A little more on the polls

Of course, I saw this piece from Nate Cohn right after I posted my last one.  My favorite part is that he systematically goes through all the reasons to expect the polls to be more accurate this year:

Of course, the polls could be even further off this time than four years ago. But there are also many reasons to think they could be better this time around.

Perhaps most important, many pollsters now weight their sample to properly represent voters without a college degree. The failure of many state pollsters to do so in 2016 is widely considered one of the major reasons the polls underestimated Mr. Trump’s support. Voters without a four-year college degree are far less likely to respond to telephone surveys — and far likelier to support Mr. Trump. By our estimates, weighting by education might move the typical poll by as much as four points in Mr. Trump’s direction.

Though many state pollsters still do not weight by education, far more do than four years ago. The Monmouth poll is one example. The final Monmouth poll of Pennsylvania in 2016, which showed Mrs. Clinton up four percentage points, would have shown her with a two-point lead, 47 percent to 45 percent, if it had been weighted by education, according to Patrick Murray, director of the poll. That alone covers about half of the difference between the actual result and the final Monmouth poll, and it’s a reason to have more confidence in the new Monmouth poll.

Education weighting is not enough to ensure a perfect result. After all, Mrs. Clinton still would have led — albeit quite narrowly — in the final Monmouth poll of Pennsylvania, even weighted by education. And other high-quality, education-weighted state polls, like the Marquette Law School poll in Wisconsin, still showed Mrs. Clinton with a considerable edge in 2016. Other factors were clearly at play.

But many of the other sources of polling error in 2016 also seem less likely to repeat.

There are far fewer undecided or minor-party voters now than four years ago. These voters broke in Mr. Trump’s favor, according to exit polls and postelection surveys that recontacted pre-election respondents, helping to explain part of his strength in the final results. Undecided voters could again break toward Mr. Trump, but this time there are simply fewer of them — and therefore less opportunity for the polls to be wrong for that reason.

Another component of the polling error in 2016 was turnout. It is hard to generalize the role of turnout in the polling misfire because there are as many likely-voter models as there are pollsters. But some pollsters probably overestimated Black turnout by relying too heavily on Obama-era turnout models. This time, reliance on the last election could just as easily understate Democratic turnout. In general, the party out of power tends to enjoy relatively higher turnout than when it’s in power, and that could help Democrats compared with 2016.

In the case of the Monmouth poll, it turns out that turnout was a significant factor. Mr. Trump would have led Mrs. Clinton by one percentage point, 45 percent to 44 percent, in the final Monmouth University poll of Pennsylvania if it had been weighted by education and if the likely-voter sample had consisted only of those who ultimately voted, according to voter file records. (Mr. Trump won the state by 0.7 of a point.)

Similarly, the final New York Times/Siena College poll of North Carolina, which was conducted over the weekend before the election and found the candidates tied, would have shown Mr. Trump ahead by four percentage points (he won by 3.6 points) if the likely-voter sample had consisted only of validated voters. These kinds of findings led many pollsters to conclude that survey research wasn’t fundamentally broken after the 2016 election, and led many to redouble their efforts rather than abandon the enterprise…

Perhaps the biggest risk is one that has loomed over the polling industry for a decade: declining response rates to telephone surveys. Up until now, there has been little evidence that low response rates have endangered the accuracy of high-quality survey research. It turns out that the people who respond to telephone surveys appear to vote similarly to people from their same demographic group who do not respond.

Okay, and there’s even more.  Cohn is thorough.  And you should thoroughly believe that the polls are more likely to be accurate in 2020 than they were in 2016.

Of course, the potential fly in the ointment is the fly in all the ointments– Covid.  But still, work you butt off for Biden, but don’t think he’s in the same spot HRC was.

2020 ≠ 2016

I know it’s hard to just look at the polls and say, “Damn, it seems really, really likely that Biden is going to win this thing.  Period.”  Because we all said that in 2016 and we were wrong.  In large part, because, HRC was likely, but we all took likely to be “really, really, likely.”  But Biden really is.  I would say “a lot can change” but let’s be honest here.  Leopard…spots; Zebra… stripes; Trump…manifest incompetence.  Let’s just put it this way.  Something completely out of our experience of modern polling and presidential elections would have to change dramatically for Trump to win.  Could that happen?  Ummm… sure.  But, currently, Trump seems on a path to a historic defeat.  Two posts from G. Elliot Morris get at the dynamics:

In another recent post, he also mentioned that their model would have only had Clinton at 65% right now– far less than Biden’s 90+%.

Of course, don’t get complacent.  Work like Biden is trailing.  And, given the utter disaster for American democracy a Trump re-election would be, even an 8% chance is concerning.  But, it’s important to recognize that, for the most part, the polls really weren’t that off in 2016 and Biden’s lead is massively larger than Clinton’s.  Also, of note, he’s clearly out-performing her in the Midwest swing states which were so essential to Trump squeaking through.  I don’t love what it means for our country, but, electorally, seems to be something to the boring old white guy.

%d bloggers like this: