intelligencer chats

How Much Does One (Very Troubling) Poll Matter?

Nevertheless, she persisted. Photo: Elijah Nouvelage/Bloomberg via Getty Images

A troubling poll on Monday morning has had Democrats – at least the ones who spend quite a bit of time online – in a tizzy all week. Are the campaigns taking a calmer approach? And should they? I spoke with Intelligencer’s roving political reporter, Gabriel Debenedetti on the topic of whether freaking out is warranted.

Ben: In a big new poll on Monday, the New York Times and Siena College found some disturbing news across their survey of six swing states that President Trump won in 2016. According to the poll, which was rigorously assembled (though it’s still just one result!), the three leading Democratic contenders — Biden, Sanders, and Warren — perform quite a bit worse than one might expect against the president in Pennsylvania, Michigan, Wisconsin, and elsewhere. But while Biden and Bernie hold narrow leads in most or some these states, respectively, the news was particularly bad for Warren, who only wins a hypothetical matchup in Arizona. The foggy question of “electability” has always trailed her, to the consternation of many progressives; this scary result will supercharge those conversations, and indeed, it launched many a take this week, including in our august pages. To what extent do you think it’ll affect the Warren campaign’s viability?

Gabriel: I don’t think one poll in November of the off-year is going to do all that much to her viability, especially when other polls have shown her far ahead of Trump in these states/others. If any one of these campaigns had particularly strong feelings about these numbers, it’s Biden’s. Team Joe started fundraising off the numbers — which support their central thesis that their candidate is the electable one — pretty soon after they landed. But let’s take a step back here for a sec and think about these numbers in context. The distances between Biden’s and Bernie’s and Warren’s numbers are very similar to what they have been in other hypothetical matchup polls. What’s different here is that Trump does significantly better across the board. So there’s reason for Democrats overall to be nervous, but it’s a bit of a stretch to expect any one of the candidates to start freaking out.

I do wish they had included numbers for Buttigieg, Harris, and Klobuchar, etc. If we’re really going to have a national conversation about electability, it’d be useful to have numbers for all the candidates who are talking about it a lot in Iowa these days.

Also, side note. As I type, I see that these numbers were mentioned yet again at a Biden fundraiser in D.C. today. I don’t think we’re going to stop hearing about them from his camp anytime soon.

Ben: Well, that makes sense. Polls have shown that most Democratic voters care about beating Trump above all else, which is certainly understandable. I’ve long thought that if anything, Biden doesn’t talk about his advantage in this arena enough. It should be his main thing!

Gabriel: It’s the main thing everyone around him talks about. But another question I keep coming back to here when I think about it is: did anyone really think these swing states weren’t going to be close? That’s what “swing” means …

Ben: I think people thought that Pennsylvania and Michigan might not be all that close. They weren’t before 2016, and all of the elections in both states since (including last night’s local ones in PA) indicate that things were returning to “normal,” a.k.a. a significant Democratic advantage. The other states were less surprising.

Gabriel: That’s totally fair, but this may be where these “people” may be losing it a bit too much over one poll … 52 weeks before Election Day.

Ben: (It should also be noted that the poll said getting people to respond in Michigan was very difficult, because they were all mean on the phone.)

Gabriel: That’s understandable, since every person in Michigan has had to answer questions about Trump from political reporters at least 14 times each over the last three years.

I’ll throw another hot take at you.

Ben: Please!

Gabriel: Obviously Biden is happiest here, but I think the other campaigns who will be happy with these numbers are the “don’t forget about me!” ones — Harris and Klobuchar, and obviously Buttigieg, too. (Among others. I’m not ignoring you, Senator Booker.) Each of them talks a lot about electability, which is obviously the thing Democratic voters care most about right now. Since they weren’t included at all here, and since we’re apparently, as a media industry, determined to make this one poll ~~the narrative~~, it might not be hard for them to turn around and say to voters some version of “hey look, those front runners should make you pretty nervous going up against Trump…”

Ben: That’s true. I think Klobuchar, with her dominance in Minnesota, might have the best argument there.

Gabriel: If by “dominance” you mean “being from there,” sure. Not sure there’s much evidence she’d destroy everyone else in this field there. But that gets to another important point. Considering these numbers, I’d love to see a similar poll in Minnesota, New Hampshire, Nevada, and Georgia … and Ohio! .. .and Iowa. I would like 50 state polls, I guess.

Ben: To push back slightly on your point about this being THE narrative (and to tie it in with what you just said) — there aren’t a lot of high-quality state head-to-head polls coming out right now. And Nate Cohn, with his prescient piece on hidden white working-class voters back in 2016 and his excellent record in polling 2018 elections, has a fair bit of credibility here — even if he got 2016 largely wrong, like so many others.

Gabriel: Even the most credible pollsters want you to take their polls within the context of other ones. And here’s another one from the last few days: a Washington Post/ABC poll showing Biden up 17 points on Trump nationally, and Warren up by 15. I’ll note that both the WP/ABC poll and the NYT/Siena one just got “A+” ratings from 538. Obviously these numbers aren’t state by state, and though we all know it’s entirely possible there will be an Electoral College/popular vote split in 2020, it’s really, really, really hard to see how you could get national margins like these if the swing state margins end up where NYT/Siena is.

Anyway, I definitely don’t mean to diminish the findings of the NYT/Siena poll, at all. At all! I just want them considered in context.

Ben: If other polls were to come out showing similar struggles for Warren, do you think she WOULD change up her baseline strategy in any way?

Gabriel: Well, many polls have shown something like this, actually — Warren 2–3 points behind Biden when it comes to matchups vs Trump. See the Post/ABC poll  above. Again, the difference is Trump’s level of support. But obviously if Warren sees evidence that people are spooked by her electability, based on polls or what her opponents say, or whatever, you might see her emphasizing different things on the trail. But it’s not like she’s currently talking about Medicare for All much during her town halls, for example, unless asked about it.

Ben: So in the end, it’s less a wake-up call than just another data point.

Gabriel: I mean, you should feel free to be woken up by it. I just wouldn’t yet press the panic button.

How Much Does One (Very Troubling) Poll Matter?