B (3:19)

First time, long time, Galen. Really excited to be on the pod. Listen to you forever. So thanks for having me.

A (3:24)

Well, I'm very excited to have you. I want to get this out of the way up front. You work with Predict it, so you do have some skin in the game. You also have some unique insights into how this whole world works. So we're going to try to keep it as straight shooting as possible. I'll call you out if I think you're being unfair. Does that sound good?

B (3:41)

Yeah, you should definitely call me out if you think I'm being unfair. For sure, that's, that's the goal.

A (3:46)

I'll also say that Predict it operates in quite a different world today from the main prediction markets that we're going to be talking about. Just to give some context, we will get more into that, but I want to begin with the basic blocking and tackling. For the uninitiated, we've, we've mentioned them many, many times on this podcast playing buy, sell, hold the Scottish genes, the whole shebang. For people who have never gone to the website and, you know, certainly never bet any money, which includes me. How do prediction markets work?

B (4:15)

So let me, let me start off with a question for you, Galen. What probability would you give AOC running for president in 2028?

A (4:24)

75%.

B (4:26)

Great. 75%. I seem to think a little differently. I think it's probably closer to 50. 50. Being a senator is really cushy. Got a much better schedule than the president. So because I think it's 50. 50? You think it's 75. I'm willing to sell you my AOC will run for president in 2028 contract for 75 cents. Now, you basically are assigning again a 75% probability a few months down the line.

A (4:54)

Go.

B (4:54)

And there's rumors that she is going to, in fact run for president. And that contract goes up to 90 cents. People are now willing to pay even more for that same contract. You can then sell to sell the biggest buyer or you can wait and hold until that event actually occurs. That is when you'll receive $1 for every single contract you hold. If that doesn't happen, and she says, I'm actually going to run for Chuck Schumer's seat. He's stepping down, he's clearing the for me, then you will receive $0 and I would have made, you know, the 25 cents and gotten out at the perfect time. So that's the basic mechanic there.

A (5:35)

To make it even more basic, when it says on Polymarket Kalshi, predict it, whatever, that the contract costs 60 cents, that is the overall wisdom of the crowds, quote unquote. And we'll get into how wise that wisdom actually is, is placing a 60% chance on this event happening. And so if you purchase at that amount, you say yes. 60 cents on yes. If the thing does happen, you get a dollar. If the thing doesn't happen, you get zero dollars. And every single contract is just worth either zero or a dollar. And of course, you can sell before you reach the conclusion of whatever that market is for. Right, exactly.

B (6:14)

It's a marketplace where people are betting real money on future events. And it's a binary contract option. So it's either worth $1 or $0 at the end of its lifespan.

A (6:24)

And unlike with a casino, for example, where the house creates the odds and you decide you know how you want to bet according to where the house itself places the odds, the odds are established by a bunch of different people buying contracts and the actual prediction market. Companies make their profits from a fee charged on buying and selling those contracts, not as the house, which is actually setting the market.

B (6:53)

Exactly. And every single player has a different fee structure, slightly different, but yeah, generally it's on those fees on every single trade. There are also market makers that are going to typically open up these contracts and markets and questions and they'll offer both buy and sell, you know, yes or no contracts. So they'll make money on the spread between the bid and the asking price. But a simple thing is that if you think something is worth more than I do, I'm willing to sell it to you. And it's that times a million, essentially,

A (7:30)

or potentially a trillion by 2030. I want to do a little bit of history here because I think this is quite interesting. How did we arrive at this current moment with prediction Markets Prediction markets were

B (7:43)

founded in 1988 by Iowa Electronic Markets. It was very, very small and the Internet didn't exist. They were just trying to figure out and forecast elections on a very small basis. The current iteration of prediction markets actually started after the 2008 financial crisis when Dodd Frank was written. It was written in a way so that they could reign in all of the different riskier things that the banks were doing, including credit default swaps. And they put. They filed swaps under the cftc, the

A (8:19)

Commodity Futures Trading Commission. Exactly.

B (8:21)

It was defined as an incredibly broad definition to include things where it's associated with potential financial, economic or commercial consequence. Now that phrase is used a lot in these court cases that we'll end up discussing. But essentially Dodd Frank put swaps under the cftc. Then a lot of time went by. The big breakthrough was that Kalshi launched in 2022. They first attempted political or election contracts. The CFTC tried to stop that. Kalshee sued them and the result of a court case ending in 2024 had them launched just before the 2024 election season and the re election of Donald Trump. The key breakout there was that the Kalshi markets said Donald Trump was 60 plus 70 plus percent chance to win that night and it went up to 90% really, really fast in the night, much faster than if you were watching any election coverage on cable news.

A (9:27)

And so Kalshee is based in the United States, which is why you're sort of describing American law there and aims to abide by whatever regulations there are. Whereas predict it was based in New Zealand, Betfair based in the uk, Polymarket also based abroad. So I guess Kalshi is the one that has made this a more, I guess, legitimate practice here in the United States. For Americans to bet on those outside prediction markets, they have to use VPNs or break the law or what have you, right?

B (9:59)

Yes. So there's a dcm, a designated contract market, which is what Kalshi is. There's crypto or decentralized markets, which is what Polymarket International is. Polymarket also has a US Exchange that's much smaller, sort of failed to launch earlier this year and is not competing with Kalshi as they expected. So most of the news headlines, most of the controversies that you listed at the top are actually with Polymarket International which are crypto decentralized. There's no true regulation, there's no country that has jurisdiction over Polymarket International. And then predict it, just to add is under a no action letter with the cftc. So it's a letter basically stating as long as Predict it follows certain guidelines, there will be no regulation on Predict It. And that was originally filed with University of Victoria. Now it's actually overseen by the Predict IT Market Research Consortium.

A (10:57)

And to add a little bit of detail to that, Predict it is essentially a consortium of academics. I mean, it was started by academics and it remains a consortium of academics.

B (11:09)

It's overseen by this prediction market research consortium. And it's a long list of academics. You can go to predict.org and read all those academics and they are the governance. They basically say, yes, you can put up this market. No, you can't. And they're the guideline there. But it is an academic venture. There is a $3,500 market cap that's tied to the FEC individual campaign contribution limit. Things that are allowed, unpredicted are election outcomes, significant political questions, nominations, confirmations, SCOTUS decisions, any kind of congressional action explicitly not allowed or similar to a DCM or war, terrorism, assassination, criminal proceedings, yada yada.

A (11:55)

So if I think. Let's just pull someone out of a hat. If I think that you, Jacob, are going to win the 2028 election, and I'm really convinced about this, the most I can bet on you winning is $3,500, and that's that. And the most I can lose on you, you know, sorry, eventually not becoming President of the United States is $3,500.

B (12:18)

Yeah, I mean, it's offensive, but yes, that is the most you could lose. Yes.

A (12:21)

And again, as a juxtaposition on polymarket and Kalshi, you can bet as much money as you want.

B (12:26)

There are no limits on polymarket and Kalshi. You can picture it like limit and no limit hold'. Em.

A (12:32)

I think mostly we're going to be talking a lot about folks like Kalshee and polymarket from here on. Although happy that you figured out your sort of academic corner of this prediction market place and happy that I can continue making Scottish teen jokes once again. Okay, so what is the philosophy behind providing a prediction market? Like, what's its value to society or individuals of having this available?

B (12:58)

The thought process is that there are lots of different pieces of evidence. Let's take elections. I talk politics all day. So that's the easiest frame. You have different polling numbers. You have internal external polls, you have fundraising numbers, you have party liens, cook political liens there. You have lots of different pieces of information. The best way to synchronize all of these different forms of Information is incentivizing people to, to research themselves, understand and take honest opinions backed with their own dollars. That's the only way you can actually get people to attempt at the most accurate probability of an outcome at one time. So you're taking lots of different people who have all similar amounts of information. And that is the wisdom of the crowds. That can potentially be expert panels or polls or algorithms, because real money is going to for an honest opinion out of a wide variety of sources.

A (14:04)

Do they force an honest opinion out of a wide variety of sources?

B (14:07)

It's a good question and I, I think it's still outstanding. I did see a report recently, let's take Polymarket International for one, that 70% of trading volume on Polymarket's 2028 presidential election is on candidates with less than 1% chance of winning. So there's $48 million on LeBron James. You might have drafted him in your Democratic nominee draft. There's 9 million on Marco Rubio. That's pretty sane. But there's also $34 million on Kim Kardashian. So you are seeing different, Nice to Kim.

A (14:44)

You know, she's re, she's, she's gone through it.

B (14:47)

She'll pass the bar eventually. I, I believe in her.

A (14:51)

So you're saying no, it does not always do as promised, which is seek truth through the wisdom of the crowds.

B (15:01)

I think different platforms have had different results. There was a Vanderbilt study that, and this is not fully published, but the name of it is prediction markets question mark the accuracy and efficiency of $2.4 billion in the 2024 presidential election. It found that predict it was accurate at a 93% clip. Polymarket was accurate at a 60% clip and Kalshi was at 78%. That's one study. But the accuracy question is still outstanding.

A (15:32)

You know, one other argument that I've heard in terms of the value of this kind of market is that it allows people to hedge real world risk. So say this is a little abstract and keeping it in politics, but like say you're a business that expects to lose money because of the regulation that a Democratic president might impose on your industry. You can bet that the Democrat is going to win and limit your downside from that regulation. Now let's take it to like outside of politics. There's been a lot of attention on betting on weather. Say you're a farmer and a drought will be, will cost you dearly. You can bet on a drought happening which will limit the downside for you. If in fact there is A drought. Is this the kind of argument that we're hearing from folks like predicted Polymarket Kalshee? You can use this the same way you might use other financial tools.

B (16:33)

Certainly. And the, the example that I usually give is you have a wedding, you obviously don't want it to rain on your wedding day, so you can bet on it raining. And if it does, you make money. If it doesn't, you have a beautiful sunny wedding day. Just a personal one, but is that the argument that they're making? Not necessarily actively. I think that's more of a predicted mindset, especially when it comes to like taxes is another good one. If there's an elected official that's going to raise taxes on you, you can kind of take the other side of things and, and hedge your, your trades there. But I think we keep talking about politics, but the majority of the volume, at least in the US on Kalshi is, is on sports and it is on the gaming industry of sorts. About 80 to 90% of the volume on Kalshi is sports and parlays, which is a huge emerging category on Kalshi. Crypto is actually making up the other 10%. So politics and real world events are actually taking a backseat mostly to sports and parlays and crypto.

A (17:42)

Which gets us into the follow up question, what are the pitfalls of prediction markets? I think one of them may be that it's just gambling by another name and therefore gambling that is no longer usually gambling. Sports gambling is regulated by the states. But if the prediction markets are going to make the argument that they're not actually gambling, they are providing a futures contract. This is a backdoor into nationwide legal sports gambling, which is an argument that some states are making that we can get into. But so one of the, one of the pitfalls or cons that people might suggest is this is just gambling. There's nothing incredibly philosophical or smart about it. It's just one other way to make it easy for people to lose money. What are some of the other cons or pitfalls of prediction markets?

B (18:32)

We talk a little bit about the supposed grifting of the current Trump administration. They do have a hand.

A (18:40)

All right, that's the end of today's preview. Head over to GDPolitics.com to become a paid subscriber and catch the full episode. We chatted for about an hour and I'll say for my part, it was a really enlightening conversation. We talked about the insider trading scandals and the suggestion from some academics that insider trading is the whole point of prediction markets. We talked about how these markets might get regulated. Lawmakers are writing laws. Cases are headed to the Supreme Court. We also looked at who's actually actually making money off of them and when to trust the odds. You see, Like I said, head over to GDPolitics.com to become a paid subscriber and catch the whole thing. It's 8 bucks a month or 80 bucks a year. Paid subscribers get about twice the number of episodes, can join in the paid subscriber chat, and most importantly, ensure that we can continue as an independent podcast, prioritizing curiosity, rigor, and a sense of humor. When you become a subscriber, you can connect your account to wherever you listen to podcasts so you'll never miss an episode. There's a link in the show notes explaining how to Again, head over to gdpolitics. Com. See you there.