Prediction markets are invading everything

- I think we're live. - I see there's one person, we'll give it a second and see if anyone else, someone was waiting. - All right, we got some people coming. Is there a title on this one? Do people know what we're talking about? - I put, what did I put? I put the future of prediction markets or something like that, but to me for the live, and then you can fiddle with it if you're gonna publish it afterwards. - Yeah. - Awesome. Well, I think we've got a few people here. We can get going. I'm here today, my name's Chris, by the way, co-founder, CEO of Substack, and I'm here today with Alex Heath, tech journalist, author of sources and co-host of the weekly podcast Access, with over a decade of covering tech and media outlets, like The Verge, The New York Times, The Wall Street Journal, Alex is a legendary voice in the industry. Thank you for coming on my Substack and asking me questions. Thanks for having me, Chris. A lot of questions. The news is coming in hot and heavy. - Yeah, hit me. What do you got? - Okay, yeah, I like this. Just flipping it over to me immediately. So I would love to start with how this partnership came about, get a little bit into why you're excited about prediction markets, and then ask some concerns and questions that I've just been hearing from other substackers. If that's okay. So let's start with, yeah, how did this partnership come about with Polymarket? - How to come up with Polymarket? One bit of backstory here is I'm kind of a long time prediction market nerd. People may not realize, but before kind of the advent of the modern prediction markets, they kind of like have started to work at scale. It's been kind of like a long-term sort of sci-fi fascination of certain tech nerds, basically, because of the sort of potential, at the time, potential power of having a market that aggregates information about sort of like, you know, the common knowledge price of people's predictions for future events. And there's been kind of like a long-term sort of wonkish interest in the potential power of having a prediction market as a way to yet aggregate information about predictions, as a way to like figure out on aggregate, what people think is gonna happen is this kind of like theoretically very interesting sort of piece of technology. It's something that I think has been kind of like a long-term interests and hypothetical thing. And then within the past few years, that possibility has started to become real. There have been a number of companies that have been making kind of at scale prediction markets that are actually doing this stuff. And I would argue that some of the cool promises of the technology have also started to become real. And so the things that you, you know, I think people very forgivably look at some of these prediction markets and kind of mentally lump it in their mind. They're like, oh, this is another one of these tech money making schemes. It's like crypto or sports betting. You're like, all of these things. And it's very understandable that people would think that in my mind. And there's a truth to it, to be honest, like I think there certainly, there's people who are betting on sports on prediction markets. That's a real thing that happens. But the piece that's very fascinating to me and that I think is maybe underrated is the degree to which having a clear price on salient events that are actually important, like a legible price on not just is so and so going to win the game, but like what's going to happen in this election or what's going to happen with this geopolitical thing or what's happening. What's the real thing that we're going to discover about this technological breakthrough? We're starting to see markets like that develop and in my opinion have very useful information. And so my interest in this from a sub-stack perspective is not on the hay or people betting on stuff or they trading stuff or they like, all that stuff is happening, but the part that feels very interesting to me is being able to see and discuss the price. Being able to see and discuss like the market probability assigned to a real world event is this kind of like interesting and novel thing and I think it's going to be a more important part of how we understand events, how we understand the world and even for people who are critical of it, being able to kind of like reference it, discuss it and clearly display what's happening in these markets will be a powerful tool. And so I was chatting with Shane who runs Folly Market and we've been to, I've been kind of, I've been cooking like, we've had this integration for a while, people have been kind of using it. I've been interested in, could we do a more fulsome version of it and we figured out a way to get a shift. - Has there been a specific moment for you with prediction markets in the last year or so where you feel like it's really clicked? Usually when I talk to other, I mean, you're totally right by the way that this is a long standing fascination I think among your peer set and it's been something that-- - Us nerds. - Us nerds, but you know it's like, I feel like every CO I talk to these days is thinking about prediction markets and how to integrate them or buy a platform if they can. The hype is real and I'm curious if usually when I talk about that with them, they mention a specific moment like an election or something like that. Was there a moment where you saw either the outcome or the process and what, okay, this is really clicking for me? - There were probably two moments that I think of as for myself kind of seeing this new version. I mean, one was, you know, the last federal election, presidential election. You know, I think I was sort of glued simultaneously. I was glued to the Nate Silver Silver bullet on Substack and in my other tab, I kind of had like the prediction markets. Like, hey, what's the, you know, the way that you would previously have dug into like the New York Times Needle or something? The what is happening aggregate information, I think it was like a very powerful up to the minute source of just like which race is gonna go which way. It was sort of interesting to see that in action. And I think a lot of people kind of had that experience where that was sort of like the place where they found the earliest, the earliest kind of like consolidated signal of what was going on. And then the other moment that happened earlier last year after that was, I was trying to figure out what was happening with the TikTok ban for a while where it was like it had been banned, but it was legally banned, but it wasn't actually banned. Is it gonna go off, where's it gonna go on? And there was a moment where, you know, I looked at the prediction market odds that there was basically a market on what are the chances that this service actually shuts off for some amount of time. And the odds of the time were, you know, it wasn't a majority case, but it was substantial. And it made me, it was enough that I was like, oh, I think this might actually happen. And could we put some effort into like preparing for that and being ready for that to happen and being able to help people who are, you know, suffering from that. And it ended up that it did happen and that it was like, it was down for like an afternoon or something, basically. But the fact that I could get, you could see kind of a market signal on, hey, there's a substantial chance that this event is gonna happen and then potentially change how you plan your business based on having that kind of like probability estimate. It felt very powerful. - Yeah. Okay, so we've talked about prediction markets broadly and why you're into it. This deal specifically, I think people have a lot of questions about. So why are you doing this as an exclusive with Polymarket? If SubSack is a platform and infrastructure for creators like you talk about, why not do CalChi as well? Why is it just Polymarket? - I think the interesting part of the integration to me is having the price, like having the, the well displayed thing. And I just think this is like, the integration we've been able to do is gonna provide that really well. - And you don't think you could have gotten that integration with CalChi as well? - I think it's possible. I mean, I wanted to find a version of the partnership that worked really well. - Yeah, is SubSack getting paid by Polymarket for the integration? I'm not gonna get into the financial details. One thing we did announce is that they have joined our sponsorships beta. So there are paying people on SubSack in the pilot as part of this. - Because what I've heard in the context of X and some other platforms where Polymarket has done similar deals, though this is a deeper product one that they are paying and that CalChi does not. And that's a difference in how the two prediction markets approach these partnerships. Any comment on that? - I think my take on this is, I wanted to make a version of this integration that I thought was a great product. And I think we've been able to do that. - Okay. So I think the reason I'm also asking that There's a credible argument from creators on the platform that if Substack is a creator, friendly creator, first platform, and Substack to be getting paid by something like Polymarket for this, should that money occur to Substack? Should it occur to the creators? And yeah, we can talk about the ad piece as well, but can you respond to that first? - Yeah, this was part of what, you know, I'm not gonna get into the details of the thing, but this was part of what made this appealing and made it like a worthwhile thing is we do have this sponsorships beta that we're running, and they are participating in that in a way that I think it's gonna be, it's gonna be cool. - Yeah, and I was, for full transparency for everyone, I was approached by your team about that and about embedding a chart and was gonna get paid for that. And I thought about it and I thought, you know, I mean, I have a perspective as a journalist where I try to be careful about, because I am a journalist, I wanna be careful about the brands I'm associating with if I, you know, cover the space. For example, I've covered Colchian Polymarket in the past. So that was another reason to give me pause, but I also just wanted to give pause because I feel like it's very early, and I feel like people have a lot of, I think, very credible concerns about prediction markets as you set up top. And it was interesting to see you all thread together, you know, the substacts and the announcement posts that have done the deals with Polymarket so far, and the examples of the kind of charts they're embedding. I'm curious, like, was that ad piece a key part of the deal for you if you were gonna do something like this, or was that something that like Polymarket offered? Like, was it something you wanted? Like, what, take me into that a little bit more? - Yeah, I mean, you kind of mentioned this. It's like, I see the substack, the value that substack rates is kind of the value that people are getting paid. Like, our flywheel is, you know, people being able to make money doing the work they believe in. And so the piece that's always the most exciting to me is, is there a way that the people that are making stuff can get paid doing this? The other part of this just to be really honest is, I love, I'm very excited for this integration. I think the tools are really cool. We wanted to make it completely optional. So these are things that are available. If people want to use it, there's a sponsorships pilot that's available if people are interested in it. But there's no, we wanted it to just be like, if you're not interested in this, if you don't wanna participate in it, there's kind of just like no change in your life whatsoever. And we've been able to put it together that way in a way that I'm pretty excited about. - Yeah, I'd love to come back to ads, but as a journalist, there was a quote in the announcement that really stuck with me that I wanna hear you expound on, which is, journalism is better when backed by live markets. And this is in the polymarket announced. This is an announcement, excuse me. Why is that the case? Do you agree with that? - You know, that's not the framing I would use to express that sentiment. I think that there's, I think that there's a really healthy skepticism of prediction markets and the particular prediction market companies among journalists and writers. You've seen this, I see this in the, it's been a very interesting, I've had kind of like a very polarized reaction from people on Sub-Sac about that. Some people are like, this is awesome. This is so cool. Finally, I can put this thing and some people are like, "Ah, this is horrible." This is horrible. Like we hate these people. And I think that's, well, that's not the right way to, I think that's like a, not the way that I would put it. The way that I might put it is, I think having access to prediction market data can be a very powerful tool for journalists. If it helps them do the journalism they're doing. And so having that be available, having that be something that people can use, I actually think that is true. I think that it's, you know, the thing that a prediction market does and the thing that journalism is, is fundamentally different. It's not the same thing, but it can be, depending on the journalism, complimentary, and it can be a powerful source or a powerful thing to comment on or thing to write about. And so I do think that the way I might put it is, prediction markets can be a powerful tool for some journalists. - I think that may be the case. I'll be honest, so the two main concerns I have about prediction markets are not the overall idea because it's a very cool idea. I agree with that. I think it's the practicality of two key things, which is who is on the other side of the trade and how our disputes resolved. And those are two very key things and they actually differ based on the prediction market. So-- - And you mean disputes in terms of the resolution criteria? - At some point, the market has to end. - Correct. - And it's like, okay, well, who did this happen on how are we judging that? - And for people who don't understand, I mean, Pauli Market was historically, you know, it has not operated out of the US, inside the US, excuse me, legally, until very recently. And they would use a essentially crypto blockchain mechanism globally to resolve disputes. And there's been credible research pointing to a tremendous concentration on that token and the fact that people who have large amounts of it can effectively run the network with this way that's gonna be different in the United States. They're working with the CFTC and all of that. But have you looked at that? Have you spent a lot of time looking at that mechanic and does that give you any kind of worry about just the scalability of this? - I haven't got-- I think it's a very interesting problem and a very real problem. Like certainly if you're thinking of kind of the engineering of what does it take to make a prediction market that actually works. One of the hard problems is how do you do resolution at scale? And in particular, how do you do resolution at scale about the kinds of questions that people are most interested in asking? There's a tension between things that are easy to resolve, but those might not be the things that people would be most interested in finding out the information about. So I think it's a very interesting problem. I haven't been deep into all of the ins and outs of the crypto mechanism that Polymarket uses. But I do think it's like a very interesting problem. One thing that I think is true is that anybody that's going to have a market has to have a good answer to this and has to ultimately, if they have a resolver that doesn't work, like the market won't work. - And I think people are pointing to early examples of how Polymarket and Calci do get things right as like, okay, there's something here. But yeah, I just, I worry that there's going to be a huge trust moment, credibility moment, crisis. If people do realize that this isn't actually the wisdom of the crowds, it's the wisdom of like who has the most tokens. And that's a concern that I think is very legitimate. - Yeah, I think if for these things to live up to their promise, they need to have trustworthy resolution. - Yes. - And that's a non-trivial problem. My assumption from the outside is they're working really hard on that. And that's like a major part of what makes these things great. It can be an access of what makes these things hard. It can be an access of competition. - Yeah. - I don't think it's unsolvable. And I don't think that it's the fact that there are challenges with it means that it's, you know, you throw the baby out with the bath water. - Do you know if this partnership is the global Polymarket to tier the US only one? 'Cause I believe they're fairly split. Like is this, do you know if the Polymarket trading and markets is the global network or the US network? - I'm not gonna get into the details of the partnership. I can talk about like what the integrations we've done are. - Well, I think it matters to the integration 'cause it's like the kind of markets you can put on the platform, right? I think they're different. Maybe I'm wrong. - Yeah, I think you can put all the interesting ones. - Do you think this gets to a place where sub-stack writers create their own markets for their coverage areas? Like do you see it going that deep? - I'm very interested in that. And again, this is something that people can be a part of if they want to and are interested. But to me, one of the sharpest critiques of prediction markets is, hey, this is just for, this is just kind of like an excuse to have trading on some non-event, right? It's like, hey, we're gonna be a sports thing or it's gonna be like there's gonna be some sort of a guffin thing that's, you know, you're having the, all of the risk taking with none of kind of the social value of having it be a question of actual importance. And to me, the place where the interesting intersection happens is, are these things, are the questions, are the salient questions ones that matter? And so the idea of having, you know, people who are writing about the world and recovering the world have some say in what markets get created does seem very interesting to me. - I agree. Again, if that resolution dispute and who's on the other side of the trade thing gets fully resolved, I think to me until that, I know for certainty that it is a fair playing field, I think I'm just taking away and see approach to this. And like, did you weigh that as a platform owner? Like when you were thinking about this integration, like the fact that there are people like me do feel that way, like, I mean. i'm wondering if if you think it's just better to experiment in the real world and figure it out iterating, you know, out and and and shipped versus the weight and sea approach. yeah i'm a you know i'm a big ship and fight out guys you can probably tell it was like let me put it this way the way that i do think it's very important for us is you know we wanted to make this completely an optional opt-in thing right i didn't want to be foisting any of these features on people that don't want it i kind of wanted it to be like hey here's a a powerful tool that we've created that i think is going to be potentially very valuable and that if you want to try out you can try out if you want to use you can use and if you don't want to you don't have to yeah that to me feels like a a good balance of those things it's where we're getting contact with reality we're building the thing that works in the actual world people can try it if people are enthusiastic about it they can give it a go we can sort of like start to find out what's actually where the value actually is and what's actually happening and if people don't like it or they try it and it doesn't work for them they don't have to use it is this the deepest kind of integration like this you've done with an outside partner i believe so do you want to do more like these not just with prediction markets but are you is this signaling any kind of a bigger thing happening with sub-stack and partnerships or is this more of a one-off i'm very interested in anything that brings creative and expressive power to people who are writing and making things on sub-stack and i think you know whether whether we do that kind of first-party where we're marshaling technology that does this or whether we do integrations or partnerships that kind of there's so many things beyond prediction markets in the world right now that the technology is changing very fast on and one of my obsessions is like what you can do theoretically is increasing so quickly but then like actually giving that the power to make it usable to people like make it so that if you're you know somebody who's trying to make something it's actually there at your fingertips in a way where you can easily play around with it you can easily kind of like experience some of these crazy futuristic things i think it's a really it's an important part of what we want to do okay so maybe more interesting product things to come there with the time we have since this is part of the sponsorships program how is that going it's funny like when i talk to different members of your team it seems like there's been a i don't know what the right word is there's been a slow consensus forming on the strategy but you guys maybe started this early with the goal of getting feedback and figuring out like how we actually are going to scale this and i'm still a little unclear i guess i would say on like the high level view of sponsorships on sub stack and how it's going to play out um i would say it's very early and we're getting a lot of good signal yeah um and you know to me the center of this is just like we exist here in service of creators and people making things and we believe that you should be able to make money doing the work that you believe in um and there's just there is a big a big interest in demand for this do you see a marketplace this ending up in like a a true marketplace where it's it's direct you know access to any creator who wants to be a part of it or any brand who wants to be a part of it versus the closed thing it is now i think it's probably too early to prognosticate exactly what it will be um but you know i think i think it's going to be good i think it should be interesting and it's going to be in service of uh people who are making great things respond to the obvious uh critiques that this is something that you know you need to do to justify your valuation or that there's business pressure on you to do something like this yeah i kind of like those don't really make sense to me i guess i see the you can see the logic right i can see the logic i can see the logic where it's like oh yeah you know something something vcs to me the core of subset the thing that substack is at its root is an economic engine for culture and that our entire mission in the world is help people go independent make money doing the things they actually believe in and so my obsession is not how much money we make as substack my obsession is how much money are we making for people who are making things on substack and there's there's tremendous brand interest in substack i can the test to that first hand and independent media and it's growing so the timing is good i mean i've had companies invite me into come speak since i went independent in september or about how does substack work how can we as a brand just see like show up see what's going on so yeah you are tapping into like a moment and i do think like for me when i launched i thought oh subscriptions subscriptions will probably just naturally become my main source of income and i think my revenue mix is probably gonna be minority subscription forever i think i mean we'll see i'd like to have subscription be durable and be something that can weather an add-down yeah or something like that you will definitely go through that you'll notice that the subscription thing is very is very uh yeah but the but the brand opportunities real for certain kinds of creators and eventually yeah if you guys start taking a cut of that it aligns i guess in the same way that the subscription model does right i think that's the key and this is why we're sort of it's so important to build these things in the real world and and figure out the thing because i think that there is a real thing here and the key to for substack is to make the version of it that's actually good and that actually serves the creator and actually helps people you know make money doing the things they believe in because there's so many ways that it can go wrong obviously you didn't talk about that for years i mean that was substocks positioning at the beginning was almost anti ad driven platforms and i think a lot of that like that's that's still accurate right there's there's still ways to do this stuff that breaks the the fundamental compact in a way that you know wouldn't actually help help anybody in the long run but i think they're the things that's increasingly clear is there's also ways to do it that are that actually help and one of the things that is motivating us here is you know there are so many of the of very good people on substack who are also figuring these things out for themselves yeah like me and so it's not i don't see it as my role to come to you and say hey like you know you can only ever have this one way that your business works yeah but i do think it's my role to say hey there are ways that you could have your business work that are aligned with the dream you have and the quality you want to create and it's going to help you make money making the thing you really believe in and you know we can help with that that's kind of where i where i want to show up and is alignment for you where the end result of this that could be very bad is substock starts optimizing for engagement and the feed and everywhere else right and that the advertisers become your customer and not the creators is that the main thing for you to like keep from happening or is that too simplistic yeah that's one that's one angle on it that's there's like there's so many ways it can go wrong the way that it goes right is when the subscriber and the creator and the platform are all aligned which maybe hasn't been done yet well that's why it's important to have a such deck okay well i feel we're doing an interview on your channel but are we at we at time do you want to time thank you for thank you for letting me interview me after introducing you yes so weird move but thanks for thanks for answering the questions at least most of them all right all right thanks thanks bye cheers you