fbpx
5 Jun.

Al Hunt: Hey Democrats: the problem is not the polls, its Biden

Frequently someone asks, “Can you really trust the polls?” They are Democrats or anti-Trump Republicans who want a negative answer.

Sorry, but the major national polls are pretty much on the mark.  With some variation, Donald Trump enjoys a small, but steady, lead of a point or two.

With events and campaigns, this can change. Two candidates held bigger leads at this time; Michael Dukakis in 1988 and President Jimmy Carter in 1980. The contest today is less fluid, and President Biden’s ceiling is lower than winning candidates in those years.

Source: RealClearPolitics

 

My colleague, James Carville, gave an excellent polling tutorial on Substack three weeks ago. (Link here) I want to add a few particulars to Professor Carville’s wisdom.

I’ve been fortunate over the years to work closely with three of the very best pollsters: Peter Hart, the late Bob Teeter and Ann Selzer. Among the many lessons they taught, was that for all the understandable focus on the horse race, the internals, or demographics, can be more instructive.

Over the next five months, you want to see how Biden and Trump are faring with various groups compared to past performances. For Biden, he is not close to getting 75% of voters of color or 60% of young voters, which he got in 2020 and needs this time. For Trump, a challenge is to win the white women’s vote as he did in the last two elections.

A key, as usual, will be the suburbs, where Biden is running behind his previous performance, and Independents, where one pollster says Trump is “showing some wear and tear.”

On the reliability of the major national polls, there are two caveats.

One, as Republican pollster Dave Iannelli says, is turnout:  “notoriously difficult to gauge because it requires sampling a population that does not yet exist.”

In 2016, he notes, some pollsters underestimated the turnout of non-college educated white voters favoring Trump, and in 2020, there was some underestimation of overall turnout.

Also, as was dramatically underscored in 2000 and 2016, it’s the electoral college that decides the Presidency, so pay more attention to state polls in seven states: Pennsylvania, Michigan, Wisconsin, Nevada, Arizona, Georgia and North Carolina.  Quality here varies; there are some good ones like the Marquette Law School poll in Wisconsin.

In Senate races, where in competitive contests, Democrats are running well ahead of the President, take seriously the standings of well-known incumbents. In Pennsylvania, for example, Democrat Senator Bob Casey, whose family is a popular fixture in the Keystone State, has a five-to-seven-point lead over super-rich hedge fund executive David McCormick who has moved back to Pennsylvania from Connecticut.  I doubt McCormick, even with a massive super PAC largely funded by billionaire friends, can successfully paint Casey as a radical left-wing cultural warrior.

Discount surveys that, in this polarized partisan environment, show a large undecided bloc.  A recent survey of the North Carolina Governor’s race showed the Democratic Attorney General Josh Stein and Republican Lt. Governor Mark Robinson tied at 37%. A quarter of Tar Heel state voters are not undecided.

That race is close, but Democrats have more to work with as more voters become aware of some of Robinson’s extreme assertions. Stein launched his first attack ads this week on abortion. Robinson is seen saying to women that he doesn’t support “killing the child because you weren’t responsible enough to keep your skirt down.” (Link here)


On the Presidential race, there isn’t likely to be any such new information; both these candidates are more than well known. We don’t know yet the impact of the Trump felony conviction. At the Wall Street Journal, Teeter and Hart advised to usually wait a couple weeks after a big event to get a reliable measure.

Overall, skeptics say, remember how wrong your polls were eight years ago when Trump pulled a shocking upset. I was directing polling then with Selzer for Bloomberg News. On the final 2016 poll, days before the election, Selzer had Clinton winning the popular vote by 2.4 percentage points; she won by 2.1 percentage points. You can’t get much closer.