Episode Transcript
[00:00:00] Speaker A: Predictive programming episode 13 and we got a guest today. We have Gaten Douglas, a prediction market trader, one of the top Spotify traders out there, but I know he's, he's dealt in other markets before. I've enjoyed reading his posts over the last few weeks highlighting various issues with some of the markets that he specializes in. So very happy to have him here. Gaten, welcome.
[00:00:24] Speaker B: Hey, thanks for having me on.
[00:00:26] Speaker A: I think we want to start by just asking you a little bit about who you are, where you come from, how you got into PMs, what's your story?
[00:00:33] Speaker B: Sure. So online in Gate and Dugas, but in real life. My name is Caleb Davis. I live in a suburb of Minneapolis.
My background was I studied finance and I got my mba. But most like my whole work career has been in a mix of it and the business. It's a role called being a business analyst.
So that's what I've done.
I got into prediction markets because I used to just be really interested in politics. So I debate with people all day online about politics. And then I saw an ad for a website called Predict it back in 2015 and it said, bet on Joe Biden if he's going to be the presidential nominee or if he'll try to, if he'll be a candidate. So I thought, oh, that looks cool. So I deposited 3, 300 or $500 to predict it. And that's kind of how I got my start. So for the next several years when it was just predicted, I focus exclusively on politics. But then once Polymarket came on board, then I started expanding my scope to some news events and some pop culture stuff. And then once Kalshee went big on the pop culture, then I've started hammering it ever since. And over the past two years on Kalshee alone, I've made around half a million dollars between politics and pop culture, in addition to a couple hundred thousand on other sites as well.
[00:01:59] Speaker C: So is that all from the initial $500 deposit or is that, have you added more to increase?
[00:02:06] Speaker B: No, over the years. The way I do it before is when it was on Predictive. One, I was a lot poorer than I was now, and two, there just wasn't interesting stuff all the time. So I'd deposit, I'd play a bit, and then I'd withdraw the money and then I wouldn't do anything on the site for months at a time. Then I'd redeposit, which was dumb because Predict it had 5% withdrawal fees. So I was getting hammered with the fees.
But recently, now that there's so many topics, I can just leave money in there.
On Kalshi, I deposited $10,000 total in March of 2024, and that's what I built into my bankroll.
[00:02:50] Speaker C: A nice little 50x return. Is I pretty good?
[00:02:53] Speaker B: Yeah, yeah, it's worked out well. I don't know If I deposited $10,000 now, if I could do that same type of return because a lot of the markets that I focus on have gotten a lot sharper over time. But when I, when, when they first introduced these markets, because I had already done them on Polymarket years before, I was just way ahead of the competition.
[00:03:14] Speaker C: Yeah, I was pulling up your profile earlier. You, you obviously dominate in the entertainment in politics, but like Basically, I mean 375,000 for entertainment, 175,000 for politics, and then everything else is under 10,000. So what is your like day to day? Like, are you just looking for like entertainment markets that you want to like attack and then when political events happen, because you know, political events are not very frequent, just then spending time.
[00:03:42] Speaker B: That's basically it. So somebody on Twitter set up an account at my request that sends notifications every time a new market is posted on Kalshi.
And it doesn't do like the hourly or the daily markets that are the same all the time. But anytime a new unique market gets posted, I'll get a notification on Twitter now. So then I'll check those out and kind of scope out anything that I'm interested in. And then as far as the repeating ones, I'll usually do rotten tomatoes each week.
I'll keep an eye on Spotify. But lately there's been kind of some bad news on Spotify for me, which we'll get into a little bit later, I'm sure. So I haven't been doing those as much, but I just keep an eye out for opportunities where things might be interesting or where I might see something that the rest of the market's not seeing.
[00:04:33] Speaker A: Okay, so first thing I want to ask, when you first load up into predicted back in 2016, are you, would you describe yourself as recreationally trading markets at that point where you like, was there anything algorithmic?
[00:04:49] Speaker B: It was hard for anybody to really do more than recreational.
If your only site was predicted, which, which mine was just because the volumes weren't that great outside of election day. It's like the individual vote markets and stuff like that, they just didn't draw that much attention. Plus the $850 limit really put a cap on how much you could earn like some people made a hundred thousand on there, maybe, but it was nothing like today where there's basically no limit, no ceiling on what you can earn.
And.
[00:05:26] Speaker A: At what point do you start feeling like, oh, I think I have an edge here. Like, I think I'm, I think I'm winning. I think this is something that I could make real money doing. And how does your approach and behavior change after that moment?
[00:05:43] Speaker B: So it mostly comes through bet sizing as far as the changes, as far as the ad. You just look back at your performance over the year and you over the years and you can see how you've done. It's like, I'm not somebody that has like a 95% win rate or anything like that, but I do consistently win more than I lose. So I'll have a down month here and there just by the nature of the, the markets that I bet in. But if I look over the long term, it's a consistent upward slope on my earnings. So. So, like, that's, that's usually not luck, if that's happening over such a long term.
And then from there it's just a matter of managing your bankroll, sizing up on your bets without going completely stupid on it so that you risk losing what you've. You've already earned.
[00:06:33] Speaker A: At what point in time, by the way, do you, do you have a day job or is this what you.
[00:06:38] Speaker B: Do for a living, strictly? Yeah. Yep. I'm, I'm still a business analyst. So normally throughout the day I've got my, my personal computer to my right with Kelshi up and Twitter up and then my work computer to my left. And I just spend all day like that. So like, I'll be in meetings at work, but keeping an eye on stuff, that type of thing.
[00:06:56] Speaker A: Are you, are you modeling any of the markets you're trading or is it more discretionary?
[00:07:03] Speaker B: Like the politics stuff isn't really modeling. All of the music and the rotten tomato stuff is. Yeah.
[00:07:10] Speaker C: Okay.
I'm curious if you don't mind, could you take me through, Like, I don't want you to leak any alpha, so don't say anything you don't want to say, but a particular trade that you made that's memorable to you or like a recent one? I want to know like your thought process is and like, do you come up with a couple questions I would have is like, do you come up with a fair price before and say, okay, I'm going to buy one side up to here? Are you trading both sides?
If you're getting a good deal on both sides. How does a typical trade work for you, and what's your process in the entire thing? If you could walk me through that.
[00:07:45] Speaker B: So my style is almost exclusively buy and hold or buy and sell if it hits a price that I think is too expensive.
I'm rarely jumping from side to side.
So I'll give a recent example, and this is one that's still very much live. There's a market for when Venezuela will have their next elections.
Trump deposed of Maduro in Venezuela like, a week ago, or maybe not even a week ago. And then there's this market on how quickly they'll hold elections. So for something like that, the first question. There's actually several questions, but the first one I'm looking at is, what's Trump's motivation in doing this? And if you look, it's kind of been all over the place.
He first said it was due to drugs, which was bullshit, because Venezuela, they're not a big producer of cocaine. They really don't.
[00:08:41] Speaker C: It might have something to do with that 17.3 trillion in oil.
[00:08:45] Speaker B: They traffic it and finance it. But, like, Colombia is way worse as far as cocaine goes, or way better.
[00:08:50] Speaker A: Depending on who you are.
[00:08:52] Speaker B: Way worse as far as fentanyl goes. So the drugs thing was. Was just nonsense. So I knew that that wasn't the reason.
So then, then there's, like, it could be geopolitical security. They just don't want Russia and China having a foothold in the Western Hemisphere. It could be that they're after oil. It could be that they want to expand democracy.
And I think if you followed what Trump himself has said, it's pretty clear that his main motivating force is oil. Like, he wants to get the resources from Venezuela, and he's perfectly fine having a dictatorship there as long as we're getting our cut. Right? So, unlike Bush with Iraq in the early 2000s, Bush was all about, oh, we need Iraq to become democratic. And the steps he took to get it there were clearly in line with that goal.
I don't see that from Trump. I don't think he really gives a crap about democracy in Venezuela unless it furthers his other desires, which is to extract resources.
So the first thing I was looking at was motivation, as I just described.
And I didn't see building democracy as a big motivation for Trump.
So to me, that means that it's going to be a slower process than if he was really interested in building a democracy there. And then the next thing I looked at is like, okay, let's say that he really wanted to do this. How long does it take to run a fair election in a country like Venezuela where the elections recently have not been fair?
It's a lengthy time period. Like the, the quickest situation like that might be like 11 months. But with Venezuela, you've got the situation where like a quarter of the country left under the past two presidents. So you'd need to track down like update your voter rolls with people that are spread out all over the world. So I, I think from a logistics perspective, if they wanted to do a fair election, it's, it's just going to take forever to, to get that set up.
And then I look at, okay, how could I be wrong? And I think the only way I could be wrong on this one is if Trump decides to hold elections that aren't fair. Like if he thought, okay, we'll just throw some snap elections, we know they won't be fair, but we'll get somebody in there that we can work with. That's the one risk. But I don't think that's very risky. So basically based on all that analysis, I said, okay, it's very unlikely they're going to hold it this year. And I just kept putting in the top bid.
[00:11:31] Speaker C: When you say this year, do you.
[00:11:32] Speaker B: Mean 2025 or 2026 before June or something like that? Before September and before the end of the year? So I would just put up 5,000 share blocks so that I always have the top bid.
[00:11:44] Speaker A: And are you basically, by the way, always doing make orders and being patient, or do you occasionally take whatever's available?
[00:11:51] Speaker B: It depends. For something like that, where it's a pretty liquid market, I'll stick with limit orders. But for something like Rotten Tomatoes or Spotify, a lot of the times I'll have a limit order there. And these are pretty low volume markets. So if I see a block like 2 cents or 3 cents above what my limit order is, I'll just grab it because there's no guarantee that I'm going to get those shares otherwise.
[00:12:12] Speaker A: Okay.
[00:12:12] Speaker B: So unfortunately, like, the bad part about that is last year I think I had like 320,000 in fees in earnings, but I paid like 80,000 in fees to Kalshi because so many of my markets were ones where I just would grab those and just eat the fees.
[00:12:30] Speaker A: Do you find though, with your limit orders, especially on something like Rotten Tomatoes, Right, where people know there are rotten tomatoes sharps out there that you can.
How often do you run into. Oh, I placed a Standing limit order. And I watched the market move off of that. Like somebody jumped my price or the other side open orders move further back.
[00:12:52] Speaker B: Because I.
Yeah, right, yeah, it's extremely common to see that. So when I do see a block that is a price below what I want to pay, I'll just grab it, even though I've got to eat the fees and maybe pay a little bit more than the market thinks it's worth right now.
[00:13:10] Speaker A: Am I correct that you used to be a little bit more of a Poly Market guy and now you're a little bit heavier on Kalshi?
[00:13:17] Speaker B: Yeah. So I switched in October 2024 when.
When I got an email saying that my name was part of a subpoena. And I'm just like, okay, well, it's. It's perfectly legal on Kalshi.
I might as well just move my money there and wait for the. For Poly Market to get to the US legally before continuing to. To invest there. And in fact, like, Kalshi's been my biggest site, but before that it was this American Civics exchange, one that's not very popular yet.
That's where I made most of my money in 2023.
[00:13:54] Speaker A: Interesting.
Was it at all part of the decision to move from polymarket to Calci, the fact that your account is public on Polymarket and people could copy, trade or see what you're on?
[00:14:11] Speaker B: Yeah, absolutely. For me, that's a big downside of Polymark and having your positions public. For example, last year I built a position on Bad Bunny being the most streamed artist on Spotify, and I started building this like the second week of January, and it wasn't a very liquid market, so I would just have to wait to get fills the whole year.
But had I gone in there, like, had people been able to see that I was building that position, they very easily could have jumped ahead of me and taken any of the shares that I was hoping to get. And since I'm fairly well known on Twitter in this space, I think my account would be targeted for that quite frequently. Yeah.
[00:14:56] Speaker A: So let's talk a little bit more about that, about account tracking and whatnot. And I think that ties in well to what's been going on with the Spotify market. So can you talk about the daily, most streamed Spotify market? What exactly it is? I know it sounds not that complicated, but often the devil's in the details.
And then we'll get into what's been going on with that.
[00:15:21] Speaker B: Yeah, absolutely. So the daily Spotify markets, there's actually several of Them. But the two main ones are just what song gets the most streams each day. And there's two markets. One is at the global level and one is at the United States level. And those markets, the tracking window for those is 7pm Eastern to 7pm Eastern every day.
And then they post the results the following day around 9am Eastern.
So there's just one data dump each day.
And basically to. To be successful at them. There's the. The main thing is just tracking trends and seeing how. How are the songs performing over time.
How does that compare to how similar songs have. Have done historically?
And then projecting out. You're really just projecting out one day or sometimes two days to. To determine if there's going to be a flip or if the song that's in first place will remain in first place.
And you just pick your song and you bet on it.
And the other part of that is that most Fridays there's new music that drops. So it drops it at midnight Eastern. So you can scope out those songs and see, okay, this artist always gets millions of streams the first day. Or this artist can do that if they've got a really good song. So you can check out that song and kind of make a call for yourself if it's going to be a hit or not and just kind of play it that way before we move.
[00:17:02] Speaker A: On too much because you just said it. But I want to dig into a little bit. So let's say like a pretty major artist like the Weeknd or Drake releases a record on Friday. Is it at all part of your process to listen to that record and just like ball knowledge be like, that's a banger. Or this is trash.
[00:17:19] Speaker B: Yeah, absolutely.
[00:17:20] Speaker A: That's really funny.
[00:17:22] Speaker B: In 2024, you probably remember this song was all over the place. Morgan Wallen Post Malone did a song called I Had Some Help. Do you remember that?
[00:17:31] Speaker A: Yeah. Oh, yeah. Who could forget?
[00:17:33] Speaker B: So that there were like grainy videos of them doing this song at a concert.
And I could tell from those clips, I'm like, oh, this is going to be huge once it drops. So once it drops, I put big money on it. Winning Billboard because I knew it was just going to be huge. And then of course it dropped and it was absolutely huge.
It's actually more common to do it the other way though is I'll find a song that's from an artist that has had success in the past and I'll listen to it. I'll be like, okay, that's horrible. There's no way that that's Going to work out. The most recent one was Dochi had a song with sza and both of them have had songs that have done very well. SZA in particular will usually light up the charts when she does a new song, but the song was pretty horrible. And this is a part was pretty small, so I knew to fade that one as much as I can. But yeah, that's a part of it. I'd say that's maybe 5% of those markets, though. And then the other 95% is just math.
[00:18:35] Speaker A: All right, I didn't mean to interrupt you on that, but I thought that was cool.
[00:18:37] Speaker B: That was a good question.
[00:18:39] Speaker C: I'm curious more about the trading part of it.
What, like, what are you using for. Are you using Kelly criterion for like your sizing when you said like big position? Like, is it big position as in like it was the biggest for you at the time, or is it big as you thought you had a big edge so you want to increase the position. What are you thinking about from that point of view?
[00:19:00] Speaker B: So Kelly criteria doesn't really come in on the pop culture markets just because the constraint is almost always the amount of volume available.
So I'll almost never like, if I were to calculate how much I actually want to bet on a market at a price, I'm almost never going to hit that volume on a culture market.
[00:19:25] Speaker A: Classic end boss status.
[00:19:26] Speaker C: Yeah, no, that makes sense.
[00:19:28] Speaker A: Who amongst us hasn't been there? It's just I take what the market can give me.
[00:19:34] Speaker C: So I guess from that point of view, how do is do you. Are you doing anything to like increase your size or anything like that? Like, I guess another question I would have is you said earlier you didn't trade both sides, but say you can't you eat up all liquid. Like if you say you believe something to be a fair of 60% and the market's trading at 45, so obviously incredible bet. So you. You trade it up to like 58% and just eat up all liquidity. And now people might be like seeing the markets moving, copy trading that not knowing it's you but seeing the movement and get above 60%. Could that not happen a lot and you increase your size that way? Or why does that not happen or have you done that?
[00:20:18] Speaker B: It could in theory in the culture markets. I don't see it very often on this week though. Here's an example of doing that.
Was the Greenland market. I don't know if you saw that one.
I saw, I think it was on Sunday.
There was some news about Stephen Miller saying that he wanted to take control of Greenland and military isn't off the table. And then Rubio said some stuff like that. And I saw that the price hadn't really moved. It was at 35 cents for. Yes, for taking control of some part of Greenland before 2029. So I bought 5,000 shares at 35, just thinking that the market hadn't priced in this new information yet. And then later in the day, it got up to 45 cents and I'm like, okay, that's too much. Like 35, it's fine. But 45, that's not a position I want to hold. Potentially to 2029. I think the 45 cents reflected that price. So I sold at that price and I think it's at like 40 cents now. So I sold it.
[00:21:22] Speaker C: Perfect. Perfect.
[00:21:24] Speaker B: But I didn't buy the other side or anything like that. Although I could have.
[00:21:28] Speaker A: Yes. Let's navigate back to Spotify. And there's one other thing I want to ask, just in the realm of foolishly obvious questions, but if you go onto Spotify, there are down to the ones place play counts on all these tracks. I'm assuming they are too asynchronous and unreliable to make part of your tech stack for pricing daily plays.
[00:21:50] Speaker B: So there's two data sources. So the. If you look on Spotify itself or you go to a web browser and look at Spotify at the individual track, you'll see a stream count. Right.
But if you look at the charts that serve as the resolution source, they won't match. And the reason is the streaming counts on the artist or on the song's profile page. Those are raw, unfiltered streams, which means that they could include botted streams which came, which was important, as we'll talk about.
Whereas the chart streams are supposed to be filtered. So they take a shot at removing these botted streams before they post them to the chart. So the unfiltered raw streams are always going to be higher than the chart streams. The other part is that the chart streams usually update first. So you'll see the charts update with the filtered streams and then they'll update the song profile pages shortly after that. So there's really no advantage to be had from looking at those pages.
[00:23:02] Speaker A: Okay, so you've been trading these daily Spotify markets pretty successfully, and then something happens. What happens?
Give us the timeline.
[00:23:13] Speaker B: So this was in the last week or so. So there was a really interesting day.
I think it was the day before New Year's, where Olivia Dean has a song Called Man I Need and it's about you, Chris. That song overtook Taylor Swift in early December. But then the holidays kind of put, put all of that action on pause and it's just all Christmas songs for a while. And then immediately after the holidays, Taylor Swift was in back, back in first place and in second. Third was the K pop demon hunter song golden and Olivia Dean's song Man I Need. So there's like this three way battle for first and Taylor Swift won it the first couple days, but her margin over Olivia Dean was shrinking every day.
So the margin got down to like 8,000 streams on one day. And then my calculation showed that she was very likely that Olivia Dean was likely to pass her the following day.
So I was buying some of that. It was at like 30 cents or something like that. So I was buying some of that. But then what happened was as soon as the tracking window closed, Olivia Dean went up to like 80 cents. Which given the data that we have, it didn't. Like nobody could be that sure.
[00:24:33] Speaker A: So you're saying like 7:01pm when there's still nothing has changed.
[00:24:39] Speaker C: The flip, how much liquidity does that, does it move from like you said, about 30 cents to 80 cents?
[00:24:47] Speaker B: It wasn't a ton, maybe 6,000 shares.
[00:24:50] Speaker A: And to just state the obvious, this rarely happens usually.
[00:24:53] Speaker B: Right.
[00:24:53] Speaker A: Like is there any spike at all at 7pm? Like does it usually. Is there any kind of shock? So 7pm is a nothing.
[00:25:00] Speaker B: I think there's jumps throughout the day, but it's a market that's based off of a single daily data drop. So 10 hours later, yeah, there's a data drop and then it, it adjusts accordingly. There's, there's some proxies that you can use throughout the day, but they're, they're not.
The correlation between a proxy like Apple Music or YouTube isn't anywhere near high enough to give you that level of certainty to bet that volume at those prices midday, basically.
[00:25:28] Speaker A: Okay, so it goes to 80% by the way. First day this happens, are you then personally baited into buying Taylor Swift at 15% or do you immediately smell something fishy?
[00:25:41] Speaker B: No. So I, I bailed. I sold my Olivia Dean shares, I took a profit, but I didn't jump into the other side because I thought it looked suspicious.
It was just one day, so I couldn't like draw any conclusions from it.
[00:25:53] Speaker A: I gotcha. 80% was way overvalued by your numbers, but it was smelly enough that you.
[00:25:57] Speaker B: Weren'T actually gonna get it. Yeah, pretty close. I had her as like a 55, 45 favorite. Because the thing is, it's like you can build the perfect trends on Spotify, but there's still a lot of variance to it. So if you're thinking like, oh, this is probably going to win by 12%, sometimes it just doesn't.
Especially when something after the holidays where you don't have a ton of historic data to be able to base it off of. And then there's also multiple trends. Like if you looked at the day to day trend, Olivia Dean would be a favorite. But if you look over the week, over the week to week trends, typically Taylor did better on Wednesdays than Tuesdays and it was going from a Tuesday to Wednesday. So it's like, okay, maybe that trend is going to, going to outweigh it. And then. But if you looked over time, like throughout the month, that Tuesday to Wednesday jump was, was dropping each week. So there's all these different trends that you can look at. You've got to decide which one makes the most sense.
But the fact that there's so many of them introduces all this variance that you just can't be like super certain of these things. So you have to price it accordingly.
[00:27:06] Speaker A: All right, so first day you see fishy behavior. Olivia Dean does indeed resolve to be the big winner.
[00:27:13] Speaker B: Yeah.
So that resolved to be the big winner. Suspicious, but it also wasn't like a confirmation or anything.
[00:27:22] Speaker A: Oh, by the way, did you post about it on Twitter that day or go to support that day or did.
[00:27:27] Speaker B: You say this is not that day? That day? It was just kind of in the back of my mind something to watch for because it was a one day thing. So it could have just been somebody that like really loved Olivia Dean. Who knows? But then the next day was New Year's, the time period that covered New Year's Eve. And there were four songs that were in contention. The Taylor Swift song, the Olivia Dean song. Because it was New Year's Eve, I thought that maybe a lot of people would listen to golden as well. And then the Stranger Things finale was that day.
And the guy that plays Steve Harrington on Stranger Things has a song called End of Beginning. So I thought all of those four were in contention.
So throughout the day, I was buying shares of the the End of beginning song between 1 and 2 cents and I was buying golden between 2 and 3 cents. So I had a couple thousand shares of each, just dirt cheap.
And then once the tracking window closed, the prices flipped again and golden took off. So golden was 3 cent shares and it just Took off and suddenly it was at 80 cents or something like that.
[00:28:42] Speaker A: Has golden ever been number one before?
[00:28:45] Speaker B: Yeah, it had been number one a lot previously, but its time has kind of passed. It had been quite a while since it was number, so because it's. That song came out in July, so it's, it's, it's kind of had its run, right?
And mostly it had been in second or third in the previous month, so would be a surprise. But there's, there's a case for it to be first just because you're getting a bunch of kids together at New Year's and they're listening to music and staying up late, but that the fact that it went from 2 to 3 cents to 80 cents, right when the tracking window closed, it's like, okay, there's no way that what's basically a lotto pick becomes a certain pick that suddenly unless somebody has access to the data.
And then the next day the same thing happened, except this time it happened throughout the day.
The, the Steve Harrington song that in the beginning started out as a lotto pick. And throughout the day its price just kept climbing and climbing until it was like an 86% favorite and that ended up winning. And then on the global side, like, there wasn't any price change throughout the day. And then when the tracking window closed, end of the beginning shot up to extremely high prices as well. So it was four.
Four underdog or long shot songs that went from underdog or lotto picks to overwhelming favorites throughout the course of the day when. When nobody's supposed to have access to that data.
So that, that tells me that, okay, the most likely.
The most likely cause here is that somebody at Spotify is just querying their database intraday and starting to trade based on that. And the other, I think, less likely scenario is that Spotify is somehow leaking this data publicly and somebody found a way to access it.
[00:30:46] Speaker C: I want to play devil's advocate here for a second.
Hypothetically, it's possible that, say, these markets became so liquid that big players like an institutional trader came in and decided to build just the most unbelievable model. And is it not possible that someone has done this and they're blowing everyone out of the water with a better model? Or do you just like, no, you refuse to believe that's even a. I think insider trading is far more likely, but I'm just trying to throw it out there.
[00:31:19] Speaker B: I don't think that we get intraday data from anywhere that has a high enough correlation with the Spotify data that could give you that level of confidence because like the. We get Apple music data, we get itunes Data, we get YouTube data, but all of those have problems with itunes and Apple music. They get updated throughout the day, but all you see is ranks.
So you see relative positions, but you don't see actual stream counts. And on YouTube you can see the view counts. But I found that the correlation on YouTube is very low.
And a part of that is because YouTube counts are global.
And just the viewing patterns there tend to be different than the listing patterns. And there's other things. Like Google Trends, you can see how a song's trending, but again, the correlation isn't that great. I've seen a lot of times where a song has done really well on Google trends and then the Spotify data comes out and it really didn't amount to anything.
[00:32:26] Speaker A: Yeah, I mean, this is like leaked draft pick kind of.
[00:32:29] Speaker C: Oh, yeah, yeah, this is pretty. Look, it's pretty. I'm just trying to throw it out there.
[00:32:33] Speaker A: No, of course. Yeah, Good question. So at this point, you stop trading the markets, you post on Twitter and you reveal your findings and say, this is incredibly suspect.
[00:32:43] Speaker B: Yeah, because I don't feel like I have an edge on this anymore.
I might be able to take some dumb money really early in the morning and. And to your understanding, not worth my time.
[00:32:53] Speaker C: This right here is clearly why insider trading is bad when. When people like poly market guys and you know, all the crypto bros, like, it's the source of truth. No. You have a trader here who's providing massive volume now refuses to do to participate because he doesn't want to be fish food for an insider. I don't blame it. I mean, it's ridiculous.
[00:33:13] Speaker A: Has the standing liquidity dried up on these, by the way? Have you seen less?
[00:33:17] Speaker B: It's hard to tell because that end of beginning song has been so dominant.
[00:33:22] Speaker A: Okay. Which kind of changes everything.
[00:33:25] Speaker B: It like opens up every day at 99 cents, so there'll still be people buying 1 cent lottos for it, but we can't really tell until we get another day.
[00:33:34] Speaker A: That's only interesting when there's a horse race for the market.
[00:33:37] Speaker B: Like, Bruno Mars is gonna drop a song tonight, and so that's probably gonna be interesting. But normally that's something that I would trade, but I won't bother this time because somebody else is gonna. By the time I'm starting to get any useful data, somebody's gonna have eight or 12 hours of actual data that they can trade on.
[00:33:59] Speaker A: So did you communicate this to anyone at Kelshi and what kind of response did you get?
[00:34:06] Speaker B: I just tweeted it out. I didn't go into the Kelshi cord or anything because somebody else said that they already brought it up there and like I'm not holding my breath that they're actually going to do anything about it or if they do do something, they'll just say this is a personal matter that will investigate and then they won't reveal any findings, which isn't very useful to me because I really need to know that there's not somebody that's looking at the data that's going to take my money if my bid is wrong.
[00:34:36] Speaker A: So how like on a scale of 1 to 10, how much of a problem do you think this is? Do you think of this as sort of like these are the table stakes of playing this game? Are you like they absolutely have to find solutions to this kind of insider trading or the platform is fundamentally broken?
[00:34:55] Speaker B: I think that it kills markets because I posted a poll on this. I don't know if you guys saw that.
[00:35:02] Speaker C: Yes, I saw that one.
[00:35:04] Speaker B: So the question is, what's the equilibrium for a market with insider traders? And the three options were a mix of insiders and non insiders, insiders only or nobody trades. And like this it's an economics problem with an actual answer because eventually you're going to run out of people that are just giving their money to the insiders.
So you're going to, you're going to have a mix of insiders and non insiders only as long as you've got new people coming in that don't know any better.
The idea of insiders trading against each other doesn't make sense because if they have access to the same data, they should know. I mean they're going to be on the same side of the bed or know the right pricing. So they're not going to trade with each other. And then so the equilibrium is that there's just no volume because the, the non insiders leave and the insiders don't have anybody to fleece anymore.
So it sounds great to say, well it's, it's insiders is it's the only way to get a truth, the truth. And that might be true in the short term, but there's a couple issues with that. One is who gives a shit what song is going to be number one on Spotify on January 12th? There's no informational value there.
And two, in the short term that's fine. But you're looking for long term success and long term volume. And that's going to dry up over time, and you're just eventually going to end up with markets with no volume whatsoever.
[00:36:36] Speaker A: Have there been any cases where Kalshi has identified insider trading and publicly acknowledged it and done something about it? Is there a confirmed case that was handled properly?
[00:36:51] Speaker B: Not that I'm aware of. In fact, it was not too. Recently, Tarik was on some show and he was asked about that and he said, like, no, there has been no insider trading. When there was a very obvious case of it back in September with the Bad Bunny announcement. Super bowl headliner.
I don't know if you guys saw that.
[00:37:10] Speaker A: I did. To me, there often are. Like, I remember the story with the Nobel Prize winner was it looked like insider. Then it turned out that there was a website that had leaked it, which is like, somebody beat the announcement, but it wasn't insider information. And so I.
I never know for sure. Like, you mentioned with Spotify, like, it would be easier to be an employee looking at a database, but it could be some broken website where in the metadata you can see something like that. There's always some chance of that.
[00:37:42] Speaker B: Yeah. And I'm actually like, the Nobel Prize thing or the Spotify wrapped leak, I'm actually okay with that. Like, if it's hard to find public information that's different than information that you have access to by virtue of an employment contract and you're supposed to be protecting, there's a difference there. But what happened with the bad money thing was that I had some money in that market, so I started getting alerts at some ridiculous hour saying that the price had gone up. So then I was watching the volume and people were putting up fairly large bids at a price that wouldn't make sense unless you knew something was up.
And then I went to Polymarket and I saw the same thing was happening there with some new accounts.
So I tweeted at like three in the morning, like an official prediction that Bad Bunny will be announced as the headliner of the super bowl today.
And that ended up getting traction. Like, the Bad Bunny fan accounts on Twitter came across this tweet and started amplifying it, and then it ended up being correct later in the day. So that was kind of cool.
But it's an example where if you didn't have that insider info and you were just doing some analysis and trying to predict who was going to be the headliner, you were going to be donating your money to somebody who actually knew it.
[00:39:07] Speaker A: Highlights attention. Because these sites, they need to be healthy. Marketplaces, but they also want to be exciting sources of news.
And inside information is great for the latter and terrible for the former.
[00:39:23] Speaker B: It's one of those things where insider information might work fine for these one off markets, like the attack on Venezuela, the bad money super bowl thing, because you're not relying on a daily cadence of this market to generate volume. Right. So it's not like a systemic problem at that point. It's a one off thing. Right?
[00:39:48] Speaker A: Yeah. And the user has no guarantee about when this information is supposed to be public. Right. The surprise announcement is priced in already.
So whether or not it's a public or private announcement is kind of irrelevant.
[00:40:02] Speaker B: From a user perspective.
You should exercise caution and you should understand like what the timelines of these things look like.
For example, there's a Grammy performer market.
Right about now is the time when lots of people are going to know behind the scenes who's going to be performing at the Grammys. So you're not going to want to leave big limit orders open in those markets just because there's going to be like a point in time where it's like flipping a switch where a bunch of people are going to know this information and you aren't and you don't want to open yourself up to be taken advantage of in that way. But you can still trade them when it's early on, like once the Grammy nominations were announced, you've probably got a couple month window or a month and a half window where you can trade because it's not determined yet. So there's no insider information to share.
[00:40:57] Speaker C: Another question I have is when you're making a lot of these open orders and limit orders, are you only comfortable if you're in front of the screen?
If you go even like say you've got to go get lunch or something and you go out for 30 minutes, are you canceling everything and making sure or are you going to bed with any open orders? What is your fear level? Just in any market in general, it.
[00:41:20] Speaker B: Varies widely by markets and whether or not I leave open limit orders depends heavily on the market. I've got limit orders on the Venezuela stuff because I don't think there's going to the odds of them sudden announcing elections within the next couple of days is almost zero.
Fair.
But I wouldn't have like open orders on anything Spotify related going into a Thursday night, for example, because they're going to announce or even a Thursday because oftentimes on Thursday they'll announce what's dropping at midnight.
I try to figure out what the dangerous times are and not have open orders for something. Like there's House votes today and there's some markets on those House votes. Like I wouldn't leave that open while I'm doing this interview because I could just get taken to the cleaners because somebody's watching the vote and I'm unable to.
[00:42:15] Speaker C: So then going back to the Spotify thing, what would you like to see from Kalshi in this thing? Like if they said we've investigated and we found no source of would that satisfy you or what would make you happy here?
[00:42:28] Speaker B: Well, like if they investigate it and say that they didn't find anything, I still wouldn't trade them because I don't think that there's a compelling case that somebody just did those trades without some sort of insider knowledge to get me back. They basically need to say here's what we found here was the punishment or they've been banned from this stuff, some sort of consequence. And I think that that type of thing would be important just as a deterrence to future ones because if people see that there's consequences to insider trading and they work at a place that gives them non public information, they might hesitate to use that. If they know that there's been a precedent of punishments or bans or whatever, hand it out. But as long as that's not there, there's really not a disincentive to use it.
[00:43:24] Speaker C: Right. So now I'll play devil's advocate again here. Going back to the Nobel thing. That wasn't insider information. That was just like a very unique spot that he was the only one to find this information.
If it's a spot like that where someone has figured out a way to get information that's very valuable and only they have what is and that is discovered. What is Kalsh supposed to say? They can't like out the guy's edge, so to speak.
[00:43:51] Speaker B: But no, absolutely not. So if they find out that that's the case, then they can say. They can basically say that's the case, it wasn't insider trading. But you still come across the same problem. And then it's up to me, okay, do I want to try and find out the source of the information or not? I probably wouldn't want to try it. One, because I might not find it and two, if I did, it would just put me on par with them instead of giving me an edge over them.
[00:44:19] Speaker A: Right, all right, we're getting towards the end here, but you are a predictor. Par. Excellent skate. And so I just want to ask for one prediction for prediction markets in 2026. Where do you think we're going? What's something you think we're going to see this year?
[00:44:34] Speaker B: Sure. So I think the big news of the past year was just sports getting on the platforms and basically overwhelming the volume of everything else.
And I think the reason that that's been been the case is because we've been in this legal gray area where it's, it's not illegal, but it's also not legal. And I think that it being that gray space has, has dissuaded some competitors from going all in on building competitive products.
So I think that this is going to be the year where either sports are ruled to be illegal and that just goes away and the volumes on these things drop dramatically or they're ruled to be legal and we see a flood of high quality competitors. And I think that the volume is going to get spread out amongst multiple platforms instead of just one like Kalshi dominating.
[00:45:33] Speaker C: I hope Novik is a part of that spread of volume.
[00:45:39] Speaker A: All right, Kaden, thank you very much for joining us, man.
[00:45:41] Speaker C: Really appreciate it.
[00:45:42] Speaker B: Thank you. Yeah, thanks for having me.
[00:45:43] Speaker A: Yeah, excellent. Take care.
[00:45:47] Speaker B: Of.
