age of terror and Wikileaks, something could happen to make us switch our vote to the other candidate that we also don't like or respect all that much. Pollsters and other experts alike could be assuming that since most voters do not personally like Trump or Clinton, they won't vote for one or the other. Political choices are ultimately emotional choices for 99 percent of us, and polls can't always gauge emotions very well.

But let's look at some other polls for a second, because there is absolutely no diversity when it comes to what the polls are telling us about the direction of the country. Every major poll taken over the past two years has shown a massive majority of the respondents believe the country is going in the wrong direction.

The average of the last nine of those polls has the "wrong direction" folks with more than a 35 percentage point majority. That's well into the territory where historically the incumbent party in the White House is toast. And Clinton has done almost nothing to try to recast herself as a change agent. This is one key wildcard the election pollsters may not really be able to adjust for when it comes to how they measure Trump's support.

And that wildcard comes mostly from economic uncertainty and pain. No one thinks things are as bad as they were during the height of the 2008-09 Great Recession. But wages are still depressed, the GDP is barely growing, and a newly-leaked email shows even Democratic National Committee acting Chair Donna Brazile telling Clinton campaign Chair John Podesta that the economy is simply not good.

Economic disenchantment has been evident throughout this campaign, especially at the Trump and Bernie Sanders rallies. Perhaps the reason Clinton doesn't draw the big crowds despite a lead in the polls is that even her supporters may not be up for hearing happy talk about the Obama economy and how she'll keep it going. Economic despair, even among Clinton fans, has to be considered a possible drag on turnout for her and a possible lightning bolt for Trump.

The bottom line is that the usual election rules are in the trash in 2016; that's been apparent ever since Trump first surged in the early GOP primary polls. The question is why so many experts are so certain that things are back to normal enough to stop worrying about a Trump win.

Commentary by Jake Novak, CNBC.com senior columnist. Follow him on Twitter @jakejakeny.

For more insight from CNBC contributors, follow @CNBCopinion on Twitter.