What If The Third Debate Were Based On Different Polls?

Welcome to Pollapalooza, our weekly polling roundup.

Poll(s) of the week

The deadline to make the third Democratic primary debate has passed, and thanks to harder qualifying rules, just 10 candidates made the stage. This, of course, was unwelcome news among candidates such as billionaire activist Tom Steyer and Hawaii Rep. Tulsi Gabbard, who were on the cusp of making the debate. Steyer needed just one more qualifying poll, and Gabbard needed two.

And this got us thinking: What would the debate stage look like if the list or type of eligible polls were different? Gabbard, in particular, has taken the Democratic National Committee to task for the specific pollsters included in its list of approved polling organizations, arguing that had the list of pollsters been expanded, she would have had at least 2 percent support in more than 20 polls conducted during the third debate qualification window. And in fairness to her, understanding how the DNC determines its list of approved polling organizations can be confusing. Gabbard did hit 2 percent in YouGov’s latest national survey sponsored by The Economist, but it didn’t count toward qualifying for the debate.

So to better understand how including different pollsters or relying on different pollster methodologies could affect who made the debate stage, we checked to see who would have qualified if:

1. all polls had been counted;
2. just polls from pollsters with a grade of at least B- or better, according to FiveThirtyEight’s Pollster Ratings (as this grade captures a mix of high-quality phone polls and respected online polls); and
3. only live phone interviews polls, which are often considered the gold standard in polling.

And in this thought exercise, we also kept many of the DNC’s requirements for the third debate, meaning we also included only polls released between June 28 and Aug. 28 — and only counted a candidate as qualified if he or she hit 2 percent support in four polls and attracted the support of 130,000 individual donors (including at least 400 individual donors in at least 20 states).¹ We also adhered to the DNC’s rules that limit qualifying polls to national and early-state surveys² and that said two polls by the same pollster in the same geography can’t be counted.³

OK, so first up: Who would have made the stage in our most generous scenario where all polls are counted? Well, maybe not as many candidates as you’d expect given the parameters. Gabbard and Steyer would make the stage with nine and seven polls, respectively. And author and motivational speaker Marianne Williamson comes a little closer to making it with two qualifying polls. (She also has met the donor requirement.) But this still leaves out seven candidates that FiveThirtyEight considers “major” as well as the candidates who have dropped out since the second debate.

So, OK — what about the scenario in which we limit our scope to pollsters with at least a B- grade, according to our pollster ratings? It makes sense that the DNC would want to limit at least some of the pollsters included. So we chose pollsters that are still high quality, but our list of pollsters ends up being a little more expansive than the list of DNC-approved pollsters. And under this scenario, the same 12 candidates would make the stage as in the “all polls” scenario, but it’s a much closer cutoff — Steyer would have ended up with exactly four qualifying polls and Gabbard five — just one fewer than former Secretary of Housing and Urban Development Julián Castro.

On the other hand, what if the DNC had been more — not less — strict with its requirements? For instance, what if the DNC had chosen to just use pollsters that use live phone interviews? Yes, these polls are facing many challenges right now, including low response rates and high costs, but they remain the best performing type of poll. So if the DNC had limited qualification to these types of polls, the number of debate participants would have actually shrunk from 10 candidates to nine. The odd man out would be Castro, who would have ended up with only three qualifying polls, ahead the two for Gabbard and Steyer.

So big picture, you could say the exact DNC rules don’t make a huge difference — most of the same set of candidates makes it on stage regardless. Of course, for the individual candidates on the edge of qualification, that give or take is everything. Suffice it to say, the rules matter quite a bit to them. And in this case, there’s an argument to be made that the DNC’s list of eligible pollsters helped make or break qualification for those candidates on the bubble — Gabbard and Steyer in particular.

Other polling bites

• New polling from Politico/Morning Consult suggests that Democrats prefer “Medicare for All” to building on the Affordable Care Act. The survey found that 65 percent of Democratic primary voters were either “much more likely” or “somewhat more likely” to back a presidential candidate who supported a single-payer health care system like Medicare for All over “preserving and improving” the ACA. Just 13 percent said such a position would make them “much less likely” or “somewhat less likely” to support such a candidate. Among all voters, 53 percent supported a Medicare for All-type health system compared to 34 percent who opposed it.
• A new report from the Pew Research Center found that 61 percent of Americans think abortion should be legal in all or most cases, while 38 percent said it should be illegal in all or most cases. And though Pew found stark partisan divides over abortion policy, it did find evidence that there was more support for policies advocated by the Democratic Party (42 percent) than the Republican Party (32 percent) – though 24 percent said they don’t agree with either party’s policies.
• In a recent report on trust, media and democracy, Gallup and the Knight Foundation found that Americans are at least somewhat worried about local news organizations being consolidated under large media companies, especially if the parent company had strong political views. Sixty-six percent said they would be “very” concerned that the political views of the parent company “would influence the fairness of news coverage,” and 26 percent said they would be “somewhat” concerned. Large majorities also said they were worried about the inclusion of more news from outside the local area and less investment in news gathering and reporting.
• Sen. Bernie Sanders said last week that he would hold companies accountable for their role in climate change, and YouGov Blue/Data for Progress found that about 62 percent of voters would support holding energy producers legally liable “if it could be proven that they misled the public about the consequences of climate change.” Another 20 percent opposed the idea. And perhaps unsurprisingly, support for the idea fell along partisan lines — 77 percent of Democrats supported the idea as did 63 percent of independents, while 39 percent of Republicans supported it.
• Ahead of Labor Day, Gallup released a survey on labor unions in the United States, finding that 64 percent of Americans approve of labor unions; that is one of the highest approval ratings in the past 50 years. Since the late 1960s, approval of labor unions has mostly hovered below 60 percent.
• Pro-Brexit Prime Minister Boris Johnson recently announced that the United Kingdom’s parliamentary session would be suspended until mid-October, not long before the Oct. 31 deadline for the U.K. to agree to a managed transition to leave the European Union. Johnson’s move gives members opposed to exiting the EU without an agreement less time to maneuver against a “no deal” Brexit, and a new poll from YouGov found that 47 percent of Britons oppose Johnson’s decision, while 27 percent support it.

Trump approval

According to FiveThirtyEight’s presidential approval tracker, 54.2 percent of Americans disapprove of the job Trump is doing as president, while 41.3 percent approve (a net approval rating of -12.9 points). At this time last week, 41.5 percent approved and 54.0 percent disapproved (for a net approval rating of -12.5 points). One month ago, Trump had an approval rating of 42.5 percent and a disapproval rating of 53.4 percent, for a net approval rating of -10.9 points.

Generic ballot

In our average of polls of the generic congressional ballot, Democrats currently lead by 6.7 percentage points (46.4 percent to 39.7 percent). A week ago, Democrats led Republicans by 6.3 points (46.2 percent to 39.9 percent). At this time last month, voters preferred Democrats by 5.9 points (46.1 percent to 40.2 percent).