Predicting the Future

In our second episode this season, we dive into the high-stakes world of online wagering. We trace the path of online sports betting from offshore sites like the World Sports Exchange to the landmark 2018 U.S. Supreme Court ruling in Murphy v. NCAA, which dismantled the Professional and Amateur Sports Protection Act ban on sports gambling and launched a billion-dollar industry.

What was once a legal grey area has moved into the mainstream, as betting doesn't stop with sports and has led to the rise of digital prediction markets like Polymarket and Kalshi, where stats nerds and political junkies trade futures contracts on real-world outcomes on everything from Billboard charts to the next occupant of the Oval Office. Are these digital markets the new YouGov poll, or just a gamified version of the public square?

These markets often beat traditional polling by aggregating real-time data and financial incentives, but are they free from their users' biases?

Special guests: 

• Seth Shankle, avid sports better
• Koleman Strumpf, professor of economics at Wake Forest University
• Tom Miller, faculty director of the data science program at Northwestern University and operator of The Virtual Tout

This episode was produced by Kirk McDaniel. Intro music by The Dead Pens.

Editorial staff is Ryan Abbott, Sean Duffy and Jamie Ross.

Transcript

(Intro music)

Kelsey Reichmann: Welcome to Sidebar, a podcast by Courthouse News Service. I'm your host, Kelsey Reichmann. Today on Sidebar: the business of being right. The digital public square has turned everyday folks into authorities on a range of topics. Which song will be on the Billboard Hot 100 this month? Will Taylor and Travis tie the knot in 2026, and who will come out on top in the all-important midterm elections this November? Echo chambers and algorithmic bias can convince all of us that we must have the right answer to each of those questions, but are you willing to bet on it? I'd wager that Lone Star State aficionado Kirk McDaniel could help us decide. 

Kirk McDaniel: Howdy there. Thanks to smartphones, we all now have a mini casino right in our pocket. 

KR: Ah, the tech bros have found another avenue to gamify parts of our everyday life. 

KM: Even from their early days, prediction markets could be used to show the thinking of the masses and even predict the future. Which begs the question: are digital prediction markets the new YouGov poll? 

KR: Before we get into this, I have to ask. Kirk, do you have much experience in online betting? 

KM: Well, no. Sports betting is actually still illegal here in Texas. And even if it were legal, I have never been much of a gambler myself. So, there's a lot that I think I needed to learn about how all this works. So, I called up Seth Shankle, an avid sports bettor whose betting goes back to his childhood. 

Seth Shankle: I got started in sports betting in high school. Me and a friend would, you know, bet between like, $1 and $5 on a bunch of games. Before that, my dad bet when I was growing up. So, even when I was like six or seven years old, he would ask me who I think would win this game or this and that, and I've always been a bit of a stats nerd. 

KM: He means that last part. Seth told me that sometimes he'll spend hours researching stats before placing a bet, and he keeps logs about past bets he made going back 10 years. 

SS: I actually have an Excel document that's quite, quite huge that I screen scrape information in from every day, like from analytics sites out there, and I kind of aggregate all of that data and try to find the right side. 

KM: But putting in all this hard work really comes down to one thing for Seth, and that's the satisfaction of knowing he made a good prediction. 

SS: For me, I think it has more to do with being right than the actual money involved. I think that's kind of how my personality is. Even in like fantasy football, I would just as much rather win a $100 league than a $500 league, just to kind of prove being right. It kind of validates your process. 

KM: But the betting environment in America has changed quite a bit over the last 30 years, with the internet playing a massive role in bringing this practice into the future. 

SS: In terms of online sports, my freshman year in college, my friend and I, we split an account. I don't know if the site exists anymore. It was called the World Sports Exchange. 

KM: World Sports Exchange was one of the first online bookmakers, which allowed people to bet on various sports from 1995 until it ceased operations in 2012. 

SS: Sometime early 2000s, I switched to betting on Bodog, which is now called Bovada. I bounced around for a few offshore sites, probably between like 2002 and 2010-type thing, and then once everything got more allowed here in the states, then I've shifted solely to FanDuel, DraftKings, BetMGM-type apps. 

KR: Why was he doing his betting on offshore sites and what does that even mean? 

KM: Starting in the early ‘90s and going into the late 2010s, states were prohibited from creating a sports betting scheme under the Professional and Amateur Sports Protection Act, or PASPA, a federal law. So, during this time, if bettors like Seth wanted to wager on games, they had to use sites located outside of the US. For example, World Sports Exchange was, before it was shut down, operated out of the island nation of Antigua. But things were destined to change in 2012, when New Jersey voters approved a constitutional amendment allowing sports gambling. This led sports leagues like the NCAA to sue, arguing that the new measure should be blocked by PASPA. In turn, New Jersey claimed that PASPA itself was unconstitutional. 

KR: If I know anything from my time at Courthouse News, it's that there's always a Supreme Court connection. In 2018, the court struck down the Professional and Amateur Sports Protection Act in a case known as Murphy v. NCAA. Justice Samuel Alito authored the 6-3 opinion stating that Congress' gambling ban improperly infringed on state authority. This rare 10th Amendment ruling protected the power of the states and launched what is now a billion-dollar industry. 

KM: Did the dissenting justices think PASPA should be upheld under the 10th Amendment? 

KR: Not exactly. Led by the late Justice Ruth Bader Ginsburg, the dissenters admonished the high court majority for using an ax instead of a scalpel. Ginsburg said PASPA could have survived even without the state regulations, noting separate provisions that applied to private parties and betting schemes. When reviewing statutes, Ginsburg said the court ordinarily engages in a salvage rather than a demolition operation. 

KM: Thanks to this ruling, watching the Super Bowl now sounds like this. 

Super Bowl ads: Betting on the right guy with Fanatics Sportsbook, explained by Kendall Jenner ... And live from the Super Bowl in Santa Clara. Tonight, live-bet the game on DraftKings to score your share of five million in prizes.

KM: Sports betting has become such an integral part of American sports today. In fact, Legal Sports Report, a site that tracks all things sports betting, reported a record-breaking $1.71 billion were wagered on this year's Super Bowl game, and people profited big time by correctly backing the Seattle Seahawks over the New England Patriots. What makes this even crazier is that many U.S. states still limit or outright prohibit this form of gambling. For instance, here in Texas, if I attempt to visit sites like DraftKings or BetMGM, I'm presented with a ‘not available in your area’ message. 

KR: So, folks in the Lone Star State have to stick to the offshore sites if they're looking to do some real betting. 

KM: That's right. But we aren't totally locked out of getting in on the surge in betting. Thanks to prediction markets, folks in states where sports betting is illegal are able to purchase what is essentially a futures contract on the outcome of events like the Super Bowl. Prediction markets like Kalshi, Polymarket and PredictIt are more akin to stock trading. So, think of these like the New York Stock Exchange. And similarly, these markets are regulated by the Commodity Futures Trading Commission, a federal agency that regulates derivatives. Since much of the regulation in these markets is in the hands of the feds, states don't have as much sway in whether their citizens can participate or not. I would like to mention, though, that the state of Massachusetts has sought to prohibit Kalshi from running sports-related markets without having a gaming license. Several more lawsuits have been filed against Kalshi by Indian tribes who claim it's an unlicensed gaming site, and it has filed lawsuits of its own against the states, arguing that it's not subject to their gaming restrictions. Being able to practically bet on anything has attracted most of the attention to prediction markets. As I said before, you can wager on the outcome of sports, but you could also wager on what the temperature will be in Miami or what movie will win best picture at this year's Oscars. And of course, you can always bet on politics. 

KR: It's almost like this term prediction markets just suddenly popped out of nowhere in the past few years. 

KM: Well, prediction markets have been with us for a very long time, but you're right. Even the economists that study these markets have noticed the sudden rise in popularity. 

Koleman Strumpf: Yeah, I'd say over the course of some point, 2024 to today, there's been a boom of interest, and some of the important facets of how these markets can help are actually coming into play. My name is Koleman Strumpf, I'm a professor of economics at Wake Forest University. I do research in a pretty wide range of topics, but one of the things I've been focusing on for many years are prediction markets, all aspects of them, how accurate they are, their history. 

KR: Hmm. What could have been happening in 2024 that made for a boom in these markets? 

KM: I'm sure none of us have forgotten the most recent Super Bowl of American politics, the 2024 presidential election. Billions of dollars were bet on who the winner would be. Markets predicting which states would go for Donald Trump or Kamala Harris saw values in the hundreds of millions. Even now, people are getting their bets in on who will win the 2028 presidential election, with that market having nearly reached a quarter of a billion dollars on Polymarket as of this recording. These markets are having a moment. I asked professor Strumpf how far back in history we can find people betting on politics. 

KS: The constraining factor is having an election, so we haven't always had elections in human history. So, you have to find a place where people start voting on things to predict it. So, in the U.S., depending on how you want to count, it definitely goes all the way back to George Washington, but sort of hit its stride after the Civil War as being something that's very important. Outside the U.S. where, you know, things, longer history was involved, again, it was the issue of, you know, where do we have elections and things like papal selections and city states in Italy, those were sort of early elections. 

KM: But you don't have to go back to the 1500s to find people betting on who the next pope will be. 

Video clip: Bro. The Pope wasn't even dead for 24 hours, and the gambling amount on who is going to replace him has now reached a staggering $4.5 million. Their current favorite is Italian Pietro Parolin, with Filipino Luis Antonio Tagle close behind. 

KR: I bet that it was quite an upset when Cardinal Robert Prevost came out of nowhere to be chosen to succeed Pope Francis and become Pope Leo, the first American-born pope, of course. 

KM: You can believe it was. But getting back to U.S. elections, betting on presidential races has also been going on throughout our nation's 250 years. In fact, Seth Shankle, the hobby bettor that we heard from earlier, told me about his first election he ever bet on. 

SS: The first politics bet I made was on the 2000 election, and that was actually on that World Sports Exchange site, which that took till like December or January to pay whenever the Supreme Court finally ruled on that. 

KM: What I find most interesting is that the means to bet on an election has practically always existed. And while advancements in technology have made that much easier, people's desire to make a wager reflects on how uncertain people feel about our current moment. 

KS: Another thing that you can get from these markets is not even just a forecast, but how much uncertainty or disagreement there is, and that sometimes this requires digging into the markets a little bit more, but you'll oftentimes see situations in which either people very different information or the way in which they process the information is very different. And you can see that there's a lot of uncertainty, both say the prices are moving around a lot or knowing that there's a lot of lack of clarity on a topic or among people, that can also be very important, I think sometimes to know. 

KR: Well, I don't really need a market to tell me that people disagree in this country. 

KM: That's fair. 

KR: What else is there to glean from prediction markets? 

KM: Aside from gauging uncertainty, these markets show politicians, business owners and regular people in real time what is a value to the populace. 

KS: So, the stock market is a casino, but it's a casino that we've decided is socially beneficial because it allows us to say, ‘Look, we need to figure out how do we allocate capital across companies, and this is the best way we can figure out of like, what are the best performing companies that you get the money.’ Yes, people win and lose money, and maybe that from a societal perspective is not great, but it provides some value added. So, the two value adds I'll mention are they provide forecasts over things like elections, international affairs, things that are related to the economy that are very beneficial to policymakers and people. The other thing it potentially provides, we're not there yet, but this is something maybe a little bit for the future, is it provides a place for people to hedge their own uncertainty. So, for example, Polymarket has opened a market on housing markets. Suppose you're not interested in betting in any of this other stuff, but you are selling your house, and most people sell their house in the summer, but here we are in January. I want to like, I'm happy with the real estate market now. I would love to sell my house now, but I can't, so I just want to lock in my price, but I can't do it. I don't know what to do. So, you could actually go to these markets and effectively lock in a price by buying and selling shares in a certain way. 

KM: And sticking with this analogy to the housing market, professor Strumpf said if these markets had been around before the start of the Great Recession, the market collapse might not have come as such a surprise. 

KS: Roll the clock back to 2007, 2008. What was one of the sort of triggers of the financial meltdown is we had a simultaneous collapse of real estate markets across the country, which we had pretty much people thought was pretty much impossible, because people thought the real estate markets were relatively unlinked and idiosyncratic. Well, these markets, if they were around, we'd be looking around and saying, ‘What? What's going on here?’ People who are probably very knowledgeable and engaged in real estate are seeing that there's something strange and it's going on everywhere. So, that I think could have been, you know, getting a week, a month, extra time to understand there's an issue going on, I think would have been very helpful to policy makers. 

KR: I have a nagging question: how do we know that these markets are an accurate reflection of what people are thinking? 

KM: You're not alone in that question, and I think we have to look back at what gives something like the New York Stock Exchange credibility. 

KS: People I would say at this point with the New York Stock Exchange would say this, well, A, I'm aware of it. B, yes, I could win or lose money, but the key thing is, C, I don't think somebody's going to steal my money. That's the thing that I think people are like, is this a big scam? So, you got to get over that. And at this point, because they're so new at my view, I don't think C is so much of an issue, but the perceptions are key. 

KM: Earlier this year, Polymarket got some heat online after the capture of Venezuela's President Nicolás Maduro. An unknown bettor walked away with $400,000 after doubling down that the U.S. would invade the country just five hours before Maduro was captured. The timing of the trade led many to speculate that someone had inside information, but as of this recording, nothing has been reported to confirm who was behind it. Like the stock market, prediction markets could lose people's trust if they are rife with insider trading. But it's how these trading sites respond that will make or break the ability to hold the people's trust. 

KS: If people think they're trading against insiders, it taints the whole entity of what prediction markets are, but here's my, you know, here's my sort of economics intuition on this is suppose we do have insiders. I view having insiders as potentially something that's self-correcting in the following sense. Imagine there's a type of topic or a site which is just rife with insiders, and over time people start to see this and then, you know, you see, like this one, or, you know, these one or two accounts making lots of money and everybody else loses money. Eventually, over time, that site's going to wither away. 

KM: And it's incredibly important that these prediction markets do what they can to preserve traders’ trust, because without that trust, it taints the ability to trust what is being predicted. That is especially important for someone like professor Tom Miller. He's the faculty director of the data science program at Northwestern University, and he runs a website called The Virtual Tout, where he analyzes prediction market data to forecast election outcomes. And since the 2016 election, his predictions have been correct. 

TM: In 2016, okay, everyone was looking at the polls and Nate Silver and the economists, they were doing their forecasts and it was, ‘Oh, it's going to be Hillary Clinton.’ It's, you know, and they were talking like 90% probability of winning. And that wasn't right. They missed it. In 2020, there was a wider range of forecasts back and forth and the market was more volatile. I was one of the few people in 2020 who was talking about prediction markets. I was using PredictIt at that time, but PredictIt had 56 electoral markets. I mean, I had a for every single one of the electoral areas that have electoral votes, there was a market and you could see what the price was. So, I was able to collect all those data from all those PredictIt markets and then feed those into a simulation model that essentially ran the Electoral College. So, in every hour, this was a crazy life in 2020, every hour I was running one million electoral college elections using those data. Every hour. And I was posting on Twitter every hour. My life was hard at that time. I would get very little sleep, okay, for about a month and a half right up until the election. And clearly that was more accurate than any of the polls, okay. And I showed that the prediction markets provided through my simulations provided a more accurate prediction of what was going to happen than FiveThirtyEight or The Economist. 

KM: This system worked for professor Miller going into the 2024 general election, when he predicted a Trump win, whereas many polls at the time were saying that the race was a near tie and some were giving Harris a slight edge. Both professor Strumpf and Miller told me that the edge prediction markets have over polls is that they're constantly being fed new data, whereas polls take time to properly collect the results and release to the public. Professor Miller also said that polls have a tendency to be slightly skewed. 

TM: Surveys will give you very often a, you could say, democratically biased result. So, they're more likely to vote Democratic than other otherwise than Republican. It may be socioeconomic kind of thing where you're essentially paid nothing, right? And it's your time, okay, so, you're willing to give your time. Well, people with lower incomes, they are more willing to give their time, okay. And as a result, turns out a lot of the polls are biased a bit in favor of Democratic. 

KM: But that doesn't mean prediction markets don't have their own leanings. 

TM: The prediction markets, what I found in 2020, was they were biased a bit in favor of Republican. And if you look at the at the research that's been done in the past, largely based on the Iowa electronic markets, they've been around for a while. It's, you know, the people who are involved in prediction markets, they have more disposable income, they're more risk takers. They're more likely Republican than Democratic, slightly. 

KR: If even the prediction markets have a slight bias to them, how could they be seen as an accurate forecast of an election? 

KM: These are all really good things for regular people like us to keep in mind while we're looking at polls and prediction markets. But professor Miller set out to make the most accurate election forecaster, so he fused all these methods together. 

TM: And what I did was I invented something. As far as I know, I'm the only person who's used it so far, which is called a prediction survey, which is conducted like an opinion poll. You try to get a representative sample, but then instead of asking people for whom they're likely to vote, you're asking them, which candidate do you think will win? And you provide an incentive for them to provide their forecast, the forecasts they really believe in. So, they have the opportunity of winning compensation if they're correct in their prediction. It's a market in a sense, but it's not a market where you have to put money in. It's a market where you're participating in a prize. 

KM: He actually employed this prediction survey during the 2021 Georgia Senate runoff elections, when Democrat Jon Ossoff challenged then-incumbent Republican David Perdue in the general election, and Democrat Raphael Warnock faced off against Republican Kelly Loeffler in the special election. While the media said that the races were too close to call, professor Miller correctly called both races for Ossoff and Warnock. 

KR: Given that we're headed into a very big election year, I get the feeling that a lot of these markets will play a big role in how we look at elections.

KM: I think that's my biggest takeaway here. We started off by talking about the betting side of things, but of everyone I spoke to, the only one who regularly engages in sports betting or prediction markets is Seth. 

KR: Neither of the experts you spoke to do any wagering? 

KM: No, they told me to be a good bettor, you have to have an edge in the market, and for them, it's all about what the markets can tell us that is most important. In fact, professor Strumpf and Miller emphasized the many non-betting applications that prediction markets can be good for. Here's an example from professor Strumpf. 

KS: My hope is we get to a point where people who are interested in the world at large will use these sites as a component to their news diet. So, like for me, every day I go to Polymarket, Kalshi, PredictIt, and at a price of $0, I get to leverage what a bunch of really smart people think is the likeliness of a bunch of really important topics, some of which are bubbling up and I didn't even know were like, as an example, like after the U.S. went into Venezuela and extracted Maduro, there were spillover effects on will the U.S. be doing something in Iran? What does this say about China's invasion of Taiwan? Well, these markets are bubbling up in Polymarket or these other sites. And you see these changes and I'm like, hmm. And then I start trying to learn a little bit more about these other topics. 

KM: And as an educator, he has observed how students have found an interest in civics through interacting in political markets. 

KS: I think this can get people more engaged in these kind of civic topics. So, I teach a course on prediction markets. In my course we have a play money market and coming we're in the beginning of the semester now, and some students that come into the class and they're interested in sports betting or something like that, and they're more interested in sports than these other things, but I think most of them think that's the only thing you could try to bet on. And then they start to say, ‘You know, you can have markets on lots of things.’ Even, for example, I wrapped up a market yesterday on how much snow we were going to get in the area. And I learned by Friday from my students, not from myself, that the snowstorm wasn’t going to be as bad here as people had thought. So, the students, I've been doing this for several years, the students get more engaged in politics, current events, economic issues, financial issues and to me, it's like this great tool for getting people motivated to start learning about these other topics. 

KR: Now we'd be remiss without mentioning that as sports betting has made its meteoric rise, so too have cases of problem gambling. 

KM: You're right. For anyone who doesn't know, problem gambling, also called gambling addiction, is where a person's gaming has a negative impact on their finances as well as their mental health. A survey released by the National Council on Problem Gambling last year found that 20 million American adults have shown problematic gambling behaviors during the previous year. That is actually down from the 27.5 million people in 2021, but the number remains high when compared to 2018, when the Supreme Court released its landmark decision. If you or someone you love is struggling with gambling addiction, there's a ton of resources to be found online, but one that I found is the National Council on Problem Gambling's hotline, which can be reached at 1-800-MY-RESET.

KR: It's about that time that we close the books on this episode, but I bet this is just the first chapter of the messy world of online betting and prediction markets. Look out for all our coverage on courthousenews.com, and don't forget to follow us on social media and all your favorite podcast platforms. Next time on Sidebar: Nearly six years after a viral video launched a racial reckoning, the nation's attention is again focused on Minnesota, where two U.S. citizens became casualties in President Donald Trump's vast immigration crackdown. But while the police officer who murdered George Floyd now sits in a federal prison, ICE agents seem to be acting with impunity. Join me for a look at the legal barriers keeping accountability for federal officers out of reach, and the movement to strip agents of a nearly impenetrable immunity shield that leaves victims clinging to legal rights that have no remedies. See you next time.

(Outro music)