Hopefully everybody's fueled up. You got your food, you got your caffeine. You've taken care of your biology breaks. We want to get started. We've got an exciting second half, and we're going to keep feeding in. We're going to keep imbibing you later. We'll have more on that as we go about here. We've got a couple of incredible conversations, including the one we're about to have right now. We'll bring it back up. My friend Contessa Brewer of CNBC, and she's going to be in conversation with Jason Robins, CEO of DraftKings. He's not paying me to say this, but, hey, instead of the stage is yours, the crown is yours.
Aw.
So there.
The crown is not really mine because the crown is over there in front.
There you go.
In the front row. I already tried to
I like it
steal that, and he was like, "No way, man. Stay away from me.
Do you need me to open the water for you?
Yeah. Should we just do that now before I can't do it.
I think so.
Before I can't do it for myself? Jason, you get a lot of coverage from me and from CNBC for DraftKings and for sports betting and the business of sports betting and the way that investors make money or not. You get far less attention for your prediction platform because most of that spotlight goes to Kalshi and Polymarket until very recently. What is it for you that is switching the What's the catalyst for you now starting to get more attention for your prediction platform?
I think the fact that we have been able to, in a pretty short timeframe, put ourselves in the top three exchanges in the market. Remember, we only launched DraftKings Exchange a few months ago, and we're already the clear number three, which is great, and continue to gain share. I think the fact that we're now a relevant player in the market and look to be one of those that is on the rise, I think is giving us more attention. Obviously, that comes with also other things we've been doing along the way to get there. I think we've done a ton on the product side that really has put our product in a position where I think we have the best sports offering in the market right now.
Did your sportsbook improvements in product and tech inform the speed with which you were able to improve your product and tech in prediction?
Absolutely. There is a couple of reasons. One, a lot of the things we have actually built are repurposable, and that actually means, yes, there is some adaptation, but you do not have to start from scratch when you are creating it. Secondly, though, there is a lot of know-how, and that is actually the most important part. We know what works. We know what customers want. We understand how to talk and message to those customers. I think that is something that over many, many years was built up through a lot of experiments, a lot of trial and error. So it is a huge advantage for us, I think.
You said you have become the number three exchange right now on sports only or overall?
No. I think overall, but definitely on sports, but I believe overall.
Okay. So how can that be possible when you are not offering your prediction platform in New York, for instance?
I know. Imagine if we were, right? We would be even bigger. So I think you are right that we have done that with a much more limited state footprint than all of the others that we are talking about competing with. So that is a testament to the fact that we have gained in popularity is the fact that in less than half of the country, we are able to put ourselves in the top three with others who are all competing in pretty much the entire country.
Yeah. I will just mention that when you guys moved into predictions, you made it clear you would not offer predictions on sports in states where you have a sports betting license, right?
Yes.
When you look at that decision overall, was that the right move to make?
Oh, absolutely. Of course. The alternative was we would have had our licenses under threat, and I think that wouldn't have been a smart move. It's a different thing. I think the positive for us on this is that we're not all in on this, which I'll come back to in a moment. We do have decided that we can't just say we're going to ignore this entire business that we've built up that most of our customers, most of our revenue is coming from, unlike some of the other predictions players. I think for the same reason, that's put us in a place where, unlike some others, where I think the outcome, assuming this ends up going to the Supreme Court, I think on both sides, there are companies that have really staked a lot on one outcome or the other.
It's a pretty big deal for them whether it ends up going one way or the other.
You mean whether sports gets stripped out of the prediction market?
Yeah, exactly, or any other type of restriction. Meaning, if that's the whole business, which for us it is not, then that's obviously critical. There's nothing to go back to. The entire sports business is gone. On the other side, if you haven't done what we've done and invested and put ourselves in a position where we're one of the leading players in predictions now, then if the court decision goes the positive way, then that's a pretty tough spot to be in. So we actually feel like we put ourselves in this win-win spot where no matter what the outcome of that whole process is, we are well-positioned going forward. I am hopeful that it ends up being something that is allowable over the long term. I also feel like if that doesn't happen, we're in a great spot.
One of the reasons why, as you mentioned, was we've really worked hard to create a really respectful relationship with our regulators to let them know that we aren't going to operate in the markets where we do sports betting, and I think that's been very well received.
I was in Macau interviewing the NBA Commissioner a couple of weeks ago, and I asked him about prediction platforms and what his stance is on it, because it was right after the NFL had filed an amicus brief with the Supreme Court making it clear they're not into it. Adam Silver told me that he has a lot of the same integrity concerns that the NFL does, but that he sees value in this being federally regulated rather than state by state. He goes, "It would be easier for me not to have to go through 40 different distinct jurisdictions." Would you rather have sportsbook regulated in all 50 states and available to you for license, or prediction platforms in all 50 states?
I would like to be able to offer all of our products to any customer that wants it. I think that
Are you saying if you had the ability to offer sportsbook in 50 states, you would still want to offer prediction platforms?
I think so, especially in non-sports. There is still a ton of content there. We just talked about elections earlier, and I think that is going to be a big deal during the midterms and obviously, a couple of years later, the presidential election. I think the financial markets have proven really interesting to us. We have actually been pleasantly surprised at the growth there, and I have got to admit, I did not expect that. I was under-appreciating, I think, that piece of the business. To me, that is an exciting thing, not just because it is a 50-state product for us, but it is also something that could be available even if the Supreme Court does not allow sports.
When we talked about the potential for the Supreme Court to crack down on the sports and prediction markets, in fact, at DraftKings Investor Day, I asked you whether you would be interested in all these other markets, and you are like, "Eh, not really.
I did not think so at the time, but
What changed?
Just data. We're always going to try everything, and if it were built on me guessing ahead of time what was going to work and not, we wouldn't be anywhere.
Predicting.
Predicting what was going to work or not, we wouldn't be very far because there's been so many times where I thought something was going to work and it didn't, or I thought something wasn't going to work and it did. And real testament to the team that they said, "Yeah, I think there's something there." We put a little bit of effort and we're seeing some signal, and now we're going to put more effort into that part of the product.
When you look at what you've offered, so now in New York, I can go on and I can make a bet on a prediction market.
A trade, but yes.
Okay. Tomato, tomato. When you look at that offering, are you seeing interest from sportsbook onto that particular event contract offering?
Well, I think mostly the people that are doing sportsbook are focused on that. We have seen some that are trying other products, fantasy sports, states where we have casino, and certainly in states where we have election markets and things like that, financial markets, we do see that. But I think really, that hasn't been as big a focal point for us, and the reason is, and this comes back to what I was saying a moment ago, we typically will throw things out there, see signal, and then invest further. I don't think we historically had looked at those non-sports products and said we feel that that is the right level of quality and that's the offering that we want to really start heavily promoting.
I think once we get the product in a little bit better place, which shouldn't be too long, I think we're talking months, then I think we're going to end up pushing it a little harder. But even without that, we've seen great signal, which is why we're so encouraged and why we're investing more in the product.
I'm just curious, with the entry into prediction markets with sports, has that created any complications, say, with the NFL? Because you're an official gambling partner of the NFL. The NFL has said, "Here's the amicus brief. We don't like easily manipulated markets, kicks, and missed balls and things like that." You're not offering that anyway because you're the official league partner.
Yeah. I think a lot of the concerns the NFL has raised, and this is why the NFL has been, I think, supportive of the legal sports betting industry, is it is a very collaborative relationship between the league and companies like DraftKings. We do discuss those things. We have arrangements with them. I think a lot of the concerns that they have, we are already doing the right things to cover. Now, obviously, as a category, they are not allowing predictions now, so we cannot advertise that directly through the NFL. I do think that we built a collaborative relationship where they feel more comfortable that we are taking the safeguard measures that they feel are necessary and maybe do not feel as much that way with some of the other participants in the market.
What are you seeing in terms of efficiencies in your unified wallet? Now when I go on DraftKings, I can get to Jackpocket, I can get to predictions, I can do sportsbook, I can do iCasino in New Jersey. What are you seeing in terms of your costs and where you are seeing benefits?
Well, I think it starts with the customer. What we had realized was, and this is not just true of us, but a lot of companies have historically said, "Okay, I have a different product, I am going to have a different app." Maybe still you have some unification with accounts or wallets and things like that. Just starting with the stance of a different app is creating more work for the customer to figure out, "Oh, I got to go there to get this. I got to go here to get that." Sometimes, as those of us working on it are so close to it, we are not really understanding that customers are just vaguely aware of the different products that are out there and do not really understand the hierarchy of how they are organized.
We started with saying, "Let us not make it the customer's problem to figure out where they find products. Let us make it our problem, and let us let the customer know they can go to the same app and they are going to have those things available." Part two of that was, okay, that is only so good if I still need to sign up for a different account and make a deposit into a different wallet. Still better to have it in the same app, but it is still creating an experience that is not ideal for the customer who is thinking, "Why do I have to put more money in to use DraftKings? I have already got money on DraftKings." So that was kind of the second piece of this evolution, but it is really rooted in how do we make the customer experience as easy and as positive as possible?
How do we make it so that things that people want to do, we are not making it so that they have to then do things they don't want to do, like make additional deposits in order to do it. We are allowing them to do it in a way that's comfortable and works for them. There's also, I think, an important point here, which is a lot of why I think companies went in this direction of different apps and different things in the first place is historically, it was a lot of load to have all these different products and all these different things in the same app, and you would really see performance deterioration if you tried to put too much in.
Now, we are able to, with what's possible today, create a very fast, very easy-to-use experience and still have all the content from all these different things in there. It also used to be that it was very overwhelming for customers to try to figure out, how do I find this or that? Now, we're able to more easily say, "Okay, we know if you're interested in elections, we're going to show you elections content. If not, there's a way to find it, but we're not going to clutter the screen by putting everything in there at once." A lot of, I think, what technology advancements, nothing to do with us, but just the underlying tech behind it have enabled, have allowed us to pursue this strategy, and I think it's the direction that most companies are going to go who have multi-vertical offerings.
Is AI helping you do that?
Yes, absolutely. Of course, we are using AI responsibly, despite what maybe some others will say, but I think it's something that is a very important part of creating a great customer experience. Customers expect that. They don't expect you to be serving them content that's totally irrelevant to them. You're a Brewers fan. If you're consistently betting on the Brewers, I'm not going to start showing you Tampa Bay Rays and trying to get you to bet on that. That's not going to make you very happy as a Brewers fan. I think things like that are just creating a much better and simpler and more, I think, what the customer's expecting. If we're not using the latest technologies to do that, we're not going to live up to our customers' standards.
They are expecting us to create the types of experiences that they are getting on all, forget gaming, on all the different types of products that they're using.
I'm like that Robert Frost poem. I've come to a fork in the road. I can either go down this path or go down this path. I want to talk about the Brewers because
I know you do.
Jason's right. I am a Brewers fan. It's not just because my last name is Brewer, and I can explain all that. I want to
You are a Packers fan, too.
A Packers fan and a Bucks fan. I want to tease out this thread a little bit. I will say it. The New York Times did a story that DraftKings uses AI to target losing customers. I want to know, one, can you even have a business if some customers do not lose?
Well, I think if you replace the word "lose" with "we need customers to spend money," then yes, and that is true of any business. Businesses do not work if customers are not spending money. I think one of the things that is a challenge is that there are a lot of people out there, all of whom really are not people that gamble, that do not like it. I do. I like it, and I want to spend my money. I think it is something that brings me entertainment, and I do not view it as losing money. I view it as spending money, the same way if I go out to dinner and I order a nice bottle of wine and end up spending $100 more than I normally would have if I had eaten at home.
I still like that experience, and I view it as something that is worth my money and I want to spend on. I think if you start from the stance that somebody who is maybe taking that position is a believer that the only reason anybody would ever want to gamble is because they have some sort of addictive personality, because why would anyone want to play something where they know they are going to lose money? You are ignoring the fact that that is not really what it is. There actually are people who like these products, who enjoy them. The vast majority of customers fall into that category.
Those people want to be able to get the experiences that come with spending money, just like if you went and stayed at a nice hotel and spent a lot of money there, you would expect them to take care of you, too. I think any rewards program on the planet, any loyalty, anything like that that you see is generally geared around rewarding the customers that spend more with you.
How does AI make marketing to customers and potential customers easier, better, more efficient?
I think the degree to which AI is used is a bit overblown. I think it is one of those things that is thrown in article headlines because it has got a lot of attention now, and people are generally fearful of it. But I think if you ask a lot of the people who are commenting on it, "What do you think that means? What is AI? What do you think us using AI means?" I think you get 10 different answers from 10 different people. So, I start from the stance of I do not really view AI as what you are saying. I view AI as the thing that is helping us write code faster and QA code and all that sort of stuff.
I think we are using machine learning in many cases to be able to say these customers have these preferences, example, you like the Brewers, and that is in the effort to serve relevant content to them. Not in an effort to try to target people to get them to do things that they should not be doing. In fact, our entire company philosophically is built around the opposite of that. We tell people that is not the way to create a sustainable business. It is not the right thing to do. Obviously, there are people that should not be using the products, and we do our best to try to help identify those people. We cannot do it perfectly. But we are certainly not trying to proactively target those people. That is crazy. It is actually kind of hurtful, too, to be honest.
I feel like people don't understand that when articles are written like that, there's a lot of people working on these things who unequivocally know that the way they're being characterized, the way we're being characterized is just not right. They're actually trying to do the opposite, and it makes you feel bad when people are out there saying we're terrible people, and it's just not. If you ask anybody on our leadership team anywhere throughout the company, they will tell you we preach responsible gaming. We have a whole team that we refer people to when we, instead of targeting them, we actually do the opposite. If we think somebody may have an issue, we refer them to a specialized team that looks into their case. It's just completely not how we're built and how we philosophically and in practice operate.
But listen, I think, again, it depends on the person. Some people just don't like gaming, and they're like, "Well, by definition, you're going to have to target people in a nefarious way because who else is using the product?" I can't really change their minds on that if you're starting from a stance of we just don't like the industry, and we think that you're bad people.
To that, you never get that from me because I also like gambling, and roulette happens to be my game, but I find it entertaining, so I don't have this moral high horse to climb. But when I read ProPublica's article written from the first person, "I'm trying to lose," and "When do the responsible gaming tools come into play?" or whatever, when that hits, is there ever for you this moment of, "Oh, we could do this. Okay, I'm going to learn from this. Here's what we need to do better," that you go back and you talk to your team and go, "X, Y, Z, let's do this differently in the future"?
So when that article came out, I went back. I personally looked at the entire history of that customer. I looked at all the notifications we sent. I looked at all the communications with their VIP host, and I was pretty proud of how we handled it. First of all, it wasn't mentioned, we sent hundreds of responsible gaming notices throughout that entire experience. Number two, the VIP host repeatedly said, "Hey, do you want to take a cool-off period? Should you set a limit? Are you okay?" I think sometimes what people want us to do, and they're like, "Nothing less than this is okay," is just cut people off. We do do that sometimes, but it's not a clear-cut case because sometimes people who show the exact same patterns of play, one of them has an issue and the other 9 out of 10 don't.
A lot of the things he was saying he was doing, I was reading, I was like, "Yeah, I do that." I stayed up late and was betting on random international sports in my day and playing casino games in the middle of the night. I thought it was fun, I liked it. Never caused me any financial problems, never had an issue with it. Maybe I didn't get enough sleep a few nights than I should've, but I could see how somebody else who has the exact same pattern, though, may have an issue. It's really tricky.
What we really try to do is we try to identify those situations and get people to then take action themselves, partly because we don't know for sure, and also partly because if people do things themselves, they're less likely if we restrict them, more likely to just go jump on another platform. If they actually exclude or cool off themselves, they're more likely to actually adhere to it. So, there were just so many things that were left out of these things, and again, nothing I can do if somebody's starting from a stance of they are trying to write a negative article. Really hard to get them to go positive.
Maybe you can give them some facts and make it a little less negative, but if the goal at the start is "I want to write a negative article," then you're probably going to see somebody write a negative article.
Jason, do you see the same tools being needed, the same approach needed for prediction markets that exist for gambling?
Absolutely. I think for one, we're putting our money where our mouth is on that. We do do that. We apply the same responsible trading tools as we use as responsible gaming tools in our predictions product. We take the same approach with, we have the responsible engagement team that gets referrals for people that we think need to have their accounts looked into, things like. It's literally the same playbook. I think this is a great example of where you don't necessarily have to reinvent the wheel here. Again, doesn't mean we can't improve on the existing system, but there is already a decent system in place for the regulated OSB market. People who don't understand that, they just don't have the knowledge. There is so much regulation already in this area.
Yeah.
There are very specific rules that many of the states give us that tell us how we have to manage responsible gaming. Now, we go above and beyond that. For example, we don't have to do that with predictions, but we are doing the same things that we do in OSB with predictions. There's also, at least in online sports betting space, very clear rules on a lot of this stuff.
I was in Las Vegas for G2E, the Global Gaming Expo, and I saw an ad for DraftKings in every state including Nevada, which has told Kalshi that it has to geo-fence the state. Tell me how it works where you're. Not just the states that you have a gaming license, but how does it work when you need to geo-fence out other products in other states?
We do not offer sports predictions in Nevada. We do offer some free-to-play sports contests. I think for a national marketing message, it's easier to say DraftKings is available everywhere than DraftKings is available in 48 or 47, or whatever it is, states.
Right.
We have disclaimers. We actually worked with the Nevada regulator, who is a great partner in this, to understand what their concerns were. We were able to put some disclaimers in our marketing. Ended up. Surprised you saw it in Nevada because we are trying to not target in Nevada, so that is probably a mistake.
Probably on a national program.
It might have been a national program, as we definitely are not targeting Nevada specifically with ads and trying to make it as clear as possible that you cannot play the paid product in Nevada. We also have to remember with national ads, you need a national marketing message, and the more you confuse people, the harder it is to get that message to work.
In Florida, where you fought hard to try and get on the ballot and get the opportunity to offer commercial sportsbook. Now you have prediction platform, and Hard Rock, which has the monopoly in Florida, is suing you. Why you and not FanDuel or Fanatics or Kalshi and Polymarket?
I don't know. You'll have to ask them that one.
No, I just want to know what's your thought about why you're the target?
For whatever reason, they decided to make us the target. I can't speak to why. I don't know what their motivation is.
That's boring.
I am sorry.
I was trying for something a little bit meatier.
But you are absolutely right that it is very interesting that despite the fact that there is a number of prediction operators in the state, we are the ones that singularly got sued. I was a little surprised they did not include others in the suit, but you will have to ask them that.
When you are looking at the economics of predictions and not having to pay, say, in New York, 51% tax on your business here, is that model super appealing for you? Would you like to have a departure from the state-by-state, "How much money can we wring out from the sportsbooks?
Well, I don't view them. Assuming predictions were allowed, we would want to offer everything, because it comes down to we want to give the customer choice. Obviously, we would like to have level playing fields with taxes. My guess is that what you're describing is not going to exist long-term, that if there ends up being, Supreme Court or otherwise, a more permanent sports prediction market, that states will turn to taxing it like they do sports betting. That's my guess anyway. I look at it as, again, starting with the customer, and I do think that at least now, and I think arguably, in perpetuity, the sportsbook product is a better product for a lot of things that the customer is looking to do. You can look no further than Europe to see that.
There's been exchange-based betting in Europe for many, many years, and it's a small piece of the market. It's really dominated by the traditional sportsbook market, and the reason for that, I think, is it's just a superior customer experience. I believe that to be the case today. I think unequivocally, the sportsbook experience is a better customer experience for most customers. Obviously, there's some that maybe prefer predictions, but for most customers. That said, there are definitely some elements to the predictions product that are better. While maybe not as an overall, some elements, and I think those products, including our product, will continue to improve. Again, it comes back to I'd like to be able to offer different options to all our customers.
An example of predictions we talked about, you can do all kinds of non-sport stuff that you can't do in sports betting, and I think that's really interesting, and there is an audience for that.
Institutional trading, do you see a future for DraftKings in accommodating institutional money?
That's an interesting one because I think, my personal belief is that the more of that you introduce into a peer-to-peer ecosystem, the faster that the retail customer will have to spend money in order to keep up. I think that that may create a worse customer experience than maybe having less institutional. We can't prevent them from being on the platform, of course, but we have, at least to date, chosen not to invest in product features that are geared specifically to algorithmic traders, to institutional investors, because we do think that creates a healthier long-term ecosystem.
Okay. I'm sorry, but we have to talk about the World Series. The Brewers have not made it this far since-
I knew we wouldn't finish without this.
Since 1982, and there's so much enthusiasm. My kids are trying to break the school uniform rules and wear their Brewers hoodies to school every day. Where's the money coming in? Tell me a little bit about this small market team, doesn't get to pay its players like Shohei Ohtani money. Talk a little bit about what you're seeing from the business side about Milwaukee Brewers being in the series.
Yeah. It is interesting you have two teams like that now.
Ooh. I am cheering for the other half.
The Rays. Yeah. No, it is a good sign because it shows that even though, yes, it may be less common that teams on smaller payroll budgets can still compete in baseball, which I think is very important for long-term competitive parity in the sport. That said, to answer your question, it is really interesting because we were just looking at this, I know, together before the panel. The vast majority of the money, almost 90% of the money for Game 1, is on the Rays, but for the series, it is on the Brewers. So, I would have figured Who is pitching Game 1 for the Brewers?
Got me.
Yeah. Well, maybe it is the I did not look at the pitching matchup. I should know that. But maybe that has something to do with it. But it was interesting. I was surprised to see that it was so heavily tilted one way on Game 1 and the opposite way on the series.
Yeah. Well, just out of loyalty's sake, I would think that my So, I said it is not because my last name is Brewer. It is because I lived in Milwaukee, and in Wisconsin, if you do not get inducted into the cult of sport, meaning Brewers, Packers, Bucks, you get kicked out, and I married a Wisconsin boy. I did not want to get kicked out, so
No
Yeah, that was the way to do it. Do you think
Well, now is your moment, Contessa
One last one for you. This is it. Do you think that the investors have really punished DraftKings because of the perceived threat from prediction markets? Do you think when you go out and you talk about your performance on prediction markets, you start to get more credit from shareholders?
That's a great question. I would have said yes several months ago. I'm not as sure now because I think people just don't love the uncertainty of, is this actually going to be a thing? So, great if I'm underwriting growth in the next year and a half, but is that all going to go away? I think the more that people come to understand that we're really well-positioned either way, I think we will start to overall get more credit for that, and I think people are starting to understand that. I think the more that we can show growth in the non-sports piece, which people, for whatever reason, think is probably safer, I think that'll help, too. I also think just getting to a point where there is a court ruling and there's just some clarity and there's You know this. Private markets love uncertainty.
That's what they thrive on. They like big beta. Public markets are the opposite. They do not like uncertainty. They want predictable, steady earnings, things like that. I do think that we are suffering from the uncertainty of it all, and I hope that we start to get more credit as we show traction in predictions. But, we might not really get full credit until there's more clarity.
Earnings next week?
Next week.
All right. We will be listening. Thank you, everybody, for your time.
Actually, week after next, but-
Appreciate it. Oh, two weeks. Yeah, two weeks.
All right, the show will keep going on. Our next panel is going to deal with, it is one thing for a big event to get somebody to show up to your market once. How do you get people coming back over and over again and doing so ethically on that? Our next panel, please welcome Kim Yuhl of NEXT.io. She will be leading this panel. Kim, the stage is yours as soon as we get out of that choke point.
Are you okay?
Yeah.
I think they wanted me to.
Tomer?
Yeah.
Under your name.
Oh.
Okay. Welcome, everyone. I would like to start with a number, 3 billion.
Hey, Joey. Nice to meet you.
That is what Kalshi traded at this time last year in one month. Last month, they exceeded $60 billion, and this month they are surpassing that number with more than a week to go. Getting people in the door doesn't seem to be the problem, but keeping them is. Our most recent public data comes from Polymarket's 2024 cohort, where of new wallets, about 50% were active in two months, but only 17% in three. What that tells us is that many first-time users are settling, cashing out, and not coming back. Just this morning, JB Mackenzie from Robinhood was saying that prediction markets has a perception problem. I can't help but wonder if that perception problem is part of the customer retention problem. That's what we're here to discuss today, the second trade.
What happens after that first event settles and people are going on the next step of their customer journey? I'm Kim Yuhl. I'm sorry about that, Director of Publishing and Affiliation for NEXT and NEXTPredict, and I'm here with this esteemed panel who is going to share some of their insights. I'm going to start, we only have a small amount of time, by throwing it to them so that they can introduce themselves, their company, their products, and give us their initial insights on customer retention. Tomer?
Sure. You hear me okay? Hi, everyone. Nice to meet you all. I'm Tomer. I'm the Head of Americas at Optimove. We're the number one customer engagement platform in this industry. We work with more than 500 brands globally, helping them to retain their players, customers, traders. We work with sportsbook, iGaming providers, and recently with prediction markets as well.
Hi, my name is Hanna Jaff, and I have a prediction market launching in Mexico. We're going to be the first ones there. My background is in politics. I used to be a politician in Mexico and influencer, million followers. I have experience in retention and growing brands and companies. Yes, thank you.
Hi, everyone. My name's Joey Levy. I'm the Founder and CEO of Betr. Betr has launched the first real money gaming super app. All within the same app in a consolidated wallet experience, we have Betr Picks, which is a fantasy sports pick-em product live in 34 states. We have a social sportsbook live in 36 states. We have a social casino live in 35 states. We have Betr Arcade, which is a peer-to-peer skill games product live in 38 states. Relevant to this conference, we soft launched our prediction markets product in a handful of states a few weeks ago and became the first consumer company to launch Polymarket-powered prediction markets. We've also publicly shared that we plan on getting into the digital instant pack ripping space. We're going to launch Betr Rips over the next few weeks here.
Once that goes live, we'll have six products live in the super app, and five of which we will have launched in just the last 15 months. We're really seeing phenomenal product velocity across the business, and we're growing really quickly, and we're excited about the opportunity in prediction markets, obviously. It's great to be here.
Joey, I am going to start with you. The second trade. Somebody's event settles, what goes in front of them next? What are you doing to keep them active?
Yeah, a lot of it varies by product, but I think what differentiates Betr relative to a lot of the other direct-to-consumer prediction market businesses is that we are very well diversified across a variety of different products, and we offer a lot of different products, right? Prediction markets, as we all know, I am assuming everybody here is obviously quite familiar with the product, right? It is a rapidly growing industry, but there are open questions relative to monetization and how much of the economics does the exchange get versus market makers who are doing really well. Kalshi obviously has a tremendous amount of market share, and others are trying to catch up to what they are doing.
I think it is pretty obvious that prediction market companies are ultimately going to need to have a suite of other products to be able to both retain customers but expand revenue per user for these customers, in addition to the fact that they need other products to hedge a lot of the regulatory risks we are dealing with. Obviously, an elephant in the room here is that SCOTUS is going to make a pretty big decision over the next couple of years, most likely with respect to will sports contracts actually be permitted the way they are today? That is a pretty material risk. So I think that there needs to be a level of thoughtfulness around hedging the regulatory exposure. It is not just SCOTUS, by the way. You also have a very favorable CFTC right now, but of course, that could change too.
I hate to be the guy at the prediction markets conference to talk about all the risks to prediction markets, but I think operators in this space need to think about what else do you offer besides prediction markets, while also preparing yourself for the opportunity to capitalize on the prediction markets opportunity if it does survive SCOTUS and the CFTC continues to be as favorable to the industry as it has been to date.
Very good. Hanna, you are getting ready to launch a platform in Latin America later this year. Obviously, customer retention has to be front of mind. You mentioned it in your bio. With that said, what are you doing now to be ready for that when you launch?
The first thing is you need to know your target audience. First off, we would launch our first vertical is sports, then it would be politics, entertainment, culture. You need to target those communities first. What I have seen in other marketing campaigns, what we have done is basically you need to get to know your user to keep them interested. Prediction markets are like information marketplaces. Some people go on prediction markets, and they do not even trade. You keep them there because it is news for them. They log in, they like a certain category, and that is how you retain them, through those special verticals. Marketing and retention is very important. You have to create trust, and that trust is obviously very important for any prediction markets and what gives it that retention, that credibility.
The trust, credibility, and understanding the user is what will keep those big prediction markets alive and active.
I am going to do a follow-up, hopefully not putting you on the spot, but from that talk with JB Mackenzie from Robinhood, he was talking about that perception problem. A big part of that is building trust with the customers. You guys can all answer this, how are we educating the customer that this is trustworthy, especially when the regulatory environment is sort of in limbo right now?
I think that it is important to get people that are already trusted as spokespeople behind to back it, I think that that is step 1. Many of these platforms, obviously, are new to users. Users don't know what prediction markets are, especially I am going to a country where people don't know what prediction market is. You are not only getting them to trust you are actually teaching an audience what it is. It is much harder. You need to get athletes or influencers or people that are already trusted to back it to sort of give you that credibility. That is basically what we do. What we are trying to do, we are going to target those communities and those people that can back those communities so those users will feel comfortable.
Obviously, having those payment rails and everything local to the local users, so they feel comfortable putting their personal information in and just those are sort of the tools you can use to keep those people comfortable and have that credibility.
Tomer, do you have any thoughts on that?
I think it is actually you mentioned Latin America, so I feel like I have to mention Brazil when it comes to trust and regulatory environment with what is going on right now there. We see what happens when actors are overplaying their game, if you will, and are trying to get the most out of the markets and how that could backlash. I think that stability is going to be very important. I would be concerned about prediction market in that aspect in terms of what happens in two years from now. If we can provide, as an industry, a stable environment that deters any bad actors away from it and makes it kind of like a very strong regulatory environment, we could build that trust that obviously is key for the long-term success of that industry. Otherwise, someone will shut it down.
Okay, I am going to stay with you for a second. We know that World Cup, Super Bowl, the big events are the big acquisition moments. We have an opportunity at that point to bring people in. Do people that come in for those events behave differently? Do we retain them differently than people coming in for elections or the Oscars or something maybe not as significant?
I think the World Cup is a good example. We have been speaking to some of the leading prediction market operators today around the World Cup, and they are all very busy with it, and it obviously generated huge amount of traffic. I think the bigger question for me is whether we are going to see the prediction market behave in the same way that the real money gaming space was six, seven years ago. I mean, Betr was micro betting app in the early days. I remember sitting in the crowd when you were talking about it, and things changed since then. I think that the gold rush that started with PASPA passed five, six years ago, it had a backlash. A lot of the companies had to change course, had to shut down. The unit economics did not work, and I think that is the bottom line, the unit economics.
When you are able to acquire players easily through a World Cup event, an NFL season, or whatnot, can you retain those players? Could you make the unit economic work for you in a way where the lifetime value works against the cost of acquisition? What we saw five, six years ago in the real money gaming space is that the answer was no, that most of those operators that raised millions of dollars had to shut down, and today probably 80% of the market share is with five operations. For me, the ability for an operator, for any trading company, if you will, to make the unit economics work to retain those customers after they make their first trade, is going to be the differentiator between the winners and the losers in that space.
Joey, with a product that has multi verticals, how do you capitalize on that? How is your strategy different in each of those verticals, but allowing you to capitalize on a shared wallet?
Yeah-
Oh, you go.
Sorry. That was directed to me, right? Your initial question to Tomer was exactly why I think the super app strategy is the superior strategy and why I think most of these products and businesses will ultimately become super apps because you can use these tentpole moments like the World Cup, like the Super Bowl, like political elections, to acquire customers. But you need to be able to pay back those investments to acquire those customers, and if they are only transacting around those moments and then they churn and do not continue playing, you are going to struggle to pay back those UA investments.
When you have a full skill games vertical or when you have fantasy pick'em or when you have social casino, pack ripping, all these other things we plan on getting into, you are able to retain customers and continue to monetize those customers. A big part of our strategy is we have a great sports business, proprietary picks app, proprietary sportsbook product, and we spend into NFL, we spend into NBA, Thanksgiving, Super Bowl, et cetera, and then we cross-sell a lot of those customers into casino, where you really expand revenue per user. I think a similar dynamic is going to ultimately play out with the prediction market businesses.
It does not make a lot of sense for there to be so many different consumer brands and prediction markets that are offering similar products but do not have these other products to expand revenue per user and better retain customers.
Hanna, as you are developing your app, are there features that go into the product that help with retention, and what are those?
Obviously, the user that logs into a prediction market that is sports or entertainment is quite different. First off, for entertainment users for, I do not know, Oscar fans are going to just log in once a year maybe, or maybe twice a month. Whereas sports fans will be actively logging into the app. The cost per acquisition for a user might be a lot more expensive for an entertainment niche user than a sports. Of course, they say that over 80% of people on prediction markets are just using sports. You can have tools and features like live data services using AI to keep those users entertained and stay on the app. There is also referral methods, where you can offer people, I do not know, a $10 voucher if they bring a family friend or they share it on their social media.
That same community you are building is growing the community. That is sort of the tools you can add, and as well, there are other marketing channels you can use as partnerships with other sports or entertainment communities. That is sort of one of the features you can add.
Joey, any comments on your product features that are specifically designed for that kind of retention?
Yeah, I am going to sound like a broken record continuing to kind of tout the super app strategy, but you could use generosity from one product to help retention for another product. I think, again, another reason why consumer prediction market platforms, I think, are going to need to go in this direction is the economics around prediction markets being a much lower margin product make customer generosity, which is the key part of the customer retention puzzle, at least in real money gaming, a lot more challenging. They are lower margin. The market makers are making a lot of money that the exchanges would otherwise want to retain for themselves.
Having these other products where you could be more aggressive with customer generosity is going to be really important for customer retention, and that is why ideally, you want a casino product, you want a DFS product, you want a pack ripping product. Then you could use the generosity from these different products to subsidize one another, improve cross-sell, and expand revenue per user.
Any thoughts?
Yeah. I think, well, at first, I will say I sat in a couple of panels today, and no one said AI, which was kind of refreshing. But I am going to be the first one potentially who does it. Look, I think that from my perspective, again, being so many years in this industry, I feel that there is always this idea here in North America that we are going to try and bring the customer, the player through one channel, through one way, and then we are going to convert them down the line to become a long-term player through whatever it is, social or micro-betting or DFS or whatever that may be. Then later down the line, we are going to make money out of them by bringing them into a more profitable product for us as an operator. Largely, it was a challenge.
I think that AI actually solved for that. I think that when we talk about even in my view, in the world of super app, you could probably use AI internally to better understand your customer preferences and come back to them with the right content in order to move them from one product line to another product line at the right moment, right? That also is very much applicable to prediction market, which you mentioned someone's coming in from a perspective of maybe the World Cup. You acquire them through that. You look at their behavior, you look at what they play on, what kind of they bet on, and you try to move them over to maybe a political market or anything along those lines based on their affinity and preferences.
I think AI can really help break those silos, if you will, that before that were very much hard to overcome from just a technology perspective, data analytics, the silos between marketing department and the data department and so on in the organization.
So right now we're seeing a lot of the same sort of playbook from a sports betting app for the prediction market app to try to attract people, bonuses, push notifications, promos, that kind of thing. Is that the way forward, or is the prediction market trader significantly different than the bettor, even though they're coming in probably for a sports event to begin with? Do we continue to promote the same way? Anybody can answer.
There's so much overlap with the products, right? I know a lot of people in general, and probably especially at this conference, want to talk about these are super distinctly different products catering to different people. To some extent, that's true, but to a large extent, it's no coincidence why 90% plus of the volume in prediction markets is sports, right? It's just you're offering a similar experience in geographies that don't have OSB or may not have some of these other products, right? So I think just being honest about that is first kind of like, need to be honest about that to answer this question, which is a lot of the same kind of promos and methodologies around acquiring customers are similar. With that said, like I've said a few minutes ago, it is a lower-margin product.
There are open questions about monetization, which is why I think these will end up being great customer acquisition products. But you need a suite of other products in order to sufficiently monetize customers and then reinvest that gross profit back into more customer acquisition, which is why I think all of these businesses will eventually become super apps.
Is there a line where we cross into being more predatory, trying to get them to trade on more markets. If we're using the same playbook, the idea is we want them to stay on the apps and use more things. Is there a line there that we should be aware of?
I think so, yes. I think the behavior in the user is we can't exploit it. We can't send notifications or urge them to go on the app and use our product unless they want to. I think there has to be some sort of line and safety measure to it where the user doesn't feel attacked or obligated to stay. Yes, I think AI is important for the safety of the apps, where if a user wants to stay on it's the social part of it. It's because it's fun. It's not like the old traditional casino betting where you just go and play with the house. You're actually seeing those trades, those markets, those charts. You're commenting, that social aspect of it with other traders. You're fighting about whatever topic or question or market you're talking about.
Yes, I think that it's quite important that those safety measures are taken seriously in any platform.
Who is responsible for that? This is the big question right now as we are in some legal limbo, at least here in the U.S. But who is responsible? Is it the operator? Is it the regulator? Is it the player? Who is responsible for those guardrails?
Well, technically, the CFTC does have some limitations on marketing and promotions, but I feel like it is changing, put it that way, and it is becoming more and more towards the sportsbook side over time. We will see how far it goes. But yeah, look, I think that in the same fashion that we expect real money gaming operators to be responsible and use the technology in a way that protects the player. For me and for us as a provider in that space, that is always a default situation, right? And we would put the same technology and modeling that we have to detect and work with VIPs to detect and work with at-risk customers to make sure we carve them out from that sequences, and we cater for that situation.
We only have a few minutes left. I have a couple more questions. I am going to try to get to them. As barriers to entry fall, competition is going to get tough. We know it. We expect it. What is going to make it so a company sticks around? Joey, I am going to guess you are going to say super app, but I do not want to put words in your mouth, so we will start with you.
Yeah. No, I think you need products that really monetize the customer extremely well and pay back very fast so that you could take that gross profit and reinvest it into growth. And we have seen phenomenal scale in our business deploying this strategy. Then really, once all the leading consumer companies have all of these different products, then it will be a game of who can spend the most on marketing and have the best brand. Right? And we were talking about trust earlier, and your ability to invest in brand is your ability to develop trust with the consumer. And there is obviously some companies that have already reached a phenomenal level of scale and have established that trust, and we are well on our way to accomplishing that ourselves.
Despite being a relatively new company that is only been around for four years or so and has only had the super app live for about six or seven months or so, I think we have done a pretty good job on that front. You could raise all the VC money you want, but at some point, you are going to need to demonstrate phenomenal unit economics, and you are going to need to demonstrate an ability to generate a pretty massive amount of gross profit off of your returning customer cohorts to then reinvest back into growth. We are very excited to be in that spot ourselves right now and growing really robust triple digits year-over-year on the revenue side at a pretty significant level of scale. We are going to keep doing what we are doing on that front.
Hanna, Tomer?
Also, this is a live prediction markets, right? Every day, events are going on. Who can promote those events the best, right?
Right.
Every day, it is like a newspaper. It is the news, right? It is who can basically get those users entertained enough to keep logging in and keep predicting what is in those trades every day.
Tomer?
Yeah, I think that's to Joey's point as well. It doesn't matter how much money you raised, what's the valuation of the company, who is the celebrity you have, who's promoting your brand. The unit economics is going to be the king in that space. It's going to differentiate the losers and the winners, and a holistic retention program is going to be key for that. Whether it's product-led, whether it's marketing-led, this has to be part of your strategy to be successful.
Okay, five seconds. We are at a prediction markets conference, so I would like a prediction from each of you as to what we'll be talking about in terms of customer retention at this time next year.
It's going to be pretty much along the lines of what I said so far. I think that the signals I'm seeing right now are very similar to the signals I saw six years ago in the real money gaming space, which was a hot topic, which was a gold rush, and not much of it was left as of today in terms of market share and operators. I think in 18 months from now, we'll probably talk about how we make the unit economic work in that space, how do we invest, how we don't promote too much in order to make sure of that, and how we retain those customers in an effective and efficient and sophisticated way to make sure that the unit economic works.
You guys want to add anything?
I think that is super well said.
Yeah.
I think we are going to talk about how to make these-
AI. Yeah, how we are going to-
We are going to talk about how to make these businesses more sustainable via great unit economics. We are going to talk about product expansion, and we are going to start noticing how more of these consumer companies are becoming super apps.
Yeah. How prediction markets
To also hedge the regulatory exposure. I am sorry to cut you off. Go ahead.
How prediction markets are going to be more, 18 months from now, are even going to be more a part of everyday life. It is going to be our everyday newspaper.
Well, thank you guys very much. Thank you guys for being here, and that wraps it up for us.
All right, so we are transitioning from customer retention to coming up with new markets. Sports may drive the headlines, we know. Politics, elections, sports, stock price options, those have been the mainstays. How do we get other event contracts, and where does this go next? Our next panel is on just that topic, and I want to bring up Bob Fitzsimmons of Wedbush and his panel up here to take it away.
Want a water, man? Yeah. Well, why don't we get started? We have a wonderful panel here today. We've got an exchange. We've got two investors, two broker-dealers. It's very interesting, when we were waiting out there, I don't know if you could hear us talking, Chuck Todd had real strong opinions about this. It was almost like we needed an extra chair here for him to come up. My name is Bob Fitzsimmons. I'm an Executive Vice President at Wedbush Securities. We're a global multi-asset clearing firm. So we've been involved in prediction markets since the CME launched back in December of last year and have enjoyed sort of the tremendous growth that we've been seeing.
And just before we get to our panels, I just want to see from a show of hands, how many of you are currently involved in the prediction markets from your firm's point of view? If you could just raise your hand. All right. Pretty good amount. How many are just trying to figure out where they're going with the prediction markets? Okay, fair enough. So, why don't we start with introductions? Einar, why don't you start it out and we'll just go straight down the line?
Yeah, absolutely. I'm Einar Braathen. I'm a Partner at CoinFund. We're a venture fund focused on blockchain and prediction markets. We write checks up to $50 million and have invested in a number of companies in and around the space, including Edge and Ceni, who gave a talk this morning. So, excited to be here.
Matt DeCicco, I am with Cantor Fitzgerald. I lead our prediction markets business. As of August, we announced that we are acting as an introducing broker for these markets with plans to continue expanding our footprint in the space. I will speak a little bit, I am sure, throughout the panel, but we see tremendous upside, specifically on the institutional use cases and applications for these contracts.
JJ Kinahan, Senior Vice President, Cboe Global Markets. We have just started prediction markets, and we just got approval to list corporate KPIs. We will be the first ones on the security side of the business.
I am Rob Hadick, general partner at Dragonfly. We are about a $4 billion venture fund focused on primarily the future of markets and money. We have been investors in Polymarket for a few years now. I am excited to be here talking about prediction markets.
Yeah, great. What is interesting is I love the description of this session. It is basically saying, what is next? We are always trying to figure out what is around the corner, and they look at the frontier as everything else beyond sports, which is amazing. I immediately think of Ric Best uses the example, Ric Best from Susquehanna, about the goat herding hedging. Which is a great example of a use case of something beyond sports. But we are also interested in a product that kind of bridges the traditional prediction markets and futures, which is the future sports indices that have just listed at the CME, and they just went with the NHL, I think Major League Baseball is out there. So, something to keep an eye on and something we are very excited about. Now, Rob, we are going to start with you. You are a partner at Dragonfly, early investor in Polymarket.
Polymarket, I think their numbers are about 40% sports, 60% non-sports, which is, compared to your competitors, I think they are probably at 70%-80% sports. We see a lot of advertising in terms of on the prediction side. Tell us what you are seeing both from the Polymarket side, from your other investments. Is there really a business beyond the sports markets?
It is early, but I think there is no doubt a business. To your point, I think Polymarket numbers, if you look at the Polymarket US, is much higher from a sports perspective, relative to the Polymarket International. If you look at Kalshi and others, I think it is almost entirely sports at this time, especially when you look at count parlays in there as well. But we do start to see a little bit of the early innings of interest. Polymarket started really with this idea that you could have markets for everything. The kind of breakout application or the breakout market was politics. That is a place where you brought up the goat herder hedging, but it was really a political hedge that was there because it was about a tax rebate that existed in California for these goat herders, that they were focused on.
We are starting to see a number of sort of institutional use cases. We have seen things like concerns over data center build-outs. So whether or not there will continue to be tax rebates for these data center build-outs and potentially those going away if there is a change in administration. So when you look at, okay, well, how do I hedge all of these different pieces of the economy? Politics has been the number one piece that you have started to see people focus on. But beyond that, I think we will continue to see that growth, but we need people like yourself and like Cantor and other brokers to be able to bring that and educate those net new market participants.
Great. Matt, let us go to you. Cantor Fitzgerald, 80-year-old firm, pioneer in electronic trading. How do you guys fit into the prediction markets, and what are you seeing from your institutional client base?
Yeah, listen, I think the prediction market space is a perfect example of an area where a firm like Cantor, it is in our DNA. We are on the smaller side versus our larger competitors. We are entrepreneurial. We are able to skate to where the puck is going. We did that with crypto several years ago. This is another perfect example where, unlike crypto, I think this is a space where the institutional use cases are incredibly broad and almost immediately clear to a very broad range of end users. There are institutional kind of like 3,000 plus client base that we speak to on a daily basis for traditional capital markets activity, or like your macro hedge funds that are interested in economic-related contracts, fed payroll, CPI, GDP, election-related contracts.
We speak to credit funds that where historically there are alternative products to hedge risk, like the swaps market or CDS, rather than trading a CDS basket on 15 underlying issuers. The beauty of these products, and I intentionally refer to them as products rather than an asset class because the applicability really spans any asset class across the traditional capital market spectrum, is that you have the flexibility of historically what was bespoke OTC transactions within a fully regulated CFTC framework that traditional investors are already comfortable and familiar with operating within. Back to my example on the credit side, further end of the spectrum for more macro use cases, but you could create a bespoke hedge for a credit investor on a specific issuer having a credit event or a default. That creates a much more specialized and tailored hedge to an institutional portfolio.
I think that is one of almost unlimited possible use cases for institutional investors. While I think we are on the earlier stages and there is a broad spectrum of where investors are in their level of education on the institutional side, there are fast movers that have been watching this market for some time and are just kind of dipping their toe in. But again, I think the upside potential and our view at Cantor is unlimited, and I think this market continues to develop into just another derivative in the toolkit of an institutional investor.
Yep. JJ, let us go to you. A storied career in the financial industry, a much sought-after panelist. You are speaking not just here, but I think you are on some tomorrow, if I understand correctly. But the Cboe. You are at the Cboe, they have gotten approval from the SEC, cash-settled KPI binary options. Tell us what the Cboe's expectations are on moving forward with these products.
Well, I think Matt hit on a lot of it in the fact that I think the possibility is endless, but I see the path to how we get there a little bit different. I see the path being a lot more, if you think about how zero-day options started. They started with retail. If you think about it, these things that go to $1 are truly more of a retail product at the very beginning. Where you have retail, you have market makers. Where you have deep markets, you are going to have institutional players. I think Matt's 100% right. If you think about the KPIs of a firm, this appeals to everybody.
I think one of the things, I spent a lot of my career on the retail side of the business, and one of the biggest complaints we would always get from retail traders was, "You know what? I bought this stock, they killed it on everything. Earnings were great, net income was great, sales was great, every single thing was great, and the stock was down. I do not understand it. You screwed me." Be like, "Well, no, you screwed yourself. Congratulations." But in reality, it is frustrating because on the call, maybe the CEO said that the numbers were not going to be quite what the analysts wanted to see, so the stock would be gone. Those of you who followed Micron saw this exact example two weeks ago. They killed it on every single KPI. Everything measurable by them was higher and beat expectations.
Their revenue was up 300% year over year. However, they did not quite meet the expectations that analysts had for them going forward. For an individual retail trader, it is the most frustrating thing they can possibly do. It is not only, again, retail that can use it. Institutions can use this as part of their hedging programs overall. If you think about it with Amazon Web Services, you can just say, "This is a number we think Amazon Web Services is going to be, and we are going to offer yes/no, as well as an options market on these, too." So we think that there is a lot of uses overall, and using the regular zero DTEs on earnings day versus the KPIs individual creates a lot of cross-trading opportunities.
Okay, we will go to Einar now. Einar, you are a partner at CoinFund. I think an investor in Edge Markets. We are sitting on the Edge Markets stage here. We have had products out there like weather markets, climate products. They have been out there for decades.
They have a niche little market.
What is going to be different today? What is changing that is going to make-
Yeah
These contracts more prevalent?
Well, the main difference is that you can now manipulate the outcome with a hairdryer. No. Jokes aside, the real difference here is two things. It is retail participant, building on what you were saying, and distribution. The depth and the reach of these markets is now something completely different than the niche that we have seen before, and they are much healthier. In order for people to actually use weather derivatives to hedge their businesses or stuff like that, you need them to be liquid, you need them to be actually available to buy, and you need people to be aware of it. It has been complicated, it has been difficult, it has been niche products, as you say. That is now completely changed. This is top of mind. If you walk around New York, it is impossible to not be exposed to commercials for this. Retail is there.
They are active, they love it. As a result of that, you get all the other benefits that JJ talked about. You get the market makers coming in, you get the institutionals, and you get the real use cases for hedging and for the future. That is what is different this time around.
Got you. When we were inside, we were talking a little bit about innovative markets that are out there. You mentioned Galaxy's impact markets, which I kind of think of as like a parlay or a combo for the financial marketplace. Give us your perspective on that and what sort of adoption has there been on that product?
Yeah. Now we are talking really early stuff, right? This is peeking around the corner, kind of what we are seeing in the day-to-day and what is next, and given the topic of the panel. We are seeing all these different variations of prediction markets, including stuff like opinion markets, where you can kind of take positions in different polls and opinions to what they call impact markets or conditional markets, which is really about betting on outcomes if something happens. Right? So what is the price of Bitcoin going to be if Trump gets elected? This is quite interesting from a financial perspective because if you are doing directional markets, and you are betting at the price of Bitcoin or at the price of Nvidia at the end of the year, these are already baked in. These assumptions around all these macro things that are happening.
With conditional markets, you are able to isolate some of these events much more, the impact ones, and you are able to price them differently. You are able to completely get a different risk exposure. But again, these are super early stage. The adoption is not there at all compared to what we are seeing here, but this is really kind of at least what we are seeing coming down the pipeline.
All right. We'll open this up to everyone here, and we'll start with you, Rob, on this question and go straight down the line. Clearly, the prediction markets, it's playing out through the state courts. We're seeing a circuit split. We're seeing different rulings in different states, clearly, or different districts, circuits rather, and clearly this is going to go to the Supreme Court. Let's assume that the states win, right? Sports predictions disappear, does that help the macro finance perspective or does it hurt the whole category overall?
Well, first I wouldn't assume that, but for the nature of the question, I will let that assumption happen. I also think, by the way, sports contracts aren't going away. Even if the Supreme Court says, "Hey, yes, the states should be able to go and regulate these markets," it will just mean that the states will go and ask for their pound of flesh.
Right.
Then you will have these markets being regulated by the states, and they will continue to exist. Outside of that, listen, the institutional side of what is happening in prediction markets is, from what I can tell right now, exploding. Now, it's exploding from a very small denominator. So when we look at, okay, well, people are doing billions and billions of dollars of weekly notional volume. Well, yes, it's relatively small today that we're seeing institutional equities-linked use cases or political use cases for the institutional hedging side or parametric insurance or go down the list of the things we're talking about.
But what we are seeing is we're seeing the interest and the learning and the continued integrations with the right APIs and the right brokers and the infrastructure build-out moving at a pace of a really early market that you should expect to be much, much bigger in the next year, the next two years, and the next 3 years. I mean, the fact that we've got all these people in this room, all these incredible people on this panel, kind of tells you that, right? No matter what happens at the Supreme Court, that trend is going to continue, and these markets are getting much, much bigger. The exchanges in this space are getting much, much bigger, and you can't really derail this train at this point.
I don't know what will happen at the Supreme Court. I've always joked that when it comes to sports betting, I'm a traditionalist. I like to do it with a gentleman whose last name ends in a vowel, has had his nose broken at least three times, and is loosely associated with the import/export business. But outside of that, the way we look at prediction markets, again, we're doing corporate KPIs because we see this as a natural extension of what people are doing already. And we actually see a lot of the crossover opportunity in terms of people adding to their portfolios. One of the great things about them is Peter Lynch, who we've all heard, great investor, trade what you know, and this is certainly what people know.
I think the other thing that is really interesting about this, many people ask me, "Alright, the options market has enjoyed this incredible growth over the last six years. Can it continue?" And 100%, I think it can continue. Why? Because if you look at what prediction markets are, they're nothing but giant probabilities. What are options? They're just probabilities. I think what you're going to start seeing is people using one product against another as they start to understand how to do this in a better way. I agree 100%. I think the opportunities are absolutely massive overall. And the other thing that we haven't discussed, where I think there's another amazing opportunity, is the data that's going to continue to come out of these. On that side, I'd say we don't know what we don't know.
As we start to see the corporate KPIs, is there going to be some sort of a correlated relationship between advertising on Alphabet and Home Depot lawn and garden? I have no idea, and I just made that up. That all said, there may be at some point. So the correlated data that comes out of these, I think, is also going to be some of the most interesting information we get.
Yeah. I completely agree with everything that's been said before me. I think that the ship has left port on institutional usage for prediction markets, regardless of what the outcome is at the state level for sports. I think one good indicator of that, or a lot of the feedback that we get around these contracts, one of the gating items, especially for longer-term use cases, is the capital inefficiency. If you look at what Kalshi filed with the CFTC several weeks ago for margin relief on specifically longer-dated non-sports contracts. I think those types of developments are already underway. What the response time from the CFTC is, I can't speak to. But I think you're starting to see evolutions in the market at a very, very rapid pace that create a more efficient product for institutional users.
And like I said, whether it's your traditional capital markets-focused institutional users, or I think one of the largest opportunity sets for these products is the insurance space, right? I think the realization of the flexibility and how these products can fit in for traditional institutional and corporate use cases is there. My view is, yes, while an adverse outcome at the Supreme Court level on how these are treated in the States, could that delay adoption? Potentially, and maybe there could be some concerns around reputational risk and the optics associated, but I think it's at best a delay rather than a complete brick wall for how this market evolves.
Yeah. I think I'll just echo all that and say that I think if sports goes away, obviously liquidity begets liquidity. And so having more attention and more people around the table just bleeds into everything else, so it's a positive spiral. If it goes away, the tooling is there now, or it's being built as we speak. The rails are being built and being used. The institutions are around the table. And so I think all the ingredients are there, even though sports was to go away. That said, I don't think that's going to happen.
This question's now for Rob and Einar in terms of offshore and on-chain venues can virtually list anything, right? Is the regulatory arbitrage a growth engine, or is it a liability for the category's legitimacy?
Yeah, so I can start. Listen, I think as a venture capitalist, if you look at the history of basically all technical innovation, most of it has started in a gray area, right? Look at Revolut, which is one of the most regulated businesses in the world right now, was essentially an illegal business for three years, right? That is just the way that technical innovation happens. And we've seen it with capital markets. We've seen it with prediction markets. What we're seeing today is that the U.S. regulators, and Chair Ceule very specifically, has been incredibly thoughtful about how we actually support pro-innovation markets here in the U.S., and how we potentially could allow for either on-chain venues or consolidated liquidity across international and local markets. And that is something that hasn't existed before. So, what does that mean?
That means that these exchanges can actually now build for where the puck is going in a regulated marketplace, in a place in the U.S., in a way that continues to make sure that they are within the bounds of what the CFTC has asked of them. That's really exciting. At the same time, I know this for a fact, all of the exchanges are going to the regulators across Europe and across Asia and saying, "How can we play ball in a way that makes you excited? Because we believe that this, maybe not an asset class or maybe products, are going to be tremendously useful to the market participants in your country. We want to do this in the right way. We want to make sure that we're listing the contracts that you think are appropriate.
We want to make sure that we're catching insider trading." This is what everybody is right now. Whether on-chain or off-chain, the difference is about transparency and resolution, the way you do the resolution, and potentially the way that we settle or the market or exchange infrastructure itself, but it's not going to change how people are approaching the regulatory environment in the future.
Yeah. The way I think about it is that the on-chain, offshore stuff, what's happening is really the experimentation part. Nothing starts in a regulated environment. All that wants to move regulated. What we're hearing, what we're seeing offshore is like, hey, if they could, they would offer it here. Right? Once it starts to be offered here, then it actually starts kind of going away offshore, right? Because people can now do it in a regulated environment. Overall, I think it's necessary, and it's just a natural kind of evolution of things. I think the other thing I also believe is people, especially regulators, have become very good at having a much more nuanced view and understanding of these categories than what they had 5 or 10 or 15 years ago. Right?
It's not just written off as like, hey, it's only scams and crooks over there, and this is all good. There's shades of gray. Overall, I think it's a positive for the industry and necessary for the industry.
Yeah. JJ, for you. All prediction markets today are fully paid for, right? It is a resounding complaint from all the market makers. We will use the example, Bears playing Green Bay, market maker sells the Bears, they sell Green Bay. Essentially, they are flat. But right now, they are posting all the money on, and obviously getting the credit on the sell side, but it is pretty capital intensive. Do you think this is going? Are you going to drive this to a portfolio margin? Because you are also using your own clearinghouse, which is interesting. You have Cboe Clear US, right? Which is a DCO right now and-
Not as of about an hour ago.
Oh, congratulations.
We just got approval to be on the securities side of the business.
What do you take? Tell us about it. Go ahead.
One thing I want to add to that, I actually agree. I want these offshore places to become exchanges, but operate under the same rules as exchanges, because right now they don't.
Love competition. Bring it on all day long.
But as long as we play the same game by the same rules, happy to do it. Anyway, that all said, I think it's. As I said, we just got the approval an hour ago. There was no. OCC, which is where traditionally options are cleared, and with things they have going on, was not going to be able to get to clearing these for a while. We decided, okay, we already have a clearing firm. To everyone's point up here, prediction markets are moving quickly. You can't sit and wait six, nine months or 12 months for something to happen. You have to go out and make it happen yourself if you can. Really, we've dealt with the SEC. They've been great about doing this in terms of the clearing side.
We got approval to use a futures firm, CCUS, has permission to clear securities trades. Will we go to margining? I think as we start to roll out product, we will do what is best for capital efficiency. I think regulators like when you crawl, walk, run a little bit. We want to get the product out the door, and then we will go into other things that help with that.
Got it. Matt, let us talk a little bit about the data, right? We have talked about, in some cases, is the data more valuable than the trading? We have seen examples of discrepancies or arbitrage opportunities, and it is actually interesting. In an earlier session, JB Mackenzie was talking about how they were seeing in the two Senate races in Texas and Michigan, which are toss-up states, I think on their platform, they were seeing it at 70/30. It is going to be interesting to see how that plays out. What is Cantor's perspective on this? What do you see?
Yeah, I mean, listen, I think the value in the data was identified by institutional users. I would argue a year, if not more, longer ago. You already have investors that are using the data as an input into their investment process. I think over time, right, a good example that I like to highlight is the June Fed meeting. Fed Funds Futures had the probability of a hike priced at, I want to say 36%, 37%, while Kalshi had it priced at 24%. The Fed did not end up hiking, right? That is a meaningful discrepancy in implied probabilities between two different markets and two different instruments. I think over time, right, yes, was the data accuracy of the prediction markets more accurate than kind of the traditional Fed Funds complex? Absolutely.
But over time, where I think you are going to see other opportunities and other market makers and institutional investors capitalize is also being able to monetize that arbitrage and that mispricing. It does not apply to the full suite of listed contracts.
The list of prediction contracts currently, but in certain areas where you've established markets without alternative outlets for that risk, it creates a relative value opportunity that I think is going to become increasingly more attractive to institutional users.
Let's start with you, Einar, and we'll go down the line, and everyone take a minute in terms of where are these new users. Are we going to have new users that are going to be using these products, or is it going to be embedded in existing platforms that are out there today that are going to start trading all these new products on this new frontier?
We're seeing a little bit of both. It's a bit of a cop-out answer, but I would say mostly these are new users from our perspective, right? Especially on the sports side, it's grabbing a lot of younger people, a lot of people, making it much more accessible and really immediate. So on the retail side, it's certainly new. On the institutional side, I think it's by definition all new users. So, you can take that as you want, but yeah.
Yeah, I agree. My view on the institutional side, it's a whole new subset of users that will find new and novel use cases on how these event contracts can fit into their portfolio. Again, I keep coming back to insurance, but I think the insurance market is a perfect example of where these contracts, the bespoke nature of how they can be constructed and put up on exchanges to perfectly tailor the specific risks that an insurer or reinsurer is trying to hedge. That's an untapped vertical or use case that I think is going to drive just one of many examples that will drive institutional adoption over the next. Again, I think the evolution is going to happen much faster than probably folks outside of this room think.
Go ahead, JJ.
Yeah, I would agree, it's primarily new users, both on retail and, as Einar said, by definition, it has to be institutional generally. What's really going to continue to happen out of this, though, is the new users of these products are going to be new users of many products, which is going to be good for the entire ecosystem overall, especially when you start to put. I love Matt's insurance example. As we start to get more product out there, people use them differently, and the brokerage firms will start to educate people how to use these as part of their portfolios, so you will have natural growth in a lot of different areas. You think about so many people who have homes in Florida for hurricanes or things like that.
Those just are such natural hedges for the individual, and once brokerage firms start educating people how to use hedges in their own life day to day, I think that's really going to be where you see these explode.
Yeah, I don't have a lot to add. It's absolutely a market expanding, both from a set of users, but also from the amount of things that people can trade. When you talk about having direct exposure to basically any event, it's very clear that that is going to, one, create just more capital in the system, but two, also create net new users. I think we're in this point in time where for the first time in a while, there is a very large piece of capital markets that is being essentially innovated and upended by new entrants. We're having innovation happening, and people are thinking about capital markets in a very different way than CME or Cboe, or all these other guys have been for a long period of time.
I know, JJ, you said you're excited about the competition, but you can see it happening.
Not upset about competition.
Excited about the competition.
Love competition.
Excited about the competition. Anytime you see net new entrants into markets as well, you see that be a market expanding as well. That's what we are really excited about, and I think this is going to be, like I said maybe three times on this panel, so much bigger in the next two years.
We only have a few seconds left here. I want to go quickly to Rob and to Einar. What are the new products that you are. Not new products, but sort of investment opportunities, right? There is a lot of exchanges that are out there, but what is new in terms of prop trading, settlement oracles?
Yeah, there is lots of exciting things. I think entrepreneurship is really about jumping off a cliff and building the plane as you fall, and that is kind of where this industry is right now. So really, it is about rails and tooling. So everything from indices and things around that, ways to offer leverage, to stuff that Edge is doing, for example, moving money in and out of different markets. That is at least where I am seeing a lot of activity personally, as a venture investor.
Yeah, I maybe have a little bit of a different view, which is that, yes, you are seeing net new oracles, yes, you are seeing net new tooling. When I think about my job, and my job is to get what I call a venture return, right? So the venture return means it can return my fund. I think these markets are going to be tremendously big. I think there are going to be net new exchanges that are going to be tremendously big, or already are at this point. I think we are going to see net new brokerages that are going to be tremendously big because of it. I think a lot of the tooling is probably small business, to be honest. But it is needed. We do need better oracles, and people should definitely be building there.
But it's probably not where I, as a venture capitalist, am spending or as excited about.
Last question, and I know we've run out of time a minute ago, but let's go quickly down the line. What percentage will non-sports be of the marketplace in 2029?
I'm going to go 60%.
60?
Yep.
I'll say 50.
I'll say 75.
I hope you're right.
66.
Well, thank you very much. You've been a wonderful audience. Please give our panel a hand. Well done, gents. All right, we're going to keep it going here. Now the question is, how do we get prediction markets so my mom understands them, your mom understands them, and they don't get scared away?
How do we get prediction markets beyond the early adopters? I want to welcome Chris Grove of Acies Investments, Jeremy Levine of Underdog, and Joe Anzures of Crypto.com. We're going to talk about that, getting the prediction markets mainstream so we can explain to our mothers what we're up to.
All right. Before we get started, if you could join me in offering a round of applause of thanks and gratitude to The NEXT Team and also the staff here at Convene. I think this has been an exceptionally well-run event. We only notice those folks when something doesn't go well, so I think it's appropriate to join me in a round of applause. All right, my name is Chris Grove. Joining me today to talk about the mainstreaming of prediction markets, we've got Jeremy Levine from Underdog. We've got Joe Anzures from Crypto.com.
Before we get into that, obviously, big things happening for both of your companies. Joe recently got the $400 million investment from Citadel.
Yeah.
That is a lot of millions. Even a lot of hundreds of millions. Jeremy, obviously, the acquisition with IG. The first question kind of plays off of that. We are asking this idea of how can prediction markets be brought into the mainstream when you see numbers like that from Citadel, when you see big acquisitions like that from IG? How much more mainstream is there, really? Is there an argument that we are already at that point of saturation? If not, how much more room is there to grow? How much more road is there to run? Joe, what is your thoughts on how much more there is to do?
Yeah, I still believe there is a lot to do. There is a lot that we can get to. Yeah, sure, I think we have hit a good chunk of the market today, retail market. I think that a big question is when the institutions come in, and I think that is where we will see the market really grow at a faster clip. Again, getting the markets, it really comes with the trust and then the real clarity on regulation and where that sits, some of the capital constraints. That being said, I think that is probably the next phase that we will see to really take the market even further where it is today. I think we are in a really good spot from the retail side of the business.
Jeremy, you agree or?
It's pretty damn mainstream. If you think about the companies in the space.
Right
The amount of money being spent right now, look at NFL season and the amount of money being spent on marketing. There's been well over $1 billion spent on marketing prediction markets already this NFL season. That's damn mainstream. I don't think there's many marketers out there spending, or many categories out there spending more money. I think prediction markets is such a big, broad thing. There's, hey, the core sports experience, there's innovative sports experiences, there's games, there's institutional, there's non-sports, there's politics, there's finance, there's probably more stuff that's going to be created. There's so many places it can go, but is it mainstream? I think it's super mainstream.
Joe, with the opportunities you see for growth, walk us through who are the customers that are not currently utilizing prediction markets today? That could be consumers, that can be institutional. What does that customer look like or what do those customer cohorts look like?
Yeah, again, I go back to the institutional side, and I go a lot from this from our experience, too, from starting a large crypto exchange 10 years ago now. We really focused a bunch on the retail side, and we've seen that we really attacked the retail market, and did really well in the retail market globally. We're growing that side of the market. But it was really when we hit the institutional side and became, I would say, the trusted exchange. Everyone probably remembers a lot happened in the crypto space, and there was a lot of players in there and a lot of, maybe I would say not so above-the-line players.
Once that kind of exited the system, then you had your trusted players, and they are the ones that built the strong compliance programs, had the deep liquidity, and the funding rails, the trusted funding rails, that you always had access to your funds 24/7. We never went down. As the institutions started coming in, again, that is where I focus on a lot, just seeing the previous transition that we have seen from the crypto side, and I think that is the biggest segment that I see missing for me. From the retail side, I think we are, to Jeremy's point, we are pretty mainstream on that for sure.
On the institutional side, what are the missing pieces of the puzzle there? Does it simply get solved with time, or do you see big things? I think the parallels to crypto are really instructive. What are some of the other things besides, obviously, trust, the biggest players? Are there pieces missing or does this just get solved as more time passes?
Time. I think time is a big component of it because it is on the regulatory front. They are kind of jumping in, but I think before they really jump in, it is going to have to be really clear from the regulatory front, and the oversight pieces that need to happen. Again, I think there is also a component when it comes towards the capital. There are some inefficiencies in capital, of course, because obviously being dollar collateralized and obviously in the pure derivatives markets, we have margin and what have you right now. So, I think there are definitely some of those, but some of those are being addressed today by some companies in the space.
I know that is one of the things that we hear a lot from some of the different market makers and the institutions in the place, is really locking the capital up for extended periods of time.
Jeremy, so same question, same theme, but on the retail side. What is it that is missing from the current experience, the current marketing, the current branding? What is it that is going to pull the customers who are not currently participating in more of an entertainment experience into the prediction market ecosystem?
I am not sure I want to give you what are the real answers in my head, because those are things we are going to go do.
Start with the fake answers.
Hopefully they work.
And then we will pull the real ones out of you slowly. That is what the lights are for.
Look, it is always about innovation, right? You go back to just what is the customer doing, what is the behavior they are doing, and then how do you build more for them? Prediction markets, the behavior is not nascent. The behavior has been around for decades. There have been versions of exchanges. There has obviously been sports betting. There has been a lot of similar things. There are new things that are possible. There is something we are launching now-ish. I think it just hit my phone in an admin account probably yesterday. That is a new market type.
Oh.
For sports fans, that is going to be a really fun way for them to express opinions. I am excited for that because it is, I think our first truly new market type. I am curious to see how people react. I am curious to see if we can communicate it in a really clean way to customers where they understand the benefits of it and understand why they would use it, and we can create a new way for them to express an opinion. Look, Kalshi was just out a year ahead of everyone, in a way. Now, there is a handful who have figured it out and are here, and you all can see them in the rankings. We are talking on one hand of companies that have figured it out at this point. Now it is like, okay, the core experience is there for all of them. What comes next?
What can you build that adds to that experience, that creates a new way for the customers to enjoy it? I think that is what we will start to see more of over the next year or two, where, again, for us even, the last years have been like scramble to get to where we got for-
Right
The start of NFL season. Now we've got the thing. Now it's okay, what can we build and innovate? A lot more of our energy can go to innovation rather than building the core necessary platform.
Jacob from Novig talked yesterday about how he was interested in some frontier player prop markets. Hearing you talk about new kinds of markets. If we fast-forward a year from now, two years from now, how do you think about the volume contribution of what we'll call novel markets, things that are not part of the base product today? Is that something that is a 5% contribution, or do you see it being far more substantial in terms of the total makeup of volume on the exchange, or just new customer acquisition, new customer engagement? What kind of role do the new and novel products have to play going forward?
It's so hard to speculate because it takes innovation and it takes getting that to click and it takes all the hard work that comes with that and making sure the customer understands the benefit of it. It's always really hard to speculate on that. What we've seen, and we all see how much sports drives prediction markets. One of the beautiful things about sports is they're happening every day and they're turning over fast. If you just had one sports game every year, like you have one election every four years, the turnover just can be way slower. It doesn't resolve and pay everyone out and get to go again. You have sports go every day. If you can build into that behavior, obviously, the numbers can get way larger.
My passion and background in trying to build a sports stock market, and that's something where I'm like, "Okay, that could be possible." There's collateral challenges. There's tying up money for a long period of time challenges. Is there a way to do that where you can make a short duration and thought? We're still exploring. We haven't found the answer yet or else we'd be doing it. I think there's the chance there's some breakthrough innovations, but it's tough to prognosticate.
Yeah. Joe, are you seeing demand for new novel markets on the OG side, on the direct-to-consumer side? Is that something customers are making clear to you that there's an appetite for, or does this seem more like a nice-to-have versus a need-to-have and the bread and butter is really going to be what it's been for sports in the U.S. for the last 10, 15 years at this point?
I would say, yeah. We are seeing a little bit of demand, not only just from the OG side, but also from a lot of partners that we support today, too, as well. We do lots of discussions with Jeremy and obviously in FanDuel, DraftKings, and what have you, and having those conversations. So we see a lot of demand. I'd still firmly believe right now, though, it seems like that it's really a lot in the core markets and
Yep
The core sports market today.
You guys are in an interesting position because as you mentioned, you've got the direct to consumer with OG, and then you also serve as the B2B background, the exchange platform for a number of partners. Some of them are on the sports side, and they're more centric. Some of them, like Robinhood, are on more of the financial side. Are you hearing different things from those different groups about the kinds of markets that are of interest to their customers, whether it's in terms of velocity or duration or just the nature of the markets themselves? Or is it all converging across a pretty similar set of core markets that are driving the interest?
Core markets, but I think probably a big one has been the crypto markets. That one has been a pretty strong one for us, getting that down to the 15, 5-minute crypto markets have been, I would say, probably the biggest one that we have seen from a demand perspective. We have had some other companies also, too, come up to us and you probably saw the one with payments. They are launching the whole AI platform as well or that market, I should say. They are doing a lot of stuff behind that and provided the data source components there. I think that one will be an interesting one to see where that one goes. But I think the biggest one that we are probably seeing today is really in the crypto markets, and we have seen a lot of activity there.
Before letting that thread go, just one more question. The shorter duration, higher velocity, is that a component, a theme that you expect to drive more mainstream adoption? Or is that something that is limited more to a very small and very vocal and valuable cohort, but does not necessarily appeal to the mainstream customer?
I think it appeals to the mainstream customer. Now we are seeing a lot more of that short duration component for sure. I think that is going to be another big driver of the markets going forward because we have seen that one. That is probably got one of the largest growth markets that we have seen on our platform.
Jeremy, any analog on the sports side in terms of, I do not know if we would call it explosive growth, but definitely measurable and meaningful as we keep seeing the durations drop down for up-down cryptocurrency markets. Clear analogs on the sports side or not as easy to say, yeah, the A to B here is that market to this market on sports?
Look, to my point earlier, sports are naturally short duration, and we're always seeing the predominance of customer entries always coming within the couple of hours before the game into the game. I think the first killer feature for prediction markets, where prediction markets right out of the gate were better than sports betting for customers are live. Live in-game trading, and you see that behavior and even more of the volume in prediction markets versus on the old school formats, sports betting or anything before is happening live and in-game. Obviously that's short duration that's resolving sometimes 5 minutes later when the game ends.
Are there any of the, you mentioned some of the institutional barriers, or the friction I guess, which is an interesting one to dig into. Discuss some of the things that are maybe inhibiting the rollout of additional short duration markets or novel markets that specifically are focused on market makers or other liquidity providers. What are some of the big unlocks there that if we see these problems start to get solved, we'll also see a corresponding unlock in either the number of these kinds of markets or the liquidity that's available in these markets?
Yeah, I would say, probably on the combo side is where we've seen a lot of, probably, some of the, from the constraint side. I think, from a capital perspective and things, if we had that liquidity option to some of these market makers, I think we would see a lot more liquidity coming through on the combo side because I think that's been a bit of a challenge to date.
What's the mismatch there? Because as a story, as a narrative, it sounds appealing. You would think money would just sort of be rushing in. So what's the difference?
I think there is some hesitation, too, because I have been in some of those conversations. I think there is some hesitation on the institutions that want to provide potentially that capital to the market makers on where the certainty is around regulation going forward. I think that has been a little bit of a hesitation from some of those guys. But again, there are a few that have jumped in. Not enough yet, but I think that is, again, that is what it comes back to, again, is more on the certainty from the regulation side. If they are also going to be putting their capital into this market and then they are going to want certain returns back, of course, and how is that valued today? Whether it is just I am just lending you capital or if I want to maybe potentially do a JV with you, right? Then how do I value that?
I think that is where I have seen is where they have kind of been getting stuck.
Are you seeing that attitude change at all with the Citadel announcement? Obviously hard to get more blue chip than that. That is an incredible amount of validation. It was specifically focused on, at least a component of the investment was focused on the prediction market product.
Right.
Did that change sentiment or have you not really seen any sort of alteration of the prevailing attitude after that?
I wouldn't say I saw anything that was a drastic change.
Okay.
Yeah, not yet. But I think we've seen a little bit, but nothing that's really been a drastic change.
I'll flip back to the consumer. Jeremy, we talk a lot, I think Joe used the word trust earlier. When we're talking about trust and credibility, I think these are words that are thrown around pretty often. Talk about building a brand, winning consumer affinity, getting them to try a new product and stick to that product. Taking it from the abstract to the concrete, what do you think of as the top two, maybe three components of the product experience that most engender that trust and credibility? What have you got to get right from the product experience perspective to win that trust and to develop that credibility with consumers?
I'm going to give you an answer and a hot take.
Sure.
I had this, I think a couple of years ago, it kind of hit me that a lot of the early challenges from kind of internet 10 years ago, building tech businesses 10 years ago, that I was trying to solve 10, 15, 20 years ago, at my companies then, aren't really hard challenges anymore. I think trust is actually one of those. 15 years ago, if you were building an internet company, like trust, are they going to see my money? Am I going to get my money? Is the money going to be there? That was the number one thing you had to solve for your customers. Today, customers trust. They're used to being on the internet. They're used to putting their credit card on the internet. They've got Apple Pay. They've got the ways to do it.
It doesn't feel to be the thing you have to get over as like a nascent startup.
Interesting.
Now, I think there is trust and the other, this isn't the question you asked, but the other thing that I just think is interesting too is you think of the other challenges of scaling tech companies. Technical scale also was a huge challenge. Think 15 years ago, we had the fail whale and Twitter going down every day, and sites don't go down because of load anymore, or are very unusual. Where 15 years ago, that was happening all the time as companies would get scale. Those challenges have gone away. As I think about probably what you're asking about in trust, it's can the customer do the thing they want to do? Is there a friction? Are you able to remove the friction? Are there times where they want to go do something and that thing's not there?
In prediction markets specifically, and part of it is the collateral challenge that Joe was talking about. Customer wants to create a parlay and they go do it, especially in-game because it's even harder in-game, and they do this thing. They spend however much time coming up with the opinion they want to express, doing the work to express that opinion, and then they can't get a quote. That breaks down the experience. That's a waste of their time, and I actually think wasting their time is probably one of the fastest ways to destroy trust.
Interesting. Trust being repositioned less as the classic idea of, am I safe here? Is my money safe here? My data safe here? And more about is there an alignment between the expectations I have about what I can do here and what I am actually allowed to do.
When I go to do the thing,
Yeah, that it works.
Does it do the thing? In the early days of prediction markets, especially across a lot of the different people as they were building into what they needed to be, can the customer do the thing they want to do? The more you can match that expectation, the better your product is going to be, and that is the thing we focus on relentlessly is making sure the customer can do the thing they want to do without friction.
Joe, you co-sign there, or are you still seeing that some of the things like payment reliability, quick cash out, the more traditional hallmarks of, again, that bifurcation of am I safe here, is this a legitimate place, versus can I do the things I want to do here? Do you agree with Jeremy, or are you still seeing consumers want to have that feeling of being safe and then worrying more about alignment of the experience they want to have?
Yeah. No, I agree with what Jeremy said, too, because, again, it's good point there on the wasting your time. You get through that whole process, and you don't get the quote, right? That's not a great experience, and that does cut into the trust of the platform. I would say, but we still put a lot of emphasis on the payment rails and the access, because I think at the end of the day, that's another big component for trust. It's the funds, and if I'm locking my funds up with the platform or sending my funds to a platform, I want to make sure I have 24/7 access to them, and I want to make sure it's easy and simple. Plus, they're not charging me crazy fees or top-up wallets or even get withdrawals or whatever it may be.
I think that is a key component. It's one thing how we've always built on the crypto exchange, too, and back in even PayPal days and what have you. It's really about making sure the customer does have the least amount of friction and always knows that they can get their funds whenever they want. I think that goes a long way on the platforms, and that's the way that we built out Crypto.com from day one as well, too, and-
If you think about PayPal, that was the first to be like, "Hey, you can trust moving money online here.
Right. Yes.
That was their whole premise, and now that's such a commodity.
Right. Yeah.
It's not hard to move money online anymore. PayPal literally invented the way to move money online back in the day.
Yeah. No, and when we added it to the Crypto.com platform, there is that perception. Everyone feels like they have extra added security with PayPal, which you don't. But it's there, and again, but it also comes back in how you service the customer. So if there is an issue, PayPal is going to take care of you from that perspective, because it is a brand and reputation thing, and it's the same thing that we would do on the platform as well. If there was potentially ever an issue, it is how we service the customer, and we would take care of our customers at the end of the day, which does, again, develop that continual trust and loyalty to the platform.
All right. We got about 10 minutes left. I got a couple more questions, but why don't we just bail on all those and talk about PacRips? What do you guys think?
Prediction market conference, though.
Yeah. I think there's a secret PacRips conference happening here in the shell, the hermit crab climbing into the shell of a prediction market conference. Sponsorships, I think, are often used as a way to try and build either familiarity or some version of trust, some version of comfort. Joe, I think you've seen it all, done it all when it comes to sponsorship. Jeremy, you mentioned the amount of marketing dollars that are going out the door this season. A lot of that is in the form of either partnership or sponsorship. So Joe, without necessarily picking on you, curious, as someone who spent a lot of time and high profile, UFC, F1, what are some of the missteps that you see? Where's the inefficiency, and where are some prediction markets likely to look back on the spend and go, "That was a misallocation.
That was not the best use of funds.
Everywhere.
Yeah. I would say, obviously you guys, you probably know what we did. There was a lot that the crypto markets were doing back in the day. We still have some big ones with rename the Staples Center and Chris mentioned F1, UFC, a lot of different things. Big dollar items, UEFA, whatever. But they were really good. They were great, I would say, getting our brand out there, and especially for where prediction market is right now and where crypto was back then. When I joined Crypto.com and overseeing the Americas region and starting the U.S. and Canada here, we were about 10, 12 million or so users and quickly went to 100 million within two years, and we are over 175 million now globally. It was wild.
I still remember when we signed the Staples Center deal, renamed it to Crypto.com, and people were signing up for our prepaid card, and they were nice. There was lots of rewards on them, metal cards. We probably, at the time, were about maybe a two-month backlog because metal cards take a while to produce. After we signed that, within two weeks, I had almost a nine-month backlog because everyone was coming to the platform quickly, but they are expensive, too. It is a big number to get your name out there, but it did work well. We were giving out crazy rewards, but some of the missteps was, is probably thinking about that and thinking about it is great during the bull runs, but then what does that look like when there is a bear market?
You still got to activate those things, and you still got to commit to activate those things to feel like you are really getting an ROI out of those investments. I think that is some of the things that you will see as they might look at as missteps going forward.
Jeremy, you saw version one of this post-PASPA with OSB, similar sort of opening of the money spigot. Looking back, I think mixed feelings about the efficacy of that spend. Prediction market operators learning anything from that or just repeating the mistakes of the past all over again, maybe slightly different denominations and slightly different focus?
Multiplying it, I think.
All right.
The amount of money being splashed around right now, it is going to work for one or two. It is not going to work for all.
So then, because you've done some sponsorships, partnerships, you're not anti, right? Same question for you, Joe. Without giving away the playbook, if we see all of this money being spent at the headline level with the most prominent partners, what are some of the areas, what are some of the themes where there is a greater opportunity for efficiency? Again, not naming names, but where do you look for efficiency? Where do you look for value? Do those pockets still exist because of the tendency of some of the larger brands to only want to go after the largest sponsorships, the top tier?
Yeah. Look, it's kind of the same as everything. You've got to build. You've got to have good ideas and you've got to build. Look, a philosophy of ours always has been, we don't want to just compete on dollars. I hope we can do something better than dollars. If it's just competing on dollars, anyone can do that. It's not hard to know, "Hey, you can go spend money on this. We can all go compete for the New York Knicks deal." We all know that's going to be expensive. We all know what it is. But is there something we can build, we can be creative, we can come up with a specific concept that makes it work and makes it all work for us and our customer and our offering, and then build together? There's the concept of kind of zig while others are zagging.
One of the things we're spending a lot of our time on, we've built a team called The Rascals, and their mandate is just to have fun. Have fun with sports, have fun with sports moments, try to make sports more fun for our customers, try to bring a smile to our customer's face. We've done that to varying degrees of success with different stunts. We did something with Lane Kiffin or making fun of Lane Kiffin when Ole Miss and LSU played. That had a lot of interest and attention.
We did it with SGA during the finals where we built a If you guys remember Operation, if you played Operation as a kid, we took the Operation game and called it Unethical Hoops and made it with SGA and got a rival of his, Dillon Brooks, to kind of release the game, and that obviously went super viral, got a lot of attention, I think made a lot of people smile. So I think doing things like that and just trying to bring more fun to sports fans is a lot of what we focus on.
Bringing more of the personality of the brand, the identity of the brand, pulling that through either the partnership or the attachment to a name as opposed to a formal partnership in that case. Joe, what would you advise, the benefit of your experience, current environment, what are the untapped pockets of opportunity that can maybe deliver outsized returns from a partnership or sponsorship opportunity? Yeah, I think it is probably taking a look at the company and your identity and who is currently, I would say, probably coming towards your platform. One of the things there, and just like in any business, there is a certain segment, right? It is like the 80/20 rule, and plus probably in sports books and even on the crypto side, it is a lot less than that.
It is probably maybe the top 5% to 8% that are actually driving the majority of the revenue on the platform. So it is looking at maybe even that. Do we hone in on certain segments versus, to your point, am I going to go out and spend dollar for dollar against these guys like we did early on in the crypto space? No. But now we are looking at those types of opportunities and how we dig into those specific segments and what is important to them. I think it goes back to now a lot of today is on how we service the client and how are they getting unique access and what can I provide there.
Because I think for us, we are probably, again, we have been around for 10 years on the crypto side, so we do have a lot of experience and a lot of those customers that we are catering to now are leveraging our platform today. That is where we are really probably spending a lot more of our time is, again, is focused on some of those interests for them, which are a little different than what we have done in the past.
Last question, same one for both of you. I will start with you, Joe. Make the case for why OG is likely to be a leader in expanding the appeal of prediction markets beyond the current audience that is being reached today. Why are you going to be able to drive this uniquely into the mainstream?
Well, I would say how we're going to do that, why we're the ones is, again, because I think we're one of the most trusted ones in the space. A lot of our team has been just within the crypto space. From the compliance to the legal side, we've been doing this for 20+ years. I'd say we have the expertise. We're the ones that are most ingrained throughout the industry, have the strongest relationships, and have the liquidity, have all the funding rails. We have the reliability, and I think we've proven ourselves in how we grew up in the crypto space is why we're still here in the number two or three market, depending on the day, on who's trading what. Globally for crypto, I think, and we've done the same thing and taken the same approach in the prediction space.
Jeremy, how does Underdog leapfrog the early adopters here?
Can I hijack? I'll answer your question in a second.
Yeah.
Just one thought that's been in my mind that we haven't quite got to that I think is interesting is if you look at prediction markets and what's happened today, obviously, the first two companies out of the gate, Kalshi and Polymarket, were startups. Right now it's all incumbents. I think something in terms of as it goes mainstream and where innovation's going to come from, the startups are coming. There's startups that have all started within the last year or so. I know you're super passionate about it. I see Noah there who works with us at Underdog, but also is at UFO Holdings and has invested in a lot of them.
It's going to be really interesting to see the innovation that comes from startups in this space, and I think that's where a lot of what is next is likely to come from, as is the case in any fast-growing space. But it hasn't happened yet, and I think that's a super interesting thing. They're coming. They haven't happened yet. What does that bring to the market? As for Underdog, for us, it's always really simple. It's just how do we build the best experience for our customers. We feel like we know our customer, we know the sports fan in America better than anyone else, and how do we build a great experience for them and the best experience for them is always the focus for us. Luckily for us, there's a-
There's many millions, many tens of millions, probably hundreds of millions of them out there that want to engage with the products and have fun with sports, and that's our mission.
Millions now, tens or hundreds of millions to go. Join me in thanking for their time Jeremy Levine from Underdog, Joe Anzures from Crypto.com.
All right, we got a powerhouse closing panel here, moderated by my friend Charles Payne. It's the wisdom of crowds. One of my favorite expressions in politics came from an old politician named Mo Udall. He lost a primary and he said, "The voters have spoken, the bastars" But there is wisdom in crowds, and one thing about the prediction markets we're going to find out is just how much wisdom is there. For instance, one study said prediction markets have beaten Wall Street analysts at forecasting earnings 78.5% to 43.7%. Are the crowds just that much smarter? With that, we bring back my friend Mr. Payne and his panel to discuss that. Charles, stage is yours.
Good afternoon, everyone. Okay, we got it right?
Right on.
Yes.
It's hard to see.
We've only got 30 minutes, so everyone here is accomplished. They have their titles, their primary. By the way, they all have about 4 million different titles and jobs and accomplishments. So the wisdom of the crowds, beating Wall Street, beating weather forecasts, beating everyone. You know, it's interesting. My background is the stock market. One of the first books I read was "The Madness of Crowds." Popular delusions and the madness of crowds, and for Wall Street, the crowds have always been wrong. Tomorrow, today, I didn't get a chance to see it, but the AAII number for individual investors is out, and if it's too bullish, guess what? Pros go short and vice versa. But maybe collectively it's not. We know that things have changed recently, and I'll just start with you, David. Extraordinary popular delusions, the madness of crowds, the tulip bubble.
People don't know, the experts know. More recently, though, the tide has turned and people are seeing the wisdom in crowds and where do we go from here?
Well, first of all, thank you all for being here and this is exciting to talk about. Wisdom of the crowds obviously is a pretty old concept and we're an old group that have been playing around in prediction markets for decades.
Decades
Some longer than others. Impressive resumes on this. This ties back to academic papers over 100-plus years old. Prediction markets on the street corners in the early 1900s predicted the presidency with great accuracy. Move forward and you see a lot of different versions of this type of work. To see the modern versions of it take a lot of traction is, of course, very exciting. It is a great way to incentivize people to come in, to utilize their dispersed information, their idiosyncratic information, their ability to pull in different models, and ultimately bringing new data and models to the business makes more accuracy. The flip side of it, this modern set of prediction markets is very heavily driven towards volume, and being driven towards volume is not necessarily optimized towards accuracy.
There's a bit of a trade-off and a lot of questions, which I know Emile and others will have about is it a great thing for wisdom of the crowds to have so many people coming in, which may be driving some dumb money that brings in a lot of smart money, but also may anchor prices on some sides, may lead to manipulation concerns, which drives people off into different directions, and ultimately, you could have smaller prediction markets, even play money that can do very effective if it's set up right to optimize over information versus money flow.
Robin, you're nodding your head vigorously over there.
Prediction markets are known to be accurate by comparison with other especially non-market mechanisms. A large literature that is robust. Compare it to Wall Street, that is another market. I am not sure the prediction market is going to beat Wall Street market to market, but when you do not have a market on something where you just have a question, that is where the prediction market wins. It is just better than not having another organized forecast. It is just more robust. A key question for this panel, though, is how much does this world care about accuracy? Accuracy is the measure that we are arguing over here, and yes, there are lots of ways to make these markets even more accurate that we are not using, but do they really care?
Well, Emile, I will ask you because it feels like accuracy is what sort of has given it the gravitas, and it is what is drawing in the big money, and it is taking it sort of to the next level. People who have been around it for a while have talked about its wisdom for a while. The first person told me about it was a reporter named John Stossel.
Yep.
This is maybe 25 years ago, and it was like, "These prediction markets are better than surveys. They can tell you who is going to win presidency." Everyone just kind of like, "Okay, thanks, John. Did he take his meds today?" But the accuracy has built the credibility that is making this the explosive movement that it is.
Absolutely. The accuracy is the key. 100 years ago, there was a great Frenchman also called Emile, by the way, which said, "To govern is to predict." By that he meant that unless decision-makers care about accurate probabilities on which to base their decisions, they are just going to run us into the walls. We can see that happening very often, whether it be government, companies, or even our own individual private selves making bad decisions for lack of accurate predictive anticipation. What prediction markets can do, besides making loads of money for lots of people and taking a lot of money from the pockets of most people-
Right
is also provide excellent, best-in-class probabilities on which to base decisions to actually run the world better. Lord knows it needs to be run better these days.
I'm going to read a brief description of a book. The premise of a book. "Defense doesn't win championships, teamwork isn't what it's cracked up to be, momentum is a myth, and the Chicago Cubs aren't cursed, they just stink." Toby, that was part of your book "Scorecasting," so you've been pushing against the grain, against popular beliefs out there, and of course, we know sports has been the epicenter for this predictions market. How do you tie it all together?
Yeah, it is related to something Robin was just saying and others alluded to, which is we want these markets to be accurate, but part of that is there are distortions in markets, right? There's behavioral biases people believe. They misconstrue signal with luck or skill with luck, if you want. They focus on outcomes and not recognizing it's a function of both. Because of that, prices are always going to deviate a little bit from the perfect price. This is sort of an age-old market efficiency question. But they go hand in hand because you need incentives to make markets really accurate, which I think is critical for decision-making. I absolutely agree with Emile. But you need the incentives of other market participants to make those markets accurate.
There's a famous paper by Grossman and Stiglitz, a Nobel laureate, which is you need enough inaccuracies in the market to warrant people to come in and make them more accurate and more liquid. I think they go hand in hand, and we're going to see this. I think it's a misconception to say there are people going to be losing money. There always are in markets. There are people making money.
Right.
You need to have that.
Although, is it extraordinarily skewed right now, though? Particularly on the sporting side? It is just a tiny fraction according to what I have read, and it is so skewed that it feels like maybe that could be a barrier to general public trust.
Compared to what? Do you think it is less skewed than the stock market? It is the same.
Yeah.
Well, at least today, if you own some CoreWeave, you can just sit on it. But if you miss the game, if the Jets didn't win last week, that is over, and it is over and done. I think that might be a distinction.
Well, I think there is two distinctions. One is if you look actually at the mutual fund industry, the hedge fund industry, the stock market in general.
Right
I think the statistics are the same in terms of who is actually winning consistently and who is not. I think a broader point is there a risk premium, something that you could just buy and hold, which you can in the stock market? There are some things in prediction markets that are like that, and there are definitely some things that are not, and I think we could have that discussion.
David, so I want to go back to Robin. You mentioned, Robin, something about you could put up fake money or play money. Is the money aspect of it, or at least the general public has been led to believe that one of the reasons that this is so much more effective than, let us say, a survey where someone is waiting outside of a supermarket to ask you, "Consumer sentiment, how are you feeling?" "Not too good." "Okay, all-time low." But because we are actually putting our hard-earned money to work, does that lend to being taken more seriously by the participants?
I think that one of the most exciting things we have seen in research over the last few years in this space has been to explore the value that people can create if they are properly motivated, whether or not it is money or whether or not it is just fame and interest, or whether or not it is simply ego. And there has been some really awesome work from Emile and others, especially Phil Tetlock and the Good Judgment Project, and others that have said, "Let us take a smaller group of people, let us incentivize them, let us make them into a domain they may not have that much information of, but really have them research very well." And I think something Robin said very nicely earlier, it is just about putting a proper mechanism in place.
Maybe that's prediction markets, maybe it's other scoring rules, but it is very impressive how the wisdom of the crowds can work in those situations. Look, it is super exciting to have billions of dollars which drive all sorts of people coming in. The question comes in, the dumb money does drive in a lot of big money, so people invest very heavily, and they can find new and exciting details and data that could really get a little alpha at the end. These are really core trade-off questions because they also incentivize all sorts of other things that make us worried. This question about how we're incentivizing people to actually create new information to bring something new to the table is, I think, a critical trade-off question. I'll add one more thing. You guys were talking about sports versus politics.
Look, different domains are also going to have very different potential marginal values from a retail versus a main investor. In sports, it's one of those things where the data is so centralized and so readily available, it is very hard for a retail person to just have something new and meaningful. So they are going to lose because the data is there for model makers who have a lot of processing power to just get it right. In politics, maybe there's a little more things you may know.
Right.
In questions about economics, in questions about the market, in questions about dispersed companies around the world, people may be bringing legitimate information into the market structure, and we should be thinking about how these different domains really behave differently when it comes to information finding.
Well, I know I've read where in a recent study, the prediction markets call the weather beat all the weather forecasts like six out of seven years. So I guess the bookies are better at their job than the weather forecasters are at theirs, because it's so much harder to get that Edge, it sounds like you're saying, in sports. But Emil, something I read here, you say, how the retail prediction markets can beat the experts, and you say there's a phenomenon and it's actually a math equation. Accuracy equals expertise plus diversity.
Right.
Talk to us about that.
Yeah. That's the Scott Page's diversity theorem. Basically, the idea is that in terms of prediction markets, the accuracy of a prediction is going to be determined by two things, and it's going to be determined equally by these two things. It's not like one is more important than the other. One is the average knowledge or expertise of the participants. The other, and obviously when there's more expertise individually in the panel, then the accuracy of the collective prediction is going to be better. That's totally obvious and not very interesting. But the interesting thing that's very counterintuitive is that the more diversity there is in the opinions, in the predictions from everyone, the more they differ, which means the more some people are totally wrong, because not everybody can be right if everybody says something different.
The more the collective aggregation of these different judgments is going to be accurate. The more people disagree, the more likely it is that the collective is going to be accurate, given the same amount of expertise.
Right
That's distributed. There's loads of beautiful data in practice that actually shows that. Which goes to something that David said earlier as well, which is that when you attract a lot of new money, of maybe less informed money into the market, then you're driving the diversity, which is good, but you're also reducing the average level of expertise, which is bad. So, depending on how that equation or that compensation goes.
Right
Basically, diversity can replace expertise can replace diversity, and they are both cumulative and interchangeable.
Toby, I moderated the opening panel and we talked about how this is going to grow. Estimates are to $10 trillion, and sports will only be maybe 30% of it, so there are going to be a lot of other opportunities out there. How does this apply? Let's say, for instance, I would assume something is up already on the next White House press secretary. So, I would imagine that would attract a large enough pool of bettors. But some of these others, these niches, seem to be so small that, A, do they have the same ability to be accurate? Also, from a business model point of view, would it be enough to drive this business the way people think it's going to grow?
I think that's a good question. Look, to a first approximation, the more liquidity or the more trading you have and the more volume, to Emil's equation, the more accurate the prices will be. I think what you found on these markets, though, which is interesting, is even very illiquid markets have been surprisingly accurate. It maybe goes back to some of the things David was saying, that even when you don't have money at stake, they can still be pretty accurate. So I think from that perspective, I think the data sort of suggests it's been very accurate. Now, in terms of the business sense, those small sort of contracts, that's not going to really grow this business. I think what really grows this business is if you could put contracts up there that we have a saying in economics, which is completing markets.
Robin alluded to this. Are there things out there that we can't trade easily on Wall Street, right? Things like elections, for instance, that people really do care about. They matter for society, they matter for the economy, yet there's not a direct hedge or a direct bet that you can make there. Coming back to the diversity point, we've been talking about information diversity. There's a risk-sharing diversity that matters a lot for markets as well, right? I might care about one outcome, you might care about another. Maybe we're equally informed, but there's gains from trade from that, and you want markets to function that way, too. So if the business is going to grow, I think it's got to be in that direction, and we'll see if that happens. I think it's starting to.
Charles, I would like to disagree with a claim that Toby made previously. We do have this idea that the way we do markets today, you do not have winners unless you have losers. You need a lot of losers to fund the winners, and the more all of them, the more accurate the prices are. That is because we are not paying for the information. Today, and almost all through history, these markets all exist because of the traders, and the information is a side effect. If you were willing to pay to make these markets more accurate, you could subsidize market makers on them, and then just the informed traders could trade on them, and the other losers do not have to be there.
In fact, the Good Judgment people, the people who did Superforecasters, have shown that if you take a market of everybody and you find the people who do best, and you just make a market for the people who do best, that market is more accurate. We could be doing that all through financial markets in our entire financial world. We could have more accurate prices if we do that, but then we would have to pay, you see, to subsidize those more expensive markets.
That is a loser. Somebody is paying. You are acting like nobody pays.
Right. The point is-
Somebody pays.
If information's valuable in our world, then people who get the information should be willing to pay for that information. At the moment, we've been taxing losers to get our information.
They volunteered, so hey, fair game. Still.
You could tax somebody else. We are talking about the same things. There is a question of redistributing who bears the losses, and that is a debate we could have.
Well, it is not a loss because the people who pay for the information, they actually are getting the information, so it is a complete win-win, right? There are no losers in the equation.
Look
Well, I read this great piece over the weekend, and it was some guy, I think he was a truck driver or something, maybe I am being a little facetious, but he was not a Wall Street expert, and he made a killing on Kalshi with the CPI number. So the idea that this democratizes even Wall Street investing and we keep hearing about how AI and super intelligence are going to displace so many jobs, and all the mathematicians in the world are upset right now at this very moment. How far do you think this could go as far as being a disruptive force out there? I will start with you on this one, David.
I think that tying together a lot of what you all have just been talking about, if you want to grow the business, there are probably some types of markets that could be informative that are going to be detrimental to overall growth, such as this question about who the next press secretary is going to be. Because basically, the reason prediction markets do well and things like that is because they induce inside information. That is the only way to win that market. You can get lucky, but the idea is that you are inducing people to provide inside information. So if we think collectively it is worth knowing that, it is a way to induce that, we can know that earlier.
But when it comes to then people's perception of prediction markets overall, they're going to be thinking more about this manipulation and insider trading and all these other concerns, and less about, say, on the CPI, where you actually do have a problem in that we collect government data through surveys. These have been heavily disrupted over the last few years. There's already issues with them, but now they've been more heavily disrupted. They were never continuous, and yet they underpin economic decisions. They're extremely valuable to have a real-time CPI. And people can, with dispersed information, legitimately bring, from the retail side, I think, meaningful information there.
And so if you think about this kind of trade-off of concentrating more in growth in places where there is legitimate, meaningful information coming in rather than legitimate but kind of insider information, there is a lot of gain in trust and validity.
Right
I think those are places you'd want to invest in if you want to see this as a multi-trillion dollar industry rather than a multi-billion dollar industry.
Well, to that point, the industry is trying to police itself, and
Yeah
They are singling people out if something looks just whatever. I am not sure how they get them on their radar, but they have singled out some folks for this sort of thing. That brings to the question.
Yeah, but those are just on the margins.
Right.
They are creating whole tracks of things that are going up.
I guess the question is that invites government scrutiny, obviously.
Yeah.
To what degree do we want to see government scrutiny? We want to also have a flourishing market. We need some maybe guardrails, but to what degree that it doesn't change what the prediction markets are. Toby?
First of all, the exchanges themselves have an incentive to not want to have inside information.
Right.
Let's be clear, right? My experience is markets function much better than governments at handling these kinds of things. Having said that, I'm sure there will be some regulation, right? Our current financial markets are regulated as well. We have rules. In fact, the CFTC governs these exchanges. Those rules already exist. They just need to be enforced and implied, and I think they are. But I think that the key point you want to think about here is that the exchanges themselves, if they want people to trade, they've got to eliminate that kind of stuff because nobody wants to be adversely selected on the other side of that. The question is, how do you detect it? How do you punish it? How do you prevent it from happening? As those markets become more liquid, it's very hard to do.
The most liquid markets in the world, probably currency markets, very hard to have any inside information there and get away with it. It's just too efficient, right? We're not there yet, but I think that's where we're headed, or that's the hope.
Emil, I brought up artificial intelligence and some of these folks who really do very well right now, they use AI, they have agents, they have small teams of people betting for them. It's a serious operation, but when you start to add that into this, how do you think that will change? Because it feels like there's a certain point where people are looking at the prediction markets, and it's like cognitive surrender. I no longer believe what I thought I believed because now the prediction markets are saying something totally different than I thought when I woke up this morning.
Right. Which is good. Good for you, right? You need a second opinion anytime you can get one. But AI is truly a game changer. Look, I've been running prediction markets since 2000. There's probably nobody alive, not only in this room, but in the entire world, who has been running markets for as long as I have. It's a love story, and it's a love story because I read a paper from this man, 1999, and it just changed my life completely. I said I would dedicate a quarter century to following that prediction market dream, whatever, right? But AI arrived a couple of years ago. It wasn't able to do any kind of prediction in 2003, and then progressively it started to do adequate predictions, and now it's at the level of super forecasting predictions, right?
We can have these AI forecasters now that are just taking away completely this idea, and they're surpassing the wisdom of crowd, if you like. It's like wisdom of crowd in a box, right? You can apply it to anything you want. It's a thousand times faster, it's a thousand times cheaper than having superforecasters or running a prediction market. Prediction markets are going to be great. They have a great future for online betting, right? Because it's fun and people want to bet. But in terms of how to generate the best forecast possible on things that are important. Nowadays, and for the future, AI is going to take that over.
Right.
It will help you to place your next bet on Polymarket for sure.
You agree with that?
except everybody will do it, so.
Robin, you agree with that?
Ordinary finance companies, hedge funds, have been trying to get leverage with AI for decades. They keep trying. My understanding in the last few years, they have not had a big burst of success. They are still working at it and still on the margin, slowly getting better. That's maybe a cautionary tale that whatever success you're seeing doesn't generalize to other markets. But I think the more fundamental point is just I, as an economist and in general, want institutions that are robust to questions like this. I want to go to a prediction market price and be able to believe it, and I don't care whether it's AI or people trading. They're competing in this forum, and the whole point is this forum's going to sort out who participates. It's also true that I don't care whether it's rich, elite people or poor people trading in the markets.
I don't care if it's men or women. I don't care if it's Democrats or Republicans. The whole point of this as a robust institution is that I can see that it has incentives to sort this out, and I don't have to make second guesses about the institutions based on my judgments about who's participating. That goes for AI versus humans and all the other divisions. That's one of the wonderful things about these institutions as information sources, is they are quite robust to biases and manipulation.
I look at the AI trading similar to how I look at AI adoption across many other fields, which is to say, where is the human continuously better? Where is the AI better? Seeing, as they term, the jagged frontier will keep on moving depending on how regular the outcome is, how easy the data is to access, how idiosyncratic that data is. It's going to make it easier for AI when it has very regular things and when those regular things allow it to be judged, and therefore, correct and move forward. Right? The more irregular, the more likelihood that humans are going to have information that they've added on top of it. But I agree with Robin.
At the end of the day, for a lot of us, it doesn't really matter who the traders are as long as the information is being aggregated to the system.
But Toby, you think that matters for the general public, where there's already a massive backlash against AI in other areas, and anyone holding data center stocks this last week understand what I'm talking about. Ultimately, as a platform that could be $10 trillion, the key element there always is going to be trust, and will the general public say, "I trust this thing if there's some super intelligence that has an edge over me?
I guess it depends on the objective. If you care about accuracy, no, I agree with my colleagues up here, which is, I just want to get the best information I can to make the most informed decisions I can, and this is going to make it more efficient and you're going to get the information faster. If your goal is you want to make money on these markets, it's probably gotten a lot harder. That's right. Because you're now competing with tools that can move way faster than you can. You need resources. You need access to data. You need clever ideas. It forces you to be a bit more creative if you want to do that, but I think that's a totally different objective.
But I'm not sure that's what we want markets to function as, is a playground necessarily for. If this is a viable market, and I think it is in terms of completing lots of risks in the economy that we can't currently trade, then we just want it to function as efficiently as possible. Then, looking at it from that lens.
Right
I think it could be a very good thing. But may not be as fun to trade otherwise, but that's okay with me.
You wanted to add something, Robin?
People mind AIs doing movies and songs and math because those are high prestige things. But plumbing and cleaning sheet out of the sewers, they don't mind AI doing stuff that they don't respect and just want to get done. That's the question about these markets. Is this a high prestige thing where we celebrate the winners, or is this just plumbing that needs to get done? I think it's more the latter. For most people, they just want these markets to function, and they aren't really allocating a lot of prestige to the winners.
I'll add one more positive twist, though, which is, I don't think AI on its own, it's getting We'll see. My answer may change a year from now.
Right.
Just caveat that for a second. But my experience, and I work at a hedge fund. I'm a partner there where we use a ton of AI, and successfully. It still requires a lot of human input at this point. That may change at the rate it's going, but the benefit of it, what generates the alpha, what generates the insight is, to David's point, thinking a little differently. Using the tools because they give you a ton of leverage, right? They work 24/7. They never get tired. But knowing how to use them requires still that human input. I still think that's going to be important, but we'll see.
I have one small spin on that, is that a question we've had for a very long time is about the ability of the exchanges themselves to have algorithmic market makers inside the systems.
And where they may have an advantage over traders outside the system. Again, this is a question of transparency and/or regulation to make people feel comfortable because I think that it is impossible to regulate out AI traders because people will just get to the barrier, whatever that is. But if there is a structural advantage, that is where I think that you are going to have more doubts. I also think that there is a lot of anti-AI, anti-crypto, anti-prediction market on the left, and this is an existential threat to prediction markets in the longer term if it becomes an incredibly partisan issue, and moving into more stable and robust areas will limit that exposure.
Got a minute and a half left. You have been doing this longer than everyone else. What happens next?
AI.
Yeah.
Because for a simple reason, is that wisdom of crowds is great. Markets are great, but you need a crowd to be interested in the questions you ask. Otherwise, there is no market. What AI allows you to do, which is the game changer for me and for the business I am in, which is making forecasts for people who want to have forecasts, is that you can ask. The universe of questions that markets and crowds can answer is very small compared to the universe of questions that decision-makers in different organizations actually want to ask because they are so specialized, so complicated, maybe so long-term, that you will be hardly able to gather a crowd to be interested in answering them or wagering on them.
All right.
That is the opportunity.
Thank you very much, Robin, Toby, David, Emil. Fantastic stuff. I wish we had more time. Thank you very much. Thank you. Hey, guys.
Thank you, thank you.
Good stuff, man.
Nice to see you.
All right, that wraps it up for our first day. You got to come back here tomorrow. We've got a couple of different panels on the election. I'll be very involved in that. I know we dabbled in it a little bit. We'll do more of that tomorrow. Tonight, we got a great networking event at Elsie's Rooftop, 1412 Broadway from 5:00 to 8:00. Drinks and food. Classic OPM, right? Other people's money. Drinks and food. We will see you there tonight from 5:00 to 8:00. Elsie's Rooftop, 1412 Broadway. Doors open tomorrow at 8:00 A.M. We'll see you tomorrow. Terrific first start. Thank you for coming.