🔮 “Most Uncertain I’ve Ever Been”

He Called the Chinese Spy Balloon, George Santos, and 2020. Here's his take on 2024.

Nov 01, 2024

🔮Welcome back to The Oracle by Polymarket, where we bring you the news through the lens of prediction markets.

Gaetan Dugas is the pseudonym of a political gambler who gained notoriety for his accurate calls in the 2020 and 22 cycles. A self-described “Amy Klobuchar Democrat,” Dugas correctly predicted the exact Amy Coney Barrett Supreme Court confirmation vote total, among other calls that he shares on X and as a frequent guest on the Star Spangled Gamblers podcast.

A tech worker by day, Dugas began trading the Iowa Electronics Markets, one of the earliest real-money prediction markets, as a side hustle and has expanded from politics to predicting Spotify charts, Rotten Tomatoes scores, and other pop culture events. Dugas spoke with The Oracle about how he’s looking at the Trump-Harris matchup.

This interview with Dugas has been edited for length. All answers are his own.

The Oracle: You’ve announced that you are not betting on the Harris - Trump market. Why sit out the Super Bowl?

For me, the presidential race is just way too uncertain to make a big bet. So I’m hoping to get a few smaller wins in states that will be called early. Then I can take those proceeds and look at some of the other markets. If the polls looked like they did in 2020, my strategy would probably be different. In all likelihood, we're not going to know the winner with any certainty on election night. It might take several days.

The Oracle: If you had to put odds on Trump vs Harris, what would you give it?

I'd say probably 55-45 Harris, and that's super low confidence. I think Kamala has the advantage in the Blue Wall states, but she needs to hit all three of those, assuming she loses the Sun Belt, so that's where the uncertainty comes in. If you’re favored in three but you have to win all three to win, does that mean that you're favored overall? It's tough for me to calculate, and that's part of why I'm avoiding that bet for now.

The Oracle: Does betting help you stay grounded about politics?

The reality is that the federal government doesn't affect my life very much on a day to day basis. So I could get really worked up over stuff, and I just choose not to. So that helps me compartmentalize things. Sure, I don't like Trump, but if he wins, my life isn't going to be bad. And if he is gonna win, and I'm gonna get $50,000 as a result of that, it takes a little bit of the sting out.

The Oracle: Where does your alpha come from with your political bets?

I put in the work on research, especially with some of these more obscure questions. For example, when George Santos won his seat in Congress, and then all this stuff came out about him, there were betting markets on how long he would last.

So I did the research. I looked back a couple of decades, and found all of the different criminals in Congress and how long it took from the time that they were indicted to the time they resigned or were expelled from their position. And I realized that this is not going to happen fast, no matter how crazy it was. It just takes a long time to build a case, indict somebody, and put him through the court process.

One of the markets I'm looking at now is when the election will be called. So I looked all the way back to 2000 and researched all the different races to see how often races are called the same night. And it turns out not very often, really, and that type of information can lead me to finding markets that are mispriced.

The Oracle: Do you believe polling averages are polluted by right wing polls?

I think that was true in 2022, and I think it's true to a lesser extent now, but I also don't think it's making that much of a difference. Nate Silver looked at that in detail, and he didn't find that there was much difference. That being said, I do prefer to use someone like Adam Carlson who compiles the results only from A and B ranked pollsters.

The Oracle: What about this shy Harris voter theory? That women in the South are going to secretly pull the lever for Harris but can’t say it out loud.

Don't buy it. If you think about the situation, not only would the person have to be a shy Harris voter, but they would have to be worried that their husband would somehow find out what they're saying to a pollster about their support, and I don't buy it.

The Oracle: What about pollster herding? That the pollsters want to avoid being outliers from the averages?

I 100% believe that. Because random sampling shouldn't look like that. If the race was truly tied 47-47, you wouldn't see ties in most polls. You'd see a lot of variance, and we're not seeing that variance, which leads me to believe that people are trying to protect their reputations by playing it safe. NYT - Sienna have been posting some outliers, and also CNN had a poll in Michigan and Wisconsin that was different from everybody else. But the others have been playing it very safe.

The Oracle: How much does the early vote affect your process?

I agree with Nate Silver on this, that the Jon Ralston stuff [on Nevada] is good and pretty much everything else is garbage. Even before COVID, that state had three things that made the early vote numbers meaningful. First, they voted early in huge numbers. So by election day, like 80% of the vote would already be in. The second is that they report party registration of early votes. And the third is that party registration closely matches actual votes. In Alabama or Mississippi, registered Democrats are going to far outnumber the number of votes that you would expect for Democrats, just because for so long, the South was dominated by Democrats, and people just didn't keep their registrations up.

On top of that, Ralston's track record is really good. Another reason I think it doesn't work well in other places is because there's just so much difference election to election, so it’s hard to model.

The Oracle. You were bullish Kamala in Pennsylvania and Michigan but then backed off. Was that early vote-related?

It was the number of African American votes in Georgia. That raised my uncertainty more than anything. If I'm going to be betting a significant amount of money, I want to be fairly certain that I'm right. And when I see those numbers coming in as low as they were for the African American population, it causes concern. I knew that there was a likelihood it would turn around later, but I just wanted to see some of that evidence before making a bet. You can tell I'm playing it quite conservatively. This is the most uncertain I’ve ever been in an election.

The Oracle: What’s driving all this uncertainty? That the polls are close?

This is the first time I'm going in without any certainty on the general election winner. I was pretty certain in 2016, but I was wrong, so my certainty in the past hasn't always translated into success.

The big sources of uncertainty are that all seven swing states are polling within one or two percent either way, and the number of undecideds or third-party in those polls is higher than the difference between the two candidates.

Another one is what happens post-election. Say that Trump wins or Kamala wins 270 to 268, how confident are you that Trump isn't going to pull something off? So that's another piece of uncertainty where I don't want to hold 98 or 99 cent shares in Georgia after Kamala gets the most votes. And maybe I'm being paranoid about that, but I don't want to risk a bunch of money on something like that.

The Oracle: Why do you think Kamala’s odds are consistently lower than the Democratic senate candidates?

I think that's pretty much unavoidable. As soon as Biden decided he was going to run for reelection, he was handicapping whoever was going to replace him. This is because they weren't going to go through the process of a competitive primary that would let people get to know and like the person over the year between when they declare and the primaries are over. I think that hurts her to not have gone through the full process.

The Oracle: So you’re saying Amy Klobuchar was robbed?

Yeah, she’d be riding to a 404 point Electoral College victory for sure. The Democrats are not a homogenous party by any means. So it helps to hash that out during the primary process. With Biden and Bernie, for example, both guys got their messages out. Bernie ended up influencing Biden's platform to the point where he didn't really have trouble winning those voters back. And that just didn't happen with Kamala. Of course, she's been vice president, but vice presidents don't really do shit, so I don't think many people know what she's been doing for the last several years.

The Oracle: For the electoral college, you’ve been most positive on the Dems to win by 0-4 bracket (🔮 9% odds), that’s a squeaker.

I think the most likely outcome is a split between the Rust Belt and the Sun Belt with Trump taking the Sun Belt and Kamala taking the Rust Belt, and NE-2, which, if you add that up, it's 270 to 268, so the closest possible result. And that’s what I think people are going to start moving to. As all those higher electoral vote brackets that still have low value in them go away, I think it's going to move towards those lower brackets.

The Oracle: What are you watching in the next five days? Is there any time for one more card to come out?

I don't think things are going to change much. The ones that I think matter are what happens in Nevada the next couple of days, and what Ralston says about it. If he says that Kamala is heavily favored, then the Dems 0-4 electoral college outcome gets blown up because Nevada goes to the Democrats. In that case, there aren't a lot of realistic maps that result in that, in the 0-4 margin. And number two, I'm just looking at final polling. I want to get those final polling averages and make decisions based on those. So if it comes out that Michigan is something like 49 to 47.5 in the final polling averages, I'd probably pull the trigger on something like that.