Here’s a lightly edited transcript of my discussion with Stephen Leuchtman, a Montana pollster.

Jen: Hey, everyone! Welcome to the blog. Today, I’m talking with Stephen Leuchtman about polling, Polymarket, and how this influences the elections we’re watching. Let’s start with Polymarket—what’s going on there? Betting on elections seems like a wild concept.

Stephen: It’s a unique twist for sure. In the U.K., betting on elections is common, but here in the U.S., we’ve been more puritanical about it. The idea of betting on elections was generally frowned upon. Polymarket allows users to bet on political outcomes using cryptocurrency. It’s a playground mostly for “crypto bros” and libertarians who see it as the next big thing.

Jen: So you have to bet with crypto? That already sounds complicated.

Stephen: Exactly. You’ve got to pass two hurdles: first, you need to buy cryptocurrency, which isn’t a simple process. Second, you need to be willing to gamble on elections. Peter Thiel, who’s behind Polymarket, has essentially created a betting market for political races. It’s mostly aimed at the most degenerate gamblers, as I like to say, because not only do you have to believe in crypto as a stable store of value, but you also have to want to bet on elections. It’s a narrow market.

Jen: It sounds like it could really distort perceptions of elections.

Stephen: It does. When you look at Polymarket, it’s largely filled with users who lean toward the right and believe in a “red wave” during elections. This creates a skewed sense of reality. In 2022, many were convinced that a red wave was coming, and they bet heavily on that outcome. Obviously, that didn’t happen, but the platform’s betting trends shaped the perception.

Jen: That’s fascinating. It’s like the stock market but with elections. Speaking of which, let’s shift to more traditional polling. How does it fit into this landscape?

Stephen: Traditional polling is all about modeling the electorate accurately. You’re trying to get a sense of who will actually turn out on election day. It’s a mix of art and science. For example, when we polled Georgia in 2014, we created different models based on possible turnout scenarios. We even adjusted for past trends to make a more accurate prediction.

Jen: Trump recently claimed he’s spending $500,000 on polling. Is that realistic?

Stephen: It’s exaggerated. The reality is that polling costs have skyrocketed. In 1996, we had a 35% response rate for phone surveys. By 2008, it had dropped to about half that. In 2024, the response rate is around 3%, making polling extremely labor-intensive and costly. You’re looking at spending $25,000 to $30,000 for a quality poll now, compared to maybe $10,000 in the past. So, when Trump talks about spending half a million, he’s likely exaggerating for effect.

Jen: That makes sense. Let’s talk about how polls impact voters.

Stephen: Polls have a psychological impact. For example, in 2016, Hillary Clinton’s four-point lead in the polls led some of her supporters to think she was going to win no matter what. That gave permission for some to vote third-party or skip voting entirely. On the flip side, close polls can motivate voters to turn out because they feel their vote could make a difference.

Jen: So polling isn’t just numbers; it’s psychological.

Stephen: Absolutely. For 2024, I predict Harris will win by five or six points. Trump hasn’t made moves to win back voters he lost last time. I’ve also noticed a lack of enthusiasm in traditionally red areas, like Sanders County, Montana. When I was there recently, I saw more signs for John Tester, the Democrat, than for Trump or Tim Sheehy, the Republican candidate.

Jen: That’s surprising. What do you think that means?

Stephen: It means there’s less enthusiasm on the Republican side. But it doesn’t necessarily mean Tester will win Sanders County—it just shows that voters might be less excited about Trump and Sheehy than they were before. It’s a warning sign for Republicans.

Jen: And how about enthusiasm for young voters?

Stephen: Younger voters, especially men ages 18-29, are actually the most pro-choice demographic in America (the Harvard Youth Poll – gold standard). I’ve found that surprising, but it makes sense when you consider that they don’t want an unintended pregnancy to turn into 18 years of child support. They’re also motivated by the loss of Roe v. Wade because they recognize the personal impact it has on them and their partners.

Jen: That’s a perspective I hadn’t thought of. What’s your take on the current state of the polling industry?

Stephen: It’s not in great shape. Response rates are so low that it’s challenging to get accurate results. And then there’s the issue of media organizations using poll averages, like RealClearPolitics, which mixes high-quality and low-quality polls. This creates a false sense of certainty. We saw this in 2022 when “red wave polls” were used to dampen Democratic turnout in Wisconsin, likely costing Mandela Barnes a Senate seat.

Jen: You’ve mentioned that polls are often used for propaganda purposes. Can you explain that?

Stephen: Sure. In 2022, there were a lot of “red wave polls” designed to create a perception of Republican dominance. This was intended to discourage Democratic donors and voters. We call them “red wave polls” because they’re designed to build momentum for one side. These polls aren’t about accuracy; they’re about shaping the narrative.

Jen: So, as we head into the final stretch before the 2024 election, what should voters keep in mind?

Stephen: Take polls with a grain of salt, especially if they seem off. If a poll shows Harris with a 14-point lead among white non-college Southerners, it’s probably wrong. Use your intuition and common sense when looking at polling numbers. And remember, polling is just one piece of the puzzle.

Jen: That’s solid advice. Any final thoughts before we wrap up?

Stephen: I’d say, keep an eye on enthusiasm levels and turnout efforts. Democrats have built an effective canvassing network, while Republicans seem to be struggling with turnout operations, particularly in rural areas. This could make a big difference on election day.

Jen: Thanks, Stephen. This has been a great conversation. Let’s catch up again before the election for some final predictions!