How were the polls so wrong?

4 surprises that led to Trump's win
4 surprises that led to Trump's win

    JUST WATCHED

    4 surprises that led to Trump's win

MUST WATCH

4 surprises that led to Trump's win 01:01

Story highlights

  • Two Princeton professors reflect on one of the most unexpected outcomes in presidential history
  • Sam Wang: Perhaps it was shy Trump voters who did not give pollsters honest answers

Sam Wang and Julian Zelizer are co-hosts of the podcast, "Politics & Polls." Wang is a professor of neuroscience at Princeton University and a founder of the Princeton Election Consortium. Zelizer is a professor of history and public affairs at Princeton University and a New America fellow. He is the author of "Jimmy Carter" and "The Fierce Urgency of Now: Lyndon Johnson, Congress, and the Battle for the Great Society." The opinions expressed in this commentary are theirs.

(CNN)Julian Zelizer and Sam Wang, two professors at Princeton who co-host a podcast on "Politics and Polls," exchanged emails early Wednesday morning, as America witnessed one of the most unexpected outcomes ever in a presidential election with the victory of Donald Trump. Virtually all the polls had forecast a Clinton victory, as did forecasters including Wang.

Zelizer: I am waking up to digest the stunning news. Sam, how are you thinking about what happened. Obviously you had been very confident that Hillary Clinton would win, and it looks like she won't. What just took place?
    Julian Zelizer
    Wang: Yes, it's amazing. I am very surprised at where things are headed tonight. At the Senate level, GOP candidates outperforming polls. At FiveThirtyEight, they're talking about an electoral vote/popular vote split -- a weak tendency for a Trump electoral win, Clinton popular-vote win. I thought that would be a freak occurrence, but it's not looking that silly at the moment.
    Sam Wang
    Yes, last night's election was totally unexpected news. This failure of polling was the largest in a presidential election year in decades. Donald Trump exceeded his polls against Hillary Clinton by about 4 percentage points. That may not sound like a lot, but with such a close race, it was enough to reverse his fortunes.
    It wasn't just the top of the ticket: in Senate races, Republican candidates outperformed their polls by a median of 6 percentage points. Again, this has a tremendous impact because Senate races were very close. Democrats gained a net of four seats, getting them to 48 votes - and leaving Republicans in control.
    It's going to take some time to sort out what went wrong in the polling industry. I viewed state polls as a gold standard for monitoring elections. But this year, I was wrong. Could it be shy Trump voters? He is, after all, so unprecedented: never held office, an outsider to his own party, no organized campaign, and a fairly checkered past. And now, President-elect Donald Trump.
    There is one more piece to the puzzle: the national vote. Before the election, Hillary Clinton led national surveys by a median of 4 percentage points. At the moment, the New York Times' model is projecting Hillary Clinton to win the popular vote by a little over 1 percentage point. That's again a 3-percentage-point gap. Notably, this leaves us with a situation like 2000: one candidate wins the electoral vote and the other wins the popular vote. The last time that happened, George W. Bush became President.
    Working-class whites give Trump the White House
    Working-class whites give Trump the White House

      JUST WATCHED

      Working-class whites give Trump the White House

    MUST WATCH

    Working-class whites give Trump the White House 02:20
    Zelizer: There is a lot of shock and awe out there. Last night it became clear pretty early that the polls weren't catching something pretty significant that was taking place and Trump's appeal with Americans was much stronger than people were saying—even if Clinton ended up with a higher popular vote. I have no idea what happened with the polls; that's not my expertise. It does remind me, though, of the value of history which always emphasizes those unexpected turning points that we live through or the periods when the conventional wisdom turns out not to be as wise as we thought. There are all sorts of interesting data to digest, including his better than expected performance (from exit polls) with Latinos and even African Americans despite the campaign he ran.
    My guess is it will take some time to sort through all this, for much of the electorate to really grasp what happened. All of the political analysis industry will need to take a serious look at how this unfolded and what they were not grasping.
    We still need to see if this is more like 2000, a close contest in a polarized era, or 1980 where we see a realignment take place.
    Editor: Julian, is this a governing moment?
    Zelizer: Yes, this is a governing moment like 2009, if not as big as 1965. Republicans will unite, Trump will find points of agreement like tax cuts, and there is the potential for a burst of legislation. The only check is that Senate Democrats can obstruct and delay and filibuster.
    Follow CNN Opinion

    Join us on Twitter and Facebook

    Wang: Now do you think the majority will do away with the filibuster requirement?
    Zelizer: Yes, I think of all the potential scenarios this is the one that produces the best chance for filibuster reform. Republicans sense opportunity unlike any they have had in decades. But it could be that Democrats filibuster filibuster reform.