Trump, polls, and what experts (like me) did and did not get wrong

I was listening to the Slate political gabfest yesterday and David Plotz quite aptly pointed out that although experts were wrong about Trump, what wasn’t wrong was the polls.  Not long after he entered the race, Trump took the lead and basically never relinquished:


It’s just that experts are well aware that polls this early in a primary election season are just not all that predictive.  Things like party endorsements and fundraising tend to be far more predictive.  This time they weren’t.  That doesn’t mean everything else we know about Trump and elections is wrong.

If the polls were right about Trump and the experts were wrong for the primary is there really any rational basis to expect the exact opposite for the general?  No.  It’s well established that national primary polls the year before have very little predictive value.  It’s also established that general election polls have limited, but increasing value at this point.  Especially when you get to the point where you have the two candidates matched up as presumptive nominees.  Now, these things can change and they have in the past, but at this point, general election match-up polls are modestly predictive and they strongly predict a Clinton win.


Maybe this time all us experts are totally wrong about something different about Donald Trump.  But just because he up-ended what we know about primaries– which are very different sorts of elections– does not all all portend that he will upend what we know about general elections.  And what we know about those says he’s a likely loser.


About Steve Greene
Professor of Political Science at NC State

2 Responses to Trump, polls, and what experts (like me) did and did not get wrong

  1. ohwilleke says:

    On the other hand, a fairly complete survey of state level general election polling is pretty convincingly pro-Clinton if older polls aren’t too stale to be meaningless.

    Based on polls in all battleground or “leaning” states except CO and NV which favor Clinton based on the past two elections the tally is as follows:

    Clinton: 374 electoral votes
    Tied: 6 electoral votes (Utah)
    Trump: 168 electoral votes

    There is also no polling in Arkansas which has six electoral votes and a decent chance of supporting its “native daughter” in the race (even though she isn’t really a native), as Clinton leads in several other southern states (VA, NC, FL, GA and is only narrowly behind in MO and MS (he also trails in AZ). So, a 386-162 electoral vote landslide for Clinton is far from impossible at this point.

    The fact that Trump has to fight just to keep states like GA, MO, MS, UT and AZ may drain his resources for fights in other key battleground states.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: