E01: And we're live

<cite>Phil:</cite> <time>0:57</time> welcome to episode one of the Game Economist Cast. I am Phil at a Game Economist Consulting, which is a invisible company until today, hopefully when this cast is releasing everything has come together. formally Amazon Game Studios Dice as well as Scopely. I am joined by two amazing co-hosts who have a lot to share and I believe I can introduce you as Dr. Smith. Is that something we can do this week? <cite>Chris:</cite> <time>1:22</time> See. I don't know, because official academic people from the university have been like congratulations Dr. Smith. And I'm like, oh, cool. But I technically walked or gotten my degree. That won't happen for another two weeks. I don't know. I'm just <cite>Phil:</cite> <time>1:32</time> I think it, I think it's fine. Can you operate on a patient? <cite>Chris:</cite> <time>1:34</time> Well, I told my friends, I was like, I don't like you guys impersonating a doctor, so you need to call yourself physicians. For the ones who have an MD or a do degree <cite>Eric:</cite> <time>1:42</time> you're not president yet, you're president elect. <cite>Chris:</cite> <time>1:44</time> Yeah. Yeah. It's kind like that. I did the hard part and and now I can just sit back and relax. Well get back to work like but now my weekends are free up. <cite>Phil:</cite> <time>1:52</time> So welcome to the cast. Dr. Smith. Do you mind telling us a little bit about who you are and where you're from <cite>Chris:</cite> <time>1:57</time> Where I'm from, I'm from the us but mostly the Midwest. I've spent a lot of my time. where I come from. I did my undergraduate in economics and I went straight into a PhD program, in economics at Purdue. And during my time there,, I got involved with, game economics, the same way that a kid gets involved with drugs and like high school. I've pretty much been doing that ever since. <cite>Phil:</cite> <time>2:17</time> There's always a gateway drug into this weird subdiscipline. <cite>Chris:</cite> <time>2:19</time> Right now I'm at,, star Atlas or Atha, working on their giant space in rpg kind of trying to bring that star citizen kind of elite, dangerous type of experience, into web three. <cite>Phil:</cite> <time>2:30</time> Hailing, from the great state of Illinois. <cite>Eric:</cite> <time>2:33</time> I'm Eric, working at Super Layer, doing kind of web three economy design, X Riot, it's where I cut my teeth. Yeah, math and econ, like these folks, but, probably the least accredited. I'm, I'm just a, plug in Bachelor, <cite>Phil:</cite> <time>2:44</time> Bachelor. No, not true at all. Not true at all. I've seen, I've seen your tweets. <cite>Eric:</cite> <time>2:48</time> Yeah. <cite>Phil:</cite> <time>2:49</time> Those are tweets of a seasoned game industry veteran. <cite>Eric:</cite> <time>2:52</time> Yeah, it's seven years at Riot. Kind of see all sides of the industry, but, yeah, I love, love game economics, I think I got into it, just from the, the YA cups. Yeah. The guy, I don't know how to pronounce his name, the TF two guy was the first I <cite>Phil:</cite> <time>3:05</time> Gs. Gs, yeah. Yeah. He's many of our gateway drugs <cite>Eric:</cite> <time>3:08</time> and I was like, oh, this is like a thing. And I was like, oh, you can take games that I love and do all sorts of data stuff with them. Okay. This is awesome. And,, here I am. <cite>Phil:</cite> <time>3:17</time> And for you, Chris, did you come through Edward Cast Nova, he's the big name in academia for this, for g for game. Econ. He seems to have the most papers. <cite>Chris:</cite> <time>3:25</time> Yeah, so I was gonna say if I, if you guys have, Giannis is your gateway drug. I would say probably Ted is my, is my gateway drug. he ended up on my PhD committee, which was really cool. we, we ended up writing a paper together, so, yeah, kind of has to be him. and, yeah, I've got tons, tons of respect for the, those like, old guard guys. and now I heard from Ted. It's okay, it's in your guys' hand now. Like you. Go, continue to continue to fight for game economics <cite>Phil:</cite> <time>3:50</time> Well, hopefully you can send him this podcast. I think that gets us at least one guest appearance by him if he was encouraging you <cite>Chris:</cite> <time>3:56</time> I think he would be, I think he'd be up for it. we could definitely convince him to come on. <cite>Phil:</cite> <time>4:00</time> This guest list is already getting full. I think we have three seasons worth of guests already. <cite>Chris:</cite> <time>4:05</time> Or at least at three seasons, then <cite>Phil:</cite> <time>4:06</time> three seasons in a four, no, four seasons in a movie, six seasons in a movie. I got my franchises mixed up. <cite>Eric:</cite> <time>4:11</time> That. Last season, that sixth season was pretty bad of <cite>Phil:</cite> <time>4:13</time> Oof. Yeah, as soon as Pierce went we have three articles to talk about today. We have three beautiful articles to talk about, or at least something approaching articles. I will be talking about a piece I wrote on Marvel Snap, doing some shameless promot. For a game that I've been obsessed with and a topic that I continue to write about, it's torturing me. I need to, I need to crank out these two other pieces. When you find you, you write these pieces and they start to expand and expand, and then you break them into subparts. It's not a good idea. You think it's a smart commitment device, but it's not. It's just a torture chamber for yourself. Eric, what are you gonna be talking about today? <cite>Eric:</cite> <time>4:48</time> We can talk about Sky Weavers talk about the trends in web three games, right? It used to be like,, play to earn and now they're like, talk, everyone's talking about stability, but I think they're overshooting in the wrong direction. <cite>Phil:</cite> <time>4:58</time> So we'll check out a little bit about Sky Weavers and Chris is there a dissertation chapter in our future? <cite>Chris:</cite> <time>5:04</time> We would be a remiss not to, chat about my second chapter of my dissertation, which focuses on basically video game firm behavior. So it's kind of an IO problem. And we will, we'll explore the gambit of different, industry models. <cite>Phil:</cite> <time>5:16</time> And I would say, I think there's almost too much going on in crypto to talk about it this week, but I definitely think in the future we'll be spending a lot more time with crypto. All three of us are involved in crypto in some way. Eric, you've made the full leap over to web three from web two. Chris, you just, you just dove straight into Web three. I'm somewhere in between. I'm working with both clients. It's a very strange place. Hopefully none of the listeners know that we are fanatics. Would you describe our or web three positions as sensible, sane, hopeful, confused, sober. Sober? I think that's the great way to put it. A sober web three group. We spend a lot of time critiquing arguments about the cost and benefits of Web three relative to web two. With that being said, let's dive in <cite>Eric:</cite> <time>6:02</time> just wonder if I should have my diagrams with boxes and arrows. <cite>Phil:</cite> <time>6:06</time> Oh man, that, that would take this cast to a whole nother level <cite>Chris:</cite> <time>6:09</time> But any sort of diagram with arrows and stuff, people love that shit. They're like, oh, like, and actually I love it too. It helps me visualize, like anytime I get a G D D, I just start making diagrams. I can't keep all this in my head. <cite>Phil:</cite> <time>6:22</time> It's almost as if the process of making the diagram is more important than the output of the diagram. I, I think you really have to discipline your thinking when you make diagrams, which is a perfect segue to Marble Snap, which I did a lot of diagramming for So I published a piece on Marvel Snap trying to really just talk about the monetization of Marvel Snap. As a summary, when you think about where Marvel Snap has been, it comes from a studio of former blizzard employees. Hamilton Chu who is the CEO of Second Dinner was an executive producer on HearthStone, and Ben Brode who is the Chief Development Officer at Second Dinner was the design director on Hearthstone. And these two, I think, were really responsible for making Hearthstone the juggernaut that it is today. It's been in development for about four years, and surprisingly it was only in soft launch for six months. It was an incredibly quick soft launch. It looks like they just got the KPIs that they needed to read. I don't think monetization was one of'em. Perhaps they were just testing for retention. That's pretty common these days. And does anyone know where the, where they spent the six months, was it a tier, tier one or tier two or tier three country? <cite>Chris:</cite> <time>7:20</time> No idea. <cite>Eric:</cite> <time>7:21</time> Yeah, and it's almost always like Australia or Canada <cite>Phil:</cite> <time>7:25</time> Well, Sweden, Sweden's a nice tier too, represent, yeah. But apparently Australia and Canada are big spenders. <cite>Chris:</cite> <time>7:30</time> That's interesting. Do, do you run into any, like sampling bias biases because of <cite>Phil:</cite> <time>7:35</time> Tons. And I would say it's an unexplored problem. I would say it's ill-defined and I've been disappointed that data scientists have not gotten more involved in it. <cite>Eric:</cite> <time>7:43</time> I think they specifically choose countries that are similar to target demographics. Canada for us, or Australia for, just the western world in general. <cite>Phil:</cite> <time>7:50</time> and you don't need to do localization. So you don't have to think about language. Everyone speaks English, <cite>Eric:</cite> <time>7:55</time> Dingo ate my baby <cite>Phil:</cite> <time>7:56</time> is that English? Eric <cite>Eric:</cite> <time>7:57</time> It's localized. Yeah. Anyway, <cite>Phil:</cite> <time>7:59</time> So it was only about six months in soft launch before Marvel Snap ended up going worldwide., so far the metrics, and I think was this after maybe about a month, six weeks after launch, is looking at about, probably around 8 million downloads. Interestingly enough, they published this with Bite Dance of TikTok Fame rather than Net's. And Net's was a strategic investor in second dinner. So rather strange. And then if we had some news today that Blizzard has broken up with net's in. So they will not be able to publish games for Blizzard in China. And of course, you need a license in China to be able to sell games. There's a very limited ability to sell games in China. And so without that, blizzard is gonna be without a very large revenue stream. So we don't know what's going on behind the scenes, but I find it pretty interesting that Net's is also an investor in second dinner <cite>Eric:</cite> <time>8:48</time> Got a rumor for you is that, I heard that breakup was hot and messy. Like it was not a planned, executed thing. There was some kind of negotiation where they could not see eye to eye. <cite>Phil:</cite> <time>8:57</time> so not a strategic move, <cite>Eric:</cite> <time>8:59</time> salt on both <cite>Phil:</cite> <time>9:00</time> both. Okay. Okay. Some, some high emotions. <cite>Chris:</cite> <time>9:03</time> Where do you get your, where do you guys get the, the like, traditional industry? like juice? <cite>Phil:</cite> <time>9:08</time> t tmz.gamer.com. <cite>Chris:</cite> <time>9:11</time> what was that? <cite>Phil:</cite> <time>9:12</time> tmz.com. That's not a, that's not a real website <cite>Chris:</cite> <time>9:16</time> Oh, <cite>Phil:</cite> <time>9:17</time> Although now I want know what comes out of tmz.gamer.com. <cite>Chris:</cite> <time>9:19</time> nothing. <cite>Phil:</cite> <time>9:19</time> Bummer. So it looks like a, a messy breakup with Nets and Blizzard, which makes this strategic investment in second dinner, which was years ago, they invested all that more interesting, but they ended up publishing Marble Snap, not with Nets, but with Bite Dance. And I assume, again, this relates back to the expansion into China that I'm sure they're ironing as well as making sure that they have the license to even be able to operate Marvel Snap in China. Although I'm not quite sure how, how Marvel has done in China. I know Matthew Ball has some research out there. It'd be interesting to see how the IP resonates with that audience. But six, six, 8 million downloads probably so far. And it's hard to get good numbers on active users. we only really get installs that are direct from the first party publishers app, a data ai, whatever you'd like to call it. A lot of those are estimates when you look at active user numbers. So I'm a little bit skeptical of that. So I decided to take a look, at least for this article at STEAM Db. Which does report actuals straight outta steam. And we can see a very nice early PSU stacking chart. So player simultaneous users continues to increase day over day over day, and that is a good sign. That means that they are retaining at a higher rate than they are churning players. Let's see what happens when the UA funnel drops off. That is always the moment of truth for these games. When you pull back on spend, are the downloads still there? Can the brand carry you forward? Have you built up a large enough dau war chest to be able to monetize in the future? But so far things look good from an early engagement perspective, but I think this is where we get into the meat and potatoes, at least of my argument, which is that Marvel snap to me, they try to reinvent the wheel when it comes to monetization in a way that didn't appear to be necessary. One of the things I do in the article is spend some time working backwards from how you spend money into the game, to finally how you actually end up getting value out of the game. So when you end up going through that process, one of the things you find is that there's a core economy loop in Marvel Snap that really revolves around three components. And those three components are to be able to collect cards, variants, character boosters. And then once you collect those cards, variants or character boosters, you frame break the cards or you cosmetically upgrade them in exchange for some currency. And so by frame breaking, you then get collection points. And those collection points contribute to what is almost an account level progression system and account level reward track. The more collection points you earn, the higher up the track you move. The higher up the track you move, the more card you get. So there are a variety of rewards on that track, but most notably, that seems to be the exclusive place in which Marvel Snap new cards are sourced. If you're a player and you're not so much interested in the cosmetics of frame breaking, which just adds a small cosmetic effect around the border of the card, it might be animated. They have a 3D view, which is interesting. So if you were to tilt your camera, it almost looks like the card has a 3D effect. So they have these very small cosmetic effects. But if you're not interested in that, then the only thing they really have in terms of progression is you frame break. So even if you're not interested in cosmetics, you need to frame break. You need to frame break to be able to get the new cards. That is the primary vector in which you level up. And so my issue with Marvel Snaps monetization so much is in that framework so much as it is that seems like a rather strange way to source cards to. So, one of the things that I think we learned in Clash Royal is just really the power of marginalization or modularization. So if you have a really high value item in a game economy and you need to break up the value and pars it out, almost make it more fungible. If we were to take a word from the crypto universe, if we were to make it more fungible, then what you do is you chart it. So if I, for instance, have a hundred shards of a particular item, I might not be able to realize that item or to summon it or to cast it or to make it real. This felt like it would've been a better way for Marvel Snap to distribute cards rather than having this single reward track with kind of these Intermitent releases. And what this also does is it lets you back into loot boxes, which is something that I feel is sorely missed in Marvel Snap and almost to the detriment of the players because Marvel Snap does randomly source you cards. when you go into your award track and you've collected enough collection points and you go to get one of those core cards and your collection levels, That is a random card. Not only that, but when you're at the end of a round in Marvel Snap, when you're at the end of a match and you're collecting your booster xp, that goes to a random card in your deck. So they have these loop boxes in the game. They're just not showing you the animations for it. And I just find it really hard to believe that a team that spent so much time on animation and production quality, which is just absolutely off the charts in this game, every card has an animation. I find it very strange that this team would not put in loop boxes into the game or put in a loop box animation as an oversight. It feels like a purposeful decision to avoid the wrath of Reddit. And this was something that happened in Launch Soft Launch. They had this Nexus event, which was a high stakes loot box, events. And again,, they didn't have charting, so you were either getting the card or you were not getting the card. So it's a very high risk. So any given box just has a really low probability of you getting the card when you open it. And so they ran this event. They had a lot of community ear, a lot of community backlash against it. I think the question when you end up in this situation is always, did you get the price wrong or did you get the mechanic wrong? If you go back, for instance, to Apex season one, there was a lot of anger at Apex Battle Pass. Was it the Battle pass or was it, the battle pass being too hard. And so I think those are things you have to really think about and consider when you are met with backlash in the community, is looking past what they're saying to what they mean. And so this to me was a very strange outcome is for them to take away the lesson that, hey, we aren't just gonna do this anymore. I'd say the other thing is the price was rather high for these cards. It was about$400. Someone had calculate and read it again, I hadn't verified that independently, but about$400 to get a. <cite>Eric:</cite> <time>15:09</time> To get <cite>Phil:</cite> <time>15:10</time> $400 to get a card during the Nexus event. So they had two cards available,$400. <cite>Eric:</cite> <time>15:16</time> Were they like very powerful cards that were exclusive to the event or yeah. Why? <cite>Phil:</cite> <time>15:20</time> So,, when you think about these cards, a lot of'em are supposed to be horizontal, right? They're supposed to be horizontal in the sense that one isn't intrinsically better than the other. And you can achieve that in game design by altering the man acot right? Now, of course, like that doesn't always end up being the case at all, but that at least in many cases is the design intent. And I feel like this design intent here. But of course whenever there's a new card, it is going to grant you some. Level of vertical progression if the card is useful, like the card should be useful. So these were useful cards. In some sense it does increase your wind probability even though it isn't the intent of any one card to do so. <cite>Eric:</cite> <time>15:52</time> And, and they were exclusive to the event. <cite>Phil:</cite> <time>15:54</time> They were exclusive to the event. I think we don't know a ton yet about how Marvel Snap will start to source these cards going forward to people who weren't a part of the event. They're not gonna use the FOMO strategy. That would be rather strange for a card game. But it looks like what they do is they put past cards that were in particular events into that collection level. So you get a future probability of getting it when you get to that particular place in your leveling track. And of course, like that brings in a whole host of implic. That to me is the problem is that when you start to move away from clash royals design, you have to reverse engineer all these things that that system solved into your mechanic. This is one of the dangers of not taking an off the shelf solution and really I think focusing on where your innovation is. And to me, the innovation in this game was clear. They made it a card game that could be played in under three minutes with an incredible level of strategic depth. <cite>Eric:</cite> <time>16:45</time> I think it's really cool that they innovated on a monetization. To your point, it is risky. Like it's always riskier to innovate on the monetization than it is on the game design itself. I'm glad they did it as an observer, if I were in house, I don't know if I'd recommend that, but I think what you said about the loot boxes or the lack thereof is a hundred percent true. These, this, this is a group of X, right? Ex blizzard devs who like are very concerned about positive player monetization. They're always trying to stay ahead of the curve. When Hard Stone launched, it was seen as a very player friendly monetization. Today it's seen as very like extractive and loot box based, right? And that's just cuz of the times always change and what is friendly or not friendly is always,, like Battle Pass I think is going through this, right now where like they're starting to seem extractive and exploitive and everywhere. I think you're a hundred percent right there that they avoid. Specifically showing a chest open up and a bunch of lights come out right. They specifically avoided that cuz they didn't wanted to avoid the stigma of loot boxes. But the fact that the matter is random content distribution is a very effective way to do it. It's a very effective way to take a large pool of content and divvy it out in a way that feels better than the, a like heart. And yeah. So ultimately <cite>Phil:</cite> <time>17:51</time> don't, do you see that as a problem though? This is what I always go back to. If we just, find and replace any loop box conversation with the words card packs. It's a mechanic that's been around as long as Pokemon has been around, and it's never been a problem. It's never been a problem until it started entering some AAA video games and certain internet communities were outraged by that. that's the thing that I've always stumbled on is what's the difference between a loot box and a cart pack? And I have never had someone explain that to me because they don't seem to be able, Reddit, Twitter, the active, game communities, they don't seem to challenge these things. <cite>Eric:</cite> <time>18:26</time> I mean, what's the difference between amphetamines and coffee? It's like a lot of <cite>Phil:</cite> <time>18:30</time> Do you, do you think it's a degree? Do you think it's a, it's a matter of <cite>Chris:</cite> <time>18:32</time> No, I, I think it's different and this is gonna make me sound like a really, like huge crypto person, but when card packs were around, you could, there was a secondary market for them. So you like opened it up and then you could, you had the option to go and actually just buy the card you were looking for. So the card packs really are like, they're for a, in my opinion, they're for a particular style of games. So in Imagine the Gathering, they're designed for, they're designed for, draft and not constructed. If you wanna play constructed, you have this outside option, you have this other option. You don't have to go by.$5,000 worth of packs to get a$500 deck. You can just go to the secondary marketplace and buy that 500 deck because you're using the aggregation of all these cards within that community versus in a video game, you don't have that option to go, aggregate all these <cite>Phil:</cite> <time>19:20</time> that's where I think you're wrong, Dr. Smith And here's and here's why? FIFA Ultimate Team is the most successful game in many regards in the West? It goes out in every single year. FIFA will sell a bunch of copies and it is FIFA ultimate team. It's this segmented mode in FIFA that has more free to play like mechanics in it versus the mainline fifa. And so that has a secondary marketplace. That still gets some of the most criticism out of any single game product out there. And so I don't think the auction house is really, an eligible complaint. And I would add the, I would add the secondary issue is that, when we see games that do have loop boxes in them, they generally have some sort of pity timer, right? So Hearthstone has this, I think it's, 20 card packs without a legendary, you'll get one. Some games are even stronger in how they control probability. If you remember Overwatch if you open up enough loop boxes, you get crafting currency. That crafting currency could be used to directly purchase any item you wanted at various prices. And so even with those mechanics in it, it doesn't seem to have stopped any complaints from these communities. <cite>Chris:</cite> <time>20:19</time> I know FIFA has the online marketplace, for the players, I'm not as, I'm not super familiar with that, but is there any other, there's no card game that's ever done it except for like magic gathering online. But that's just like a, a. <cite>Phil:</cite> <time>20:31</time> tried artifact, tried outed the Web two world, that that was Val's failed game. <cite>Chris:</cite> <time>20:36</time> Oh, okay. Okay. <cite>Eric:</cite> <time>20:36</time> Rip artifact. <cite>Chris:</cite> <time>20:38</time> so you think it's purely cultural, you think it's purely like a or a psychological, <cite>Phil:</cite> <time>20:41</time> So this, this, this will be, this will be the agenda I will push on this podcast is that I, I think we have to be very careful in how we think about community. The word community to me is always used improperly in games. Really what we mean when we use the word community in the context we use it is we mean social media. That is a very particular slice of your audience that engages and I would say developers have grown more and more interested in engaging in social media. It's very vocal, it's very fast, it's all the things we talk about with social media and it's really affected how we start to make games. And I think we've started to make worse and worse choices to appeal to this particular. Rather than trying to judge the health of the game, and to me the health of the game is ultimately what generates the most value for the entire player base, not just a small section. <cite>Chris:</cite> <time>21:27</time> We'll probably we might talk about this in, later, but it's about the, it's about the marginal player, right? The vocal players are, the mo literally, we write this and in the paper with Ted, the most vocal players are the ones whose needs and wants are never met because they're always going to be, they're either always gonna be there or they're not gonna be there because they're on the very far end of the distribution. You care about the people in the middle. I guess when I say community, maybe I should come up with a new word. Player base and then community is like this loud, like fringes. And then player base is the chunk of the distribution that you actually care about. <cite>Eric:</cite> <time>21:58</time> back to the loop box point. So I think there's a few factors at play. One is the degree to which it's like a jackpot lottery style system versus horizontal. So if you look at, gas ponds in Japan, right? There's like capsule machines, right? You like put in a quarter, you turn the crank, you get a random toy. There's not like, Scarcity and oh, one of these toys is worth a hundred dollars and the rest of them are worthless, right? They're all just some cheap piece of plastic that's a little bit of fun. and because there isn't that jackpot winner loser situation, like that's less gamely. Whereas if you have one super valuable, ultra rare, ultra exclusive chase item that increases that sort of like gambling component. <cite>Phil:</cite> <time>22:35</time> So to use, is you, so to you, is it just a question of what are the rewards in the pool like that? If we could draw correlation between the rewards in the pool, the drop rates of those rewards, and let's say the community or. I would say social media pushback, would that be a testable hypothesis? <cite>Eric:</cite> <time>22:53</time> I think so. <cite>Phil:</cite> <time>22:53</time> So, Yeah. I'll <cite>Eric:</cite> <time>22:54</time> mean, <cite>Phil:</cite> <time>22:55</time> this. <cite>Eric:</cite> <time>22:55</time> I'm not saying it's the only factor. I'm just saying it's a big one though. So for example, when Magic the gathering introduced mythic, I'm sure this increased the loot boxiness of their cart backs. Before they only had, common, uncommon, rare, they added an extra tier of rarity, which are even rare, even more expensive on the open market. <cite>Phil:</cite> <time>23:11</time> My explanation for what I see when I see a lot of active gamer social media types engage in monetization is that this is, there's a public choice explanation for this. That's at least where my head space has been that this is an interest group. And really we should start to think about this, not necessarily from a strictly economics background, but a political science economics background, a public choice background, is that their interest groups. These interest groups have agendas and they're trying to promote their agenda. And one of the ways that they can promote their agenda is through low cost action in numbers, I would say a lot of things that have come out of modern social media to me have been engineered first by gamers. I don't know if we've ever gotten, I don't know if the gamer community has ever gotten credit for, I think, some of these tactics and for better or for worse it really depends on what the ends are. But I think a lot of them have been engineered in games. And so to me, I, I look towards public choice. I'd like to see a richer theory of this, a richer theory of community and what this is, because remember, we don't see this in mobile. We don't see a lot of, I think, pushback in mobile, or at least the pushback isn't as paid, as much attention to in mobile as it is in hd. That's been a pretty sharp line that I've seen throughout my career. <cite>Chris:</cite> <time>24:23</time> It's gotta be an information thing, right? so to me, I've always thought, I've always seen, free to play as basically a better, monetization strategy than premium models because it allows you to more perfectly price discriminate against your consumers. But that only works in a,, in an environment where there's not a lot of, information transfer between consumers. So if I'm playing Candy Crush on my phone, or maybe, GI and Impact or something like that, and, I don't tell other people about it. It's not a part of my personality the way that, playing, FIFA or playing, apex is I'm much more vocal with those people. They know exactly how much I'm paying. I know exactly how much they're paying. Whereas within this like little phone, These free to play environments, there's a lot less information transfer. maybe you're spending a lot, maybe you're not spending a lot. I don't know. Does that seem <cite>Eric:</cite> <time>25:06</time> I think that's right. If I have a shitty experience in a game on pc, you tab over to Reddit, make a big rant about it. If you're on mobile, you're like, ah, well that sucked. Like time to go back to brushing my teeth or whatever. <cite>Chris:</cite> <time>25:15</time> Yeah. <cite>Phil:</cite> <time>25:16</time> So you would think of it almost as a barrier to entry explanation. So basically the cost of complaining is cheaper on PC because you're on the same medium in which the negative experience is happening and what you'll post in versus mobile. Yeah. <cite>Chris:</cite> <time>25:31</time> I see it more as for example, you could think of a model where, the joy that somebody gets out of a game, has to do with the number of people that are playing that game with them. so there's this like connection, there's this warm glow from the. And basically, in these free to play mobile games, that the model doesn't include that, that warm glow effect. So there's no, there's no benefit of playing with my friends. Mainly because they're single player., that's kind of the first, that's the most obvious, empirical thing you could touch is the fact that, these free to play mobile games. There's no not that I know of. There's no like real multiplayer, versus, if you're going and you're playing, rocket League or something like that, you get together with your friends. To me it's more about that than necessarily like the, maybe like a cost or something <cite>Eric:</cite> <time>26:13</time> And on that point, the communities with the more vibrant social media, bandwagoning or whatever are the online multiplayer ones, right? people aren't freaking out over God of war nearly as much as they are over, rocket League or League of Legends or whatever. <cite>Phil:</cite> <time>26:25</time> But that's because the monetization doesn't exist there,. If you go back for instance, and you look at God, I remember when I was at Dice Ghost shipped, I think it was need for speed payback, which was the last need for speed to have loop boxes in it, predominantly a single player game. If you were to look at, let's say, share of outer spend in single player versus multiplayer, still a ton of push. loop boxes during that period. That was actually pre Battlefront too. I think a lot of people forget about that one. That one little snuck under the rug a little bit before Battlefront too. I think that's another interesting case of people pushing back. I'm sure you'd be soft, has done loop boxes and Assassins Creed. I'm not up to date on that franchise, I still co see to Erica, I would agree there's a positive correlation between, let's say community chatter about monetization and whether or not the game is single player in the size of the player base. That makes a lot of sense to me. <cite>Chris:</cite> <time>27:13</time> What I'm hearing basically is that, you need to consider who's on the intensive margin and who's on the extensive margin. Who are the people that you're actually talking about and who's the, who are the people actually playing the game like, No, the people who are playing Candy Crush are not the types of people who care about, I saw a statistic that the average, mobile gamer is actually a woman in her, middle age It makes a hell of a lot of sense. these people are, they're, they're not complaining about like$10. They don't really care, like to them. they would go out to, dinner or they would go do this. This is just one of those other things. Whereas this, a completely different sample. It's a completely different universe. There's two types here. Like studying the labor market,, you don't want to include unemployed people in your wage gap regressions, for example, or something because you're getting improper sampling. You have to be on the intensive margin studying those employed people. the other people are a different type. They're a different group. And that's why we see these dramatic differences in reaction because there's two different groups of people, <cite>Phil:</cite> <time>28:08</time> I think to your point, Chris, track the metric, track the metrics, track the size of the response relative to your player base. When you start to look at these Reddit communities or Steam or Twitter and you take them as the numerator and your player base is the denominator, it's a small relative share. That is a metric that analysts need to be delivering to game teams is helping you put all of this in context, cuz it is very scary to have a tab open on Chrome as all developers do. That is to the subreddit of your game. And a lot of angry players in all capital letters are saying nasty things. It's a scary place to be as developer. Article number two. <cite>Chris:</cite> <time>28:44</time> Let's do it. <cite>Eric:</cite> <time>28:45</time> sky Weavers relevant to loot box and card game discussion. It's a web three card game. The marketplace is built in, it's kind of in that magic online vein where all the cards you earn are supposed to be tradable. There are actually no purchased loot boxes. When you level up, you get a random new card, but, when you buy cards, you buy them from the open market. That's the primary place to buy them. so it adds a bunch of determinism there, avoids that, gambling aspect, which is, cool. You can tell this game is a backlash to some of the more like high volatility web three games. Ax Infinity is probably the poster child of this where presses went straight up and then straight down. and I'll talk about some of the systems that accomplished this, but yeah, other than that, it's a solid polished card game. so it's kinda like Hearthstone a lot more stuff going on. So like it's more involved, more like you have to, there's a lot more card text and sub abilities and that kind of stuff. But overall I've been having a good time with it. you can tell the game was well designed, made, these people know what they're doing. Let me go and compare and contrast their economies real quick. Axe was the previous big wave of, poster child for web three games or blowing up, and part of that was the price went super high, right? And one thing that caused Axi prices to go up so high, I think is how. Ax are obtained. So axs are the pets, but you can think of them like cards, to get to produce another axe in the ecosystem, you have to own an axe to axes. You have to breed them. breeding requires some sort of in game currency, and then they produce another axe. some random combination of their genetic traits, and then that axe you can sell and so the, if let's say like bird axes are very strong right now, right? People both want the bird axes to play because if you have them, you're more likely to win. If you're more likely to win, you're more likely to earn, right? So there's they essentially pay out dividends, but you also need the bird axes to breed more bird axes. So if bird AEs are strong, Crunch on two fronts. One is just the direct usage, the first order usage, and another crunch on the second order usage. If I want more bird AEs to breed more bird AEs to sell to the market. and so I think this magnifies the up and down trends of any price fluctuation, right? If we think this card is gonna be strong, then you know, it pays higher dividends from play, but also. The breeding value is much higher. And so I think this kind of accentuated all the up and down curves. <cite>Chris:</cite> <time>31:04</time> To me it seems like you're actually so I totally get this like over the market, over just completely overreacting to a, to a good card because of that, that double effect. but to me it's once you are, once that breeding process has started, should mitigate some of that? Because all of a sudden the market's flush with a bunch of these types of cards. Is that not the case or, <cite>Phil:</cite> <time>31:22</time> Equilibrium, right? Prices. Prices should law of one price in some sense <cite>Eric:</cite> <time>31:25</time> yeah, that's true. But it, it takes some time for it to reach that equilibrium. If you look at AY prices now, they're like rock bottom because people overinvested in breeding operations, six months ago and now there's just a giant surplus of axs. <cite>Phil:</cite> <time>31:37</time> I guess that would be a situation of people misaligning future expectations, like producing for future that doesn't exist. Do you buy more of a, a supply side explanation for axes though? If your reproduction rate for these axes is always above one you can always breed, if I can take two axes and I can make technically an infinite amount of axes, right? <cite>Eric:</cite> <time>31:57</time> Well, you have to use slp, so you have to pay to reproduce. There's there's a <cite>Chris:</cite> <time>32:01</time> there's also a limit to the number of times you can breathe them, right? I think so. <cite>Eric:</cite> <time>32:04</time> Mm-hmm. but there's methods where you take to, and you take all the cousins and breed them together, and you take all those cousins and you breed'em together. <cite>Phil:</cite> <time>32:10</time> the economy in Axe only had one macro lever to increase or decrease the aggregate amount of breeding, which would be the breeding currency. <cite>Eric:</cite> <time>32:19</time> Pretty much. Yeah. <cite>Phil:</cite> <time>32:20</time> What does that mean for Sky Weavers? The situation I imagine in Sky Weavers, the problem I start to see a lot of these crypto web three games is that they essentially have, an uncapped amount of supply in some sense, right? Because the assets aren't degrading. It's not like a T-shirt you wear a t-shirt a bunch, the capital naturally degrades. And we keep producing more t-shirts. There's some sort of equilibrium of t-shirts declining being thrown out, and new ones being born into existence that satisfies the popul. Feels like that isn't something that exists in any of these games, and that's a big problem. <cite>Eric:</cite> <time>32:51</time> Yeah. So Sky Wars actually has a very hard anchor here. All the cards that are produced, or most of them are through this mode that's called conquest. You can think of it like arena. If you played Hearthstone, you pay a buy-in and based on your win loss, you may earn cards. but effectively we can think of it as you pay a dollar 50 to open a random card, that's an expected value. That's what happens. And so what that means is that if the expected market value of these cards is above a dollar 50, It's profitable to play this mode, earn the cards, and sell them. So the expected value of the cards in the market cannot go above a dollar 50. <cite>Phil:</cite> <time>33:25</time> so there's almost like a price ceiling. <cite>Eric:</cite> <time>33:27</time> Mm-hmm. And also there is a price floor, because instead of paying a dollar 50 to enter this mode, you can also burn a card to enter this mode. So if any cards are cheaper than a dollar 50, you will buy them from the market, burn them to enter. And so it also puts a price floor. And so the variance on the range of these prices is gonna be very narrowly constrained around a dollar 50. <cite>Phil:</cite> <time>33:48</time> And so I can burn any card to get into the draft run. Interesting. So I imagine what happens is that you're trying to mine for the most expensive card and you're trying to burn the least expensive cards and so you'll burn through all the least ones. And so it's, if you're trying to solve, so this is the other challenge, right? This is the other challenge I've seen in web three is like, how do you set optimal supply? Or is a dollar 50 the right price ceiling is burning one card the right price floor? Is that a lever you're gonna be able to pull? Does Sky Weavers have an answer to that or is that you think that they just pick some numbers and they wanna see how things play out? <cite>Eric:</cite> <time>34:20</time> I think they picked a dollar 50, cuz I think Harton was a dollar. my guess is they can change that dollar 50 number and the whole market will shift with it. So they do have one thing, but at the end of the day, it's a bunch of extra steps to price each card at a dollar 50. <cite>Phil:</cite> <time>34:35</time> So you think what you would get out of this situation though, when you have web three and you have crypto, is really, you're just getting distribution effects, right? Who gets the card? Who wins the arena drafts? Who gets the money in the short run? <cite>Eric:</cite> <time>34:46</time> That's right. But the interesting thing is that the price variation on these cards is gonna be minimal. Like even less than what in magic or hear stone, Where. Like a place that of Tomago is like 150 bucks. Whereas in this game, like at most these cards cost maybe$4. <cite>Chris:</cite> <time>35:00</time> Yeah, <cite>Phil:</cite> <time>35:01</time> Do we know that in the marketplace? <cite>Eric:</cite> <time>35:02</time> yeah. Yeah. If you look at the marketplace, the most expense is probably like four 15, the less I checked. And the decks are all one ofs. So even if one card is super expensive, you're only buying it <cite>Chris:</cite> <time>35:12</time> so it's different for these rare cards. So I'm on the marketplace right now and for the common or for the silver? It, yeah, like four 20 is like the most expensive, but then I'm looking at gold. I'm assuming that's just like some sort of <cite>Eric:</cite> <time>35:23</time> four 20. <cite>Chris:</cite> <time>35:24</time> Jesus Christ. <cite>Eric:</cite> <time>35:25</time> Yeah. <cite>Chris:</cite> <time>35:26</time> You,, California and Europeans. I live in Indiana right now, so. <cite>Phil:</cite> <time>35:30</time> Oh, Stockholm is very conservative, my friend. You'd be surprised. it's one of the things that shocks people the most. That's for a different podcast. Stockholm, Sweden, just a very conservative place when it comes to drugs, when it comes to alcohol, when it comes to a lot of different things. Huge focus on the nuclear family. <cite>Chris:</cite> <time>35:45</time> Yeah. <cite>Phil:</cite> <time>35:45</time> podcast <cite>Chris:</cite> <time>35:47</time> Anyway, so these ones are literally, 10 x what the prices of the comments are. I'm kinda interested, is that basically just, a scarcity there? Like they're 10,. <cite>Eric:</cite> <time>35:56</time> It's scarcity. As far as I can tell. The advantage of the gold, gold border is just one cosmetic, you look cooler and two, through some of the earning channels, like you get bonus points if you have more gold cards in your deck than silver. So it's a dividend payout, like those stake and earn that crypto's been doing. But for general play, it actually has no difference. I think it's mostly a scarcity thing right now. And I think a lot of the prices propped up by people speculating that these will be valuable cuz they have a restricted supply. <cite>Chris:</cite> <time>36:26</time> I actually really like the model. I think it's really fascinating. It's a cool take on this like price control thing, which a lot of crypto companies try to do the price controls. I personally hate price controls. But, I could see'em if, if you're allowing this competitive destruction of cards. That's interesting. I think that's really interesting., so I'd be curious to take a look closer, but to me it seems like everybody gets access to every card or everybody has to buy, a collection for a hundred dollars or whatever it is. If you want to go the free to play route, everybody gets all their cards, or I guess it wouldn't be free to play. You have to pay. Like a dollar to enter the tournament. If you win, you get$2 back. If you lose, you lose your dollar. The company is making, every dollar on that. They're making all the dollars from the loss. It's just a tournament. Is this just a like more complicated way of running a tournament where there, there's, and it's not even, it's not even a, a zero sum game because the company makes money. <cite>Eric:</cite> <time>37:21</time> Yeah, I definitely think this is the way like play to earn will progress is that you don't distribute rewards to all your users. It's, it's just a tournament. They also have a leaderboard where the top a hundred or a thousand players get weekly rewards. And frankly that's been driving me to keep playing. even though it's the reward is like a dollar 50, I want to get to where I'm making a dollar 50 every week. <cite>Phil:</cite> <time>37:40</time> A super layer, not treating you well, arc <cite>Eric:</cite> <time>37:42</time> yeah, that's fine. You mean, but this is the thing about like game incentives, right? if I think of it as a dollar 50, this is a waste of my time. But if I think of it as a free silver card, I get way more excited, <cite>Phil:</cite> <time>37:51</time> That makes a lot of sense to me. But I still have questions about this overall model. It makes sense to me that you would have some way to restrict supply because if demand's gonna be what it is, you have to reach equilibrium somehow, unless you just want p price to go down to zero and supply always to move outward. But if you're, if you, if there's no difference in the card that you can burn to get into these arena runs, why wouldn't it be the case that over time what you would see is natural rarity inflation. So you're, you're always burning the least valuable card and you're being rewarded with a card that is more rare than the one you burnt. And so that to me would at least suggest that average power level or rarity, or however you wanna measure that is constantly going up, <cite>Eric:</cite> <time>38:30</time> Average rarity of cards owned. Yes, that's <cite>Phil:</cite> <time>38:33</time> or I don't think it's, I don't think it's fair. Yeah, I don't think it's fair to say power level, it's, it's more about which card is the most valuable. And I don't think that has to be correlated with any of those things. But I would guess that really this, the system does drive you towards some sort of equilibrium <cite>Eric:</cite> <time>38:47</time> To Chris' point this is a beautiful, elegant way to create price stability even more beautiful in the fact that it's all through one system,= they didn't have to have five different sources and syncs. It's just one this arena game mode. I question a little bit whether this was the right goal. Like in card games, like the prices do vary quite a lot, whereas in this game it's like a dollar to$4, right? That feels pretty small. frankly, like I feel like they could have achieved the same thing as just a web two game. Just charge a dollar per card, right? And and maybe this is the kind of the crypto DJ speaking, but I think a lot of the appeal of these market games is the market volatility is the speculation, I'm playing this resource game and how do I optimize my return? And this puts a huge damper. <cite>Chris:</cite> <time>39:27</time> I don't know. I, to me this is this is the promise of blockchain technology. I'm going on playing this card game. I've got my deck that's got, how many cars did they do? 30 30, 25 to 30 cards in my deck. And each one's worth$3. I've got a$75 deck. I literally can, very easily go and sell those cards on the marketplace with very little friction. Like I've got a bunch of junk yugi cards because I'm not gonna go through the hassle of selling those things. But I also don't wanna give'em away because they have some intrinsic value to me. I don't play ugi anymore, but if there was like a super easy way for me to liquidate my, cards and know that I could just get them right back if I wanted to. To me, that's a huge, that's that is the promise, right? <cite>Phil:</cite> <time>40:09</time> But see that's, that to me is not much of a promise. So all it means is that I can press the refund on the arcade machine button and I can get my 20, my quarter, my quarterback, <cite>Chris:</cite> <time>40:19</time> long there's somebody else on the other end of the transaction, I think it's cool. <cite>Eric:</cite> <time>40:23</time> when you were buying those UGI cards like your motivation was to play the game, you weren't thinking oh, and also I can resell these or not. I feel like that's always an after the fact consideration <cite>Phil:</cite> <time>40:33</time> to, to mom, moms and dads. Well, these are gonna be worth a lot of money one day. I mean, that, wasn't that the classic response to, spending copious amounts of Christmas money on <cite>Chris:</cite> <time>40:41</time> I, I came, I came from like a traditional Christian family, so all Pokemon cards were burned because they were evil <cite>Phil:</cite> <time>40:47</time> Oh <cite>Chris:</cite> <time>40:48</time> Yeah. <cite>Eric:</cite> <time>40:49</time> Wait, really? What's <cite>Phil:</cite> <time>40:50</time> What people about <cite>Chris:</cite> <time>40:51</time> I don't know because I was my, so my older brother had a Pokemon collection and I had a UGI collection. And if you don't know, UGI is way creepier than Pokemon. Like why the fuck where the Pokemon cards burned and the Ugi cards survived? And the sad thing is the UGI cards are worth a hell of a lot less than the Pokemon cards would've been. Anyway. <cite>Phil:</cite> <time>41:12</time> But here's what I, least argue for on the crypto side going back to our discussion, Chris, and you put in a quarter, you put in a quarter, you get a quarter back. Is that really that compelling of an experience? I always go back to the monetization angle and what I'm trying to think about is, are all of these market mechanics and that's me. What we're bringing to games, all these market mechanics, we have new challenges as game economists, as game economy designers, as monetization designers, whatever it may be. We have new challenges, which is that we not only have to manage. The actual assets we have to manage supply. Whereas beforehand, we are managing price and by managing supply, we are managing price. We're just one step removed from that. And so we have to face a lot of, I think, very difficult questions on how to set supply. And again, we were still facing those in the Web two world. It's just we are facing them via price, a more direct way. And so the question is is okay, then what is the optimal supply in web three? Just as the question was what is the optimal price in web two? And of course like that's a very complicated dynamic answer that I, I just don't think anyone's really started to research, right? What is the price elasticity of the market, right?, that's what we're trying to measure to find profitable game economies. <cite>Chris:</cite> <time>42:20</time> I, so this is something that, this is I sweat at night thinking about this problem working at Star Atlas because we've got all these ships, and it's I'm constantly worried about, There are too many ships for the number of players in the environment. like how many ships should a person have? Should they have three? They don't need more than, like this many. But also, what if they just wanna own a collection of a thousand ships? do, so I, that is like so prescient, because that is completely on the four, for me, it's one of the main concerns, in daily life is oh, we, right now, oh, we wanna raise some capital. Do we like drop some more ships? That's great. But at the same time, with every single edition of a ship in the ecosystem if there's not like the equivalent growth in the economy from the, to justify those ships, there's consequences. Yeah. <cite>Phil:</cite> <time>43:01</time> That to me is where marginal analysis really helps out, right? Is what is the value of an additional unit being released into the game economy? Does that increase total net aggregate present revenue? That was a whole, that was a whole mouthful, does it increase the value of your game economy? Your game economy? That to me is something that I think you can use marginal analysis to make a decent assessment. <cite>Chris:</cite> <time>43:20</time> I was just gonna say that,, that's why I love game economics is because that's a question you could ever answer in the real world, but you're the god of the economy. So you can look at every single thing and literally calculate that that metric <cite>Phil:</cite> <time>43:30</time> It would be interesting to have the Austrians on. There is a lot in Austrian literature that will make their boots quake, and I would love to talk to some Austrian economists on this podcast about what we're doing is building essentially planned economy. There's just no way not to build a century, the planned economy. In a digital video game, you are crafting a particular experience. When you craft an economy, you cannot avoid this. It is not, there are no natural things. There are no natural inputs, there are only artificial inputs you set. <cite>Eric:</cite> <time>43:57</time> Yeah. Yeah. I've always leaned towards the that's the decentralization angle of crypto never meshed with me cuz working on league, I was like, yeah, I'm the central planner. I tell you how much you earn on a win or a loss, but <cite>Chris:</cite> <time>44:08</time> The one thing they'll be happy about is that, in these environments we pretty much don't need like parametric estimation because like we can use non-parametric approaches. So that's the one, one good thing have for night. <cite>Phil:</cite> <time>44:22</time> Third article today, Chris, I believe it's one of your dissertation chapters from your newly dissertation <cite>Chris:</cite> <time>44:29</time> My defense was supposed to be 35 minutes, ended up going for an hour and 15 minutes. So, I'm apparently not skilled in making this quick. <cite>Phil:</cite> <time>44:35</time> Can we put your dissertation in the show notes? Can we link. <cite>Chris:</cite> <time>44:37</time> Oh, for sure. Yeah. No, no, no. I mean, it's just, it's just a PDF at this point. <cite>Phil:</cite> <time>44:41</time> Oh, I think, I think our, our legions of fans already would love to read your <cite>Chris:</cite> <time>44:46</time> Oh, they're gonna, they're gonna <cite>Phil:</cite> <time>44:47</time> The Vast Game, economist Community <cite>Chris:</cite> <time>44:49</time> So it's my second chapter is, it's a stark contrast to the other chapters, which focus on labor market discrimination using empirical data. The second chapter, however, is a purely theoretical chapter with Ted Casanova, who's kind of one of the we've already talked about, is the person who got me into this field. And it's one of the first papers, if not the only one, to to just try to show firm behavior competitive firm behavior in a, in a digital media firm, or a digital media market. So when I say digital media, technically the model would work for like a social media website, like Twitter or LinkedIn. But we, we obviously focus on video games cause that's much more interesting. And the monetization there is not just ads. so we walk through a couple of different models but it all builds off of this baseline model that says, the firm has to decide on some quality, on some price to set. And they're subject to this, constraint that the number of players, the attention of the players is a function of those two things. So as the price increases, players leave, as the quality increases, players come in. So it's classic optimization problem. You've got your production function, you've got your constraint. And so, equilibrium is where those two things touch a ency. The baseline model's actually pretty straightforward. Just says, there's some optimal quantity and there's some optimal price or fee for the game. If you're talking about a premium game and you know that players are willing to enough players are willing to come in such that profits are optimized or maximized subject to the constraint. So obviously, we get much more interesting. We talk about a play to earn model, a free to play. And, and a couple of other models, but the kind of primary ones I wanna talk about are the free to play and the play to earn. So the way we change for the free to play is we say, okay, there's two prices. There's the price from before, you can think of like God of war as a premium game, obviously. So you pay this, 60 bucks and you can play God of War and a free to play game. P here is ambiguous. It could be zero. Q, however, is a positive price and it's an optional price. And you can pay Q to get a different version of the game. So you can get a different and I'm saying Q and P you can pay this optional fee to get a better quality of the game. So if you pay the regular fee the p in this case, you're just gonna get. game version zero. If you pay, this optional fee, you're gonna get a really, better version of the game. You can think of it as any free to play model. You could think of this as dlc. P could be 60 and q could be, 60 bucks for a dlc. You're just paying for a better version. So in this model, what you're gonna find is that there's this there's this, summative price that is basically P plus Q, that's like the summative price that somebody would pay. And an equilibrium p in this case could be negative or zero, as long as Q is big enough because the summative price is P plus Q. so if Q is quite large, Could technically be negative as long as it's not negative, larger than Q. So basically we end up with this model where the summative price that the firm gets is positive. They don't care, but there's actually a zero price for the main price and the optional price is so big. You could think of a Wales and Minnows problem. You could think of an advertising fee. Advertisers are paying this queue to the firm and oodles of money. So there are a whole bunch of different ways to think about this. But so that, that's the free to play model. And when I saw this, p equal zero Ted had written in there it very subtly p could also be negative. And that's where I came in and start and I wrote the play to earn. Which basically takes this cue. It gets away from the two different fee structures, the like regular game and the really good game. And it just says now the price to play the game is strictly negative the price to play the game. You're always gonna make money when you play the game. And this, you can think of an NFT having some quality. You go into the game, you get the nft, you can sell it on a secondary marketplace. You know what that price is completely, we're not gonna try and talk about, we're not gonna try and estimate actual nominal prices, but the idea is that the firm instead of having to decide, what price and what quality to sell their game for now, decides how many shares of the of the game. You can think of these as NFTs. And I'm gonna assume some knowledge from, our, from our audience here. But the firm now is instead of setting a price, they're now setting a supply. They're issuing NFTs. So now they're in, they're intrinsically setting the price but they're more so influencing the price with the issuance of the, of the, of the of the asset in this case. So instead of, players paying a fee for a game, now they're getting paid for the assets in that game, and the firm makes money by selling those assets on the secondary market. They're gonna retain some of those for themselves. They're gonna sell them to make their money, and they're going to, the players are going to be trading back and forth. We abstract away from the players' market, the secondary market, and we just try to think about what is this? What does this fee look like if it's negative? So it's a crazy model. <cite>Eric:</cite> <time>49:28</time> So, yeah, so you're talking about how like rather than pricing on price, you're pricing on quantity, right? You're not saying this costs$10. You're saying we're gonna put 10,000 of these in circulation and, and so this might be like, bread and butter for economi literature. But, there's this thing, I remember Glen, we put me on a while back, which is if you compare this to, let's say a carbon emissions policy carbon taxes versus cap and trade, right? A carbon tax is a price based policy, which says we know how much damage a ton of carbon causes to the environment, so we're gonna price the tax accordingly. Whereas the cap and trade says, we actually don't know. And let's say there's some critical threshold where if too much carbon is emitted, the whole planet dies. We're like, okay, I don't know how to price this on a. But I do know if we mi you know, like emit too much carbon, we're fucked. So I will cap the quantity as opposed to setting a price. And depending on what your uncertainty is about the costs, right? Whether, if you're very certain that oh, every unit of carbon costs a dollar, and so we can price the tax there, then you know that the price based policy works better. And so going back to games, right? If you, if it's like a unit economics thing where oh, every shoe I make costs me$5 to make, so I know I should price them at$10 because, I gotta cover my costs. Using price makes a lot more sense. Whereas if you're uncertain about the value, which is often the case with digital goods quantity can sometimes make more sense. <cite>Phil:</cite> <time>50:45</time> but when I look at it, at the end of the day, what you observe is equilibrium price the supply and demand curve is not something that you really observe in a static piece in time, right? This always goes back to the epistemological problems of the economics model is, okay, I look at a supply and demand curve. What am I looking at? And usually what you're looking at is you're looking at a point in time. You're not looking at any time series variable, but when you can observe something, what you're observing is equilibrium price, which is supposed to be where that intersection of supply and demand are. So what I would argue back to you is you, to your point, you can use each of those variables to change equilibrium price, you can use cap and trade. You can change, you can change permits you can change the total amount of permits that people have, or you can change the price. But to your point, they achieve different things, right? Cause cap and trade gets you a fixed quantity and a pian tax does not get you a fixed qu, You can design apoian tax that approximates a fixed amount of supply. But you have to know a lot about elasticity to be able to use Theon tax to do that. Cap and trade will definitely get you a supply ceiling of carbon. <cite>Chris:</cite> <time>51:45</time> so the original model that I tried to write for the free to play or for the play to earn was keeping that p and q concept. So basically the player participates in some sort of lottery or some sort of tournament. Some players are good players and some players are bad players, or winners and losers and the winners. P and the losers pay Q. So in this case, p could potentially be negative if Q was high enough. You could, you can imagine a market where the majority of people are losers. Twitch is this way, right? Like streamers only the best of the best of the best streamers and eSports ever make it anywhere or any sport for, for that matter. A bunch of other players, a bunch of other people pay this huge cost. They this huge fee and they don't end up getting anywhere. I know plenty of people who went to play college sports ended up, it probably ended up costing them. The opportunity cost was probably by the time they're done with their life in the hundreds of thousands, if not millions. So I almost see it like that, where you have a, you have this like hyper skilled tournament that happens. the expected wage for the, for the winners was sufficiently high. <cite>Phil:</cite> <time>52:46</time> I just think about wage rates as playing a game. Like I, I derive a certain amount of utility from playing a game. That's my wage rate. And whether or not that wage rate is realized in terms of fiat currency to me, is completely secondary. It just changes the integer. So, I play Overwatch, I get$10 an hour in terms of utility. That's my, relative opportunity cost. If you hired me somewhere for$11 an hour, I would quit Overwatch and I would go do that other thing. I always think about, this implicit wage rate. When I just think about play and it changes that integer value for some, for some users. Do we get, do we get predictable? Do we get testable hypothesis out of the model you described, like what would be an example of some testable hypothesis that would fall out of the PQ firm <cite>Chris:</cite> <time>53:26</time> yeah. So, the testable. The testable there, that would be like going into look at a firm's Income statement, you would see, what's coming in, what's going out, what are people playing are paying zero what are the microtransactions queue? So the model there is just that, p is lower in the free to play model than it is in the premium model. One of the things for the free to play model is that people are less sensitive to a change in price than in the premium model, as you might imagine, right? So the attention of the players in equilibrium in the premium model are very sensitive to the change in price. P whereas in the free to play model they're not as sensitive to P or Q <cite>Phil:</cite> <time>54:02</time> What drives the relative in elasticity in the free to play environment versus the more elastic behavior in the premium fixed price game model. So we were talking about the fact that, when you're operating in this premium fixed price world players are fairly elastic. They respond strongly to price. Whereas when we think about, a free to play game, you suggested that one of the testable hypothesis that come out of this is that those players are far less price. Elastic and they don't change their quantity demanded in response to a price change very much. Sorry, was a very long winded elasticity thing, <cite>Chris:</cite> <time>54:37</time> So, so I mean, in, in terms of the the actual explanation it's that the median fee is lower, but the average fee is higher that the average player is paying a higher fee. And the, because, you're price discriminating when you do a free to play model. So you're increasing the willing, the average willingness to spend, but the medium willing to spend is actually gone down so that you're literally changing kind of the distribution of spending. <cite>Phil:</cite> <time>54:58</time> Don't, don't we need to assume in these worlds though, that we're holding all else constant. So shouldn't I imagine like the same distribution of, let's say, willingness to pay between a particular product like we could imagine Candy Crush as a fixed price game. And we can imagine Candy Crush as a free to play. Do we think that willingness to pay is exogenous from the game experience or that would change materially once you're in the experience? <cite>Chris:</cite> <time>55:20</time> I, I think it's really just about where we are. we were talking about equilibrium price before. It's where we are on the, on the demand curve, right? this was something that I, I remember one of my undergrad classes, the professor was is the demand curve more or less any elastic as prices as prices increase? And I was like, oh, the elasticity of demand is the same no matter where you are on the demand curve. And I got the question wrong because, elasticity depends on the price. It's a relative price elasticity. So I think, I think my, you I, I maybe need to think about this harder, but my initial response is that because the price is like the average price is higher, there's probably, less elasticity in, in the response because that price is higher. <cite>Phil:</cite> <time>55:59</time> Okay. So it's almost like a math trick though. It can be a math trick based on like, just the fact that the, the nuMe, excuse me, the min and max value have such large integer value ranges, right? <cite>Chris:</cite> <time>56:10</time> Yeah, it's, it's always a math, math trick, but yeah. <cite>Eric:</cite> <time>56:14</time> So maybe there's a way to model this, but what I think a lot of free to play games benefit from network effects, right? The more people that are playing, rocket League, the more fun the game is. So even if each individual user has their demand curve like that curve changes depending on the aggregate demand. <cite>Chris:</cite> <time>56:27</time> Oh <cite>Eric:</cite> <time>56:28</time> I assume you know that's like complex for this model. I probably just assume that it doesn't affect it. But how would you <cite>Phil:</cite> <time>56:34</time> how would you actually <cite>Eric:</cite> <time>56:34</time> Like, <cite>Chris:</cite> <time>56:35</time> when I said there are a couple of other models that I won't talk about, that's one of them. And I'm actually not like a hundred percent convinced on on our model. I like to think of this paper as like first attempt,at doing this and, basically a call to action for other people to say, start looking at this kind of stuff. You think about how many discrimination models are out there, statistical discrimination versus prejudice taste based discrimination. there's gonna be more models that come out. <cite>Phil:</cite> <time>56:55</time> There's nothing to explain games, right? Cause we don't see this behavior, it feels like in any other medium except games. We see this free to play purchase behavior. This is a model that is specific to games. It feels like there's few other services that are doing this. <cite>Chris:</cite> <time>57:07</time> And, and I argue that the space is becoming so impactful economically. we just saw FTX and I know that's the greater crypto like ecosystem, but, play to earn and like blockchain games have a big impact and I think that the, the financialization of in game economies will have, a larger impact. And it's gonna bring more people. It's gonna bring people on the extensive margin into the space, right? All these people who didn't play games before started playing free to play games. I know all these people who don't play games now are gonna start playing play to earn, it's growing the space. And it's also, just increasing the connectivity between, financial institutions and games. So I hope that with that, shift, we'll see more interest in this space. But who knows, economists are very academic, economists are very slow to move in terms of, following the times. <cite>Phil:</cite> <time>57:52</time> It's why we're here. It's why we're here at the game economist cast. But I think to your point, Eric, I think you can think about the network effects as just a positive externality, right? When you think about an individual player's utility curve, you would have an end variable in there that is monotonically increasing the size of the player base, right? And you would model that as a diminishing return, cause the millionth player isn't as valuable to your network as the first. But I'm very skeptical about the shape of that curve. <cite>Chris:</cite> <time>58:18</time> oh, yeah. It's a complicated problem in terms of our model the sensitivity of players to price and quality are greater in the network effects model because now when the quality goes down, somebody leaves, but now that decreases my utility because somebody left. So it's this, this compounding effect. So, sensitivity to price and quality are even higher in those, that <cite>Phil:</cite> <time>58:40</time> So that would that seems like another super interesting, predictable hypothesis coming out of this. So let's think about MMOs, we think about MMOs, like a classic networking case. There's a lot of very strong social features, certainly more than a game like Candy Crush. Would your model predict that those types of experiences guess what it would be, it would be potentially a slow ascent and a slow decline. If there are network effects. <cite>Chris:</cite> <time>59:03</time> With, with network effects, <cite>Phil:</cite> <time>59:07</time> so it depends on, first of all, it depends on what marketing spend is, right? A lot of this depends on marketing spend. Let's just assume marketing spend is constant. Would your model predict that MMOs would grow fast and decline slowly, or would they grow slowly and decline slowly or fast? Fast? <cite>Chris:</cite> <time>59:21</time> technically, With network effects those economies grow faster. And I'm trying to think of, trying to think of an example. I mean, the way that it works is let's say you've got a thousand people playing a game. If there's no network effects, when the quality goes down, 10 of them leave with their, when there are network effects, the quality goes down. And so are you asking about what is that cascading effect? Is it like a nuclear reaction where like 10 people leave, but those 10 people impact a 10 other people and those 10 people, is this degenerative equilibrium? I don't know. Certainly the model doesn't get that specific. But its an interesting, that's a super interesting question. and that's the key, like what is the difference in elasticity with, with no network effects versus network effects. And you could <cite>Phil:</cite> <time>1:00:03</time> think, <cite>Chris:</cite> <time>1:00:04</time> to study that. <cite>Phil:</cite> <time>1:00:05</time> and I think there's a lot of explanatory behavior for a model like that when we think about what's happening on the app store leaderboards. And again, I'm skeptical about the results that a lot of people tend to report, if you take it at face, The apps or the games that have been, let's say the top 100 grossing, are becoming increasingly, I don't guess, what did you say? There's a lot of staying power. They aren't moving. It's the same games every single year. It's becoming more and more frozen in stone. There's less new entrance into the top grossing charts. You might be able to argue that's a function of some sort of networking effect. That's just very hard to get people away from these networks in these games. <cite>Chris:</cite> <time>1:00:46</time> Interesting. So you're saying there's like a cost to leaving as well? Just there's a benefit of having a friend there. Cost of <cite>Phil:</cite> <time>1:00:51</time> Well, well, if n if n is really big and n is a part of your utility curve. And if n is really, really big, then if I leave, yeah, I'm gonna, I'm gonna leave that on the table. <cite>Eric:</cite> <time>1:01:00</time> And firm's ability to produce quality is also a factor, if you assume there's big heterogeneity in there or that quality is cumulative <cite>Chris:</cite> <time>1:01:07</time> yeah. <cite>Eric:</cite> <time>1:01:08</time> we definitely see this in live service games. They're way more winner takes all than titles. <cite>Phil:</cite> <time>1:01:13</time> And that's game economist, episode one in the can. So you guys next week.

Show Notes

Episode Transcript

Other Episodes

Episode

E09: Regressive & Progressive UGC Taxes

Episode

E17: Regressions, Gin Rummy, and a VERY Special Guest (w/David Nelson)

Episode

GEC BONUS EP: What's up at GDC 2023?