Take a breath. Here’s how to think about the polls.
By Ezra Levin, Indivisible Co-Founder and Co-Executive Director
We’re about two weeks out from the election, so I want to get right to it with this biweekly-ish newsletter. First, the quick reminder: We’re here to get VP Harris and Coach Walz to the White House, deliver them a Democratic Congress, and then help our new Democratic trifecta pass democracy reform and codify reproductive freedom. These newsletters are one way I try to share what I’m seeing, brag about what Indivisibles are doing, engage in some honest-to-God movement-wide conversation — and occasionally share photos of our adorable IndivisiKids. We’re almost there, so let’s hop to a summary:
The News: The media has been consumed by it: the polls! Harris is down! These demographic groups are slipping! That battleground is lost! We’re losing!!! Real talk: We’re not losing. And Trump and his MAGAs are actively building a sense of inevitability so they can turn around and deny the results after they lose. My practical advice is simple: Don’t help them. For your own mental health and for the good of this campaign, accept that the race is extremely close and focus on contacting voters. That’s the winning play.
The Brag: Speaking of doing the work…we’re doing the work. I’m writing this summary from a flight to Texas to get Colin Allred elected, while Leah is headed to Bucks County, Pennsylvania to talk about reproductive freedom. Indivisibles across the country are blowing through our door targets across the country. It’s rock and roll time.
The Discussion: I hope you’ll join Leah and me for a final coffee chat the day before the election — 3pm ET on Monday, November 4. We know there’s a lot of anxiety and uncertainty about what will happen on election day and in the aftermath. We’ll talk through what we’ve been seeing around the country, how we’ll be watching the returns, and what the plans are for potential post-election chaos. Please register here, and feel free to submit any burning questions you have — we spend most of our time on Q&A!
The News: What we should talk about when we talk about polls
The frustrating limits of data right now. Let me tell you a story that’s on my mind. It’s a snowy night outside a rowdy bar. A drunk man stands near a lamppost, rummaging through the snow. A friendly passerby notices his frantic search and asks, “Do you need help?”
“I’m looking for my wallet — I dropped it over there,” the man replies, pointing to a dark patch of ground across the street.
Puzzled, the passerby asks, “Then why are you looking here?”
The drunk man, exasperated at the question, responds, “Because this is where the light is!”
Data is great. Data can shine a light on frustratingly impenetrable problems. I love using data to make informed decisions about where and how we invest more of our time, energy, and resources to defeat the MAGAs. But just because data can generate light doesn’t mean it always tells us what we need to know. Sometimes, the wallet is on the dark side of the road.
Pollsters can tell you how their model of the electorate will vote. What they can’t tell you is whether their model of the electorate represents the actual electorate. Nobody can; it’s really hard to do, and it’s getting harder. That’s why the polls were so off in 2016 and 2020. Every year, pollsters try new tools and techniques to adapt their models, but we won’t know until the actual election whose model was accurate.
Two weeks out from the election, the daily barrage of polls is shining a lot of light and demanding a lot of attention. Politicos, media talking heads, and armchair Nate Silvers are searching around that lamppost. They’re making grand proclamations about what we’re all desperate to know: who’s gonna win?
The Democratic freakout has arrived. Along with those polls and hot takes comes the inevitable, predictable, almost sacred Democratic Party ritual: the quadrennial late stage election freakout.
Like clockwork, at this time of year we get Democrats worrying about their candidate. We read that after a comfortable Democratic lead, the Republican is making troubling gains in October. We hear alarming reports from bellwether battleground counties in the most important swing states. There’s talk that the seemingly unqualified Republican VP candidate did themselves and the ticket a lot of favors in the debate. And then there’s the key battleground state that, while it modestly favors us, we learn is far from safe. Sometimes, it’s the sinking sensation that Pennsylvania is slipping away from the Dems.
If you didn’t click any of those links, I’ll explain the gag: none of those links are about this election. Those are articles about the presidential elections in 1976, 1992, 1996, 2008, 2012, and 2020 — the last 50 years of successful Democratic presidential campaigns.
I get the urge to freak out. We all lived through 2016. We’ve been burned before, and we want to know the future, especially when we’re careening toward a future that could either be catastrophic and dangerous or inspiring and reassuring. Polls, political betting markets, “chance to win” models, and seemingly meaningful anecdotal reporting shine a light on that dark future. But the light won’t help us find our wallet or defeat the fascists. At this point, only voter contact can.
Why the freakout hurts us and helps Trump. Here’s the short version of it: The freakout is unsupported by the data and is worse than irrelevant — it’s actively harmful to our goal of defeating Trump. Want more? Here’s the long version of why we should push ourselves to reject the freakout:
- An objective look at the polls says we’re winning. If you look past the deluge of Republican polls and focus just on the average of the polls since Harris got in the race, the polling has been pretty consistent. Harris has maintained a small but durable lead nationally and in a sufficient number of the seven battleground states. Full stop. Period. That’s it. It’s a boring story, and boring stories don’t get retold as much as hot takes, but that is indeed the truth.
- The polls won’t be telling us anything new. We’ve known for months now that seven states will determine the next president: NV, AZ, WI, MI, PA, GA, NC. We’ve known for months that each of those battleground states is too close to call based on available data. The polls are uncertain and will remain so — of that we can be certain.
- Buying into the doomsaying actively helps Trump. If you didn’t catch our Women Wednesdays for Harris call, I’d recommend you watch it here. Our favorite brilliant messaging expert Anat Shenker-Osorio made this excellent point: We all know Trump is planning to contest the election results after he loses. That plan depends on building up a sense of inevitability about a win so that he can turn around and yell “rigged” after he loses. As Anat recommends, don’t carry Trump’s water for him.
We’re within the margin of effort. No poll within the margin of error will help you find what you’re looking for — but voter contact in these final days sure can help all of us. We call that the margin of effort.
With that in mind, I will share with you my thought process for consuming polls two weeks out from the election. I hope this helps.
My method for staying both sane and focused at this point in the election:
Hold your head high. We’re winning. And we’re doing the work to win.
The brag: It’s coming down to turnout, and we’ve built a turnout machine
Speaking of doing the work…we’re doing the work. You might have caught Leah on Chris Hayes or me with Alex Wagner talking about the ground game or our managing director Mari on MSNBC. I’m writing this summary from a flight to Houston for Allred, while Leah is on a train to Pennsylvania for Harris, and Mari is either in Arizona or Nevada (it’s hard to keep track).
Across the battleground states, Indivisibles are blowing through our door targets. Our relational door-knocking program — Neighbor2Neighbor — that connects volunteers to sporadic democratic voters within a block or two of where they live is on fire. We’re having hundreds of thousands of conversations with crucial, must-get voters in must-win states.
We just greenlighted more money for all this work because volunteer demand is outstripping supply. And if we can raise more, we’ll put it to good use. So if you want to help us reach more voters between now and election day, chip in here.
While everything may feel like chaos, this part of the campaign is pretty simple. The battlegrounds are known. The teams are in place. The tactics are in. The time for grand strategic debates is over. Now we just go as hard as we can.
The discussion: Let’s regroup on election eve
Over the course of the last several months, we’ve had six coffee chats — informal, unrecorded Q&A sessions to talk about the campaign, organizing, messaging, and anything else on your mind. Put it on your calendar: On Monday, November 4, at 3pm ET, we’ll hold another coffee chat to discuss what we’ve been seeing on the campaign trail, how we plan to watch returns come in, and what to expect after the election. I hope you’ll join us for this election day eve chat!
Until then, remember: We’re winning and we’re going to win, because we’re doing the work and we’re going to do the work.
