Game Economist Cast

E26: Price Theorists Battle Behavioralists for Marvel Snap

Phillip Black

Send us a text

Phil gets Eric pilled into a trio of roguelikes, where we prime the pump for Squad Buster's eventual shake-up of the genre. Chris ran another NFT sale and lived to tell about it. We Snap our fingers for Eric's take on Snap's doubling cube mechanic. Should all ranked systems use it? Should the bet limit expand to infinity? We conclude with a discussion of Eurovision's political economy.

https://ericguan.substack.com/
https://chriseconomics.substack.com/

Speaker 1:

The immigrants always have the best food.

Speaker 2:

It is so true. I actually recently went to this place called Capel, and it's the only place I've been able to find anywhere in the country, or even the region, that serves pupusas.

Speaker 1:

What is it?

Speaker 2:

Pupusa, you have pupusas. You live in fucking Los Angeles. You have pupusas. What the fuck is a pupusa? I've never heard of a pupusa.

Speaker 3:

You've never had a fucking pup. Dude, you're making that shit up.

Speaker 2:

Oh, my god so pupus is Salvadorian. Pupus is P-U-P-U-A-S. No, it's a piece of bread that has usually like meat or beans or cheese in it. It's Salvadorian and they put usually cabbage on it with a spicy sauce and you eat it all together and it always comes in this little bag.

Speaker 3:

Be honest, this does not seem like the kind of food that is essential to survival. He's the only place I can get pupusa.

Speaker 1:

He says that like it's the only place you get tacos maybe he lived next to a pupusa restaurant for five years, as I'm a shit.

Speaker 3:

I grew up eating pupusa.

Speaker 1:

I was a little kid raising me on pupusa. Let's start with utility.

Speaker 3:

I don't understand what it even means everybody has some kind of utils in their head that they're calibrating. There's hardly anything that hasn't been used for money.

Speaker 2:

In fact, there may be a fundamental problem in modeling that we don't want to model. So 27,. Here we are back in the saddle. I don't know another regular time, four weeks this time, I don't know, it's just very regular. How many weeks of break we take Loopbox every time. We have two wonderful topics to talk about today on Game Economist Cast, but before you even start that, I'm joined by my two wonderful co-hosts. There's Chris, chris, how are you?

Speaker 3:

Inspired by the conversation with who was our last?

Speaker 2:

guest Sam Rosen.

Speaker 3:

Dr Rosen, inspired by our conversation with who was our last guest, sam rosen. Dr rosen, sam. Inspired by our conversation with sam rosen about how to sell out a nft collection. We did our first nft pre-sale over the weekend and sold out of half a million in four hours so congratulations did you use?

Speaker 3:

his advice. I just, yes, I used like economic principles but yeah, but the good news is it's a pre-sale, so like we're gonna have more available, it's an asset in the game. So we're not like trying to scarcity limit these things defeats the purpose, but more of a pr thing. Hey, this looks really good. Hooray, trying to get that headline figure congratulations.

Speaker 2:

Is this, can you? I was thinking, do you roll up multiple strategies? Do you roll up like a small pre-sale and then a slightly larger one, a slightly larger one and build momentum?

Speaker 3:

yeah, so that was the hope for this one. We had a big presentation over the weekend and we wanted to follow it up with something for people to buy. So we did pre-sales for the nft that's gonna launch like I don't know, launching like a week or something like that. They're like characters that use in the game, but the thing is like you are managing a fleet of a million, like a bunch of ships sometimes, so you need like tons and tons of these guys. So some players will have like tens of thousands of these crew and there's like a big probability distribution that determines the rarity of the crew. So you on, you basically unwrap this pack of cards and you get five crew and there's some probability of them being super high rarity, like a card pack, a traditional card pack. So, yeah, it's a combination of not necessarily loot box mechanics but randomness, probability and, yeah, low price point, which helped a lot, I think lots of slots, that's great.

Speaker 3:

I mean love, loves lots when you're monetizing honestly, what was really interesting is like, typically in our pricing in the past, we have assets that are, like, exponentially more efficient at extracting resources, so you need to charge exponential prices. So, like, our primary prices go from $15 up to a million dollars and with these packs, because they were probabilistic, we were able to use more traditional pricing the base pack is $25 and the premium pack is $50, as opposed to $25 and $5,000 or something like that. So basically able to distribute that value more equally across the packs, to lower the average value of a pack and sell more of them. A lot of interesting stuff wrapped up into this.

Speaker 2:

And also we're joined by Eric. Eric, how are you?

Speaker 1:

Hey, I'm doing all right, eric, from where? From? Soon to be second dinner. So I'm joining the Marvel Snap team over at second dinner. Yeah, going back to traditional gaming for a bit. Yeah, if you have any hot takes on the game, let me know.

Speaker 2:

Do I ever Like? Where's drafting Like, where do I have an agenda for this product that we got to push? Are you playing a lot of Marvel Snap these days?

Speaker 1:

Yeah, I've been lately. I definitely churned out the first time when I realized I was only playing against bots, but fought through that this time.

Speaker 2:

I'm curious about that. They have bots. They're unmarked. I have an ethical problem with that, even if it moves. Retention Is there a retention-maximizing amount of bots? I guess it would just be whatever win rate you want, right, the retention-maximizing win rate.

Speaker 1:

Yeah, that's actually one of my first tasks when I go over there is figure out how to insert bots into matchmaking, how to optimize that. Do we need to crank up the difficulty of the bots?

Speaker 2:

Oh, no shit, no shit, that sounds really cool.

Speaker 1:

Yeah, but it's definitely an interesting lever and one of the interesting things, like we talked, like I wrote that article about the doubling cube, the all their whole rank system involves wagering and so the bots are actually a source of inflation of ranked cubes as well, which is a factor in.

Speaker 2:

Oh, it's a very interesting observation. I didn't even think about that.

Speaker 3:

What do you mean? Like rank inflation?

Speaker 1:

So in the game we can get into it in the doubling cube section, but earlier. Oh, fair enough.

Speaker 2:

No, that'll be super interesting to talk about. Wow, man, you're really like joining up your love of CCGs. Man, I couldn't think of a product that's like closer to your being than working on Marvel Snap. Are you going to the office?

Speaker 1:

Uh, yeah, I think so. It's gonna be like a 20, 30 minute drive from where I live, so I definitely requested a desk. It'll be nice to have a physical space again. Where are you going every day? Nah, probably like once a week. I'm still trying to drive in traffic every day.

Speaker 2:

Hey, we want that you have one of those auto-driving Teslas though.

Speaker 1:

Yeah, that's true.

Speaker 2:

This is a lot more podcasts.

Speaker 1:

Getting an hour podcast every day.

Speaker 2:

That sounds great, man, congratulations.

Speaker 1:

Yeah, thanks.

Speaker 2:

Good Life is wonderful. Over here we just had Eurovision, which, if Americans haven't heard of this Will Ferrell he called it Cirque du Soleil meets the Olympics, meets American Idol or something of the sort and it is that it's like all of these countries submit one song and then you vote for countries and half the vote is from judges, professional judges from each of the countries who do rank order ballots that allocate points, and then the other portion of the points is allocated by the audience and a popular app based to vote, and then you try to vote for who is the best country, and they try to be apolitical. How do you guys think that went? So there's all these voting blocks. There's actually a peer reviewed paper about all the voting blocks that appear. So, giving an example like all the Scandinavian countries vote for each other.

Speaker 2:

Finland will always vote for Sweden. So we an example like all the Scandinavian countries, vote for each other. Finland will always vote for Sweden. Sweden always votes for Finland. Norway usually gets in there, everyone. It's like high school cliques. Can countries vote for themselves? They cannot, and you as a voter can also not vote for your own country. They use IP address-based geolocation or you have to enter your country, your telephone number, excuse me.

Speaker 2:

You have to vote for other people.

Speaker 3:

So if you're a small country, if you're a large country, you're like a disadvantage because it's like one of the largest pools of voters is not able to vote on your country oh, so you mean like in terms of oh, that's a very interesting observation.

Speaker 3:

So what you're saying is, if you remove that country from like the eligible votes, yeah, if you're a tiny country, you're, like naturally advantaged because you have the largest voting pool because if it's just n minus like the eligible votes, yeah, if you're a tiny country, you're like naturally advantaged because you have the largest voting pool. Because if it's just N minus like the size of your country, if you think about whatever, if 2% of the voting pool is your country, then you have 98% of the voting pool, versus Germany, who might only have 80% of the voting pool.

Speaker 2:

But then we have to assume that, like any one country, is that big of a, that big of a percentage and your value, yeah, one country is that big of a, that big of a percentage and did your value?

Speaker 1:

yeah, isn't germany a pretty big percentage?

Speaker 2:

of the eu pretty good, like we had to do rank order 83 million. Well, here's the ironic part germany does the worst.

Speaker 3:

Germany fucking sucks like a lot of the big countries. That's my hypothesis, uh, that's my theory the.

Speaker 2:

What are the countries that do? Well? So, ireland and sweden. I've won the most.

Speaker 3:

They're both dynasties but those are small countries those are are small countries.

Speaker 2:

I'm trying to think the UK used to be good, but they haven't been good in 50 years. And you also got to remember there's time series effects. Population sizes are changing, so we should include that in our peer review paper.

Speaker 3:

I think I just figured out Europe.

Speaker 2:

Dude. It demands more economist attention and, honestly, political scientists too. Like rational choice people.

Speaker 3:

I just watch like the main, like the finalists or whatever on youtube. We just go through and we watch all the big ones and it's just we always like the fun ones, like we don't like the traditional ones. It's like that one guy from years ago that was like a vampire. I don't know. He was like from croatia or something like that, not croatia, what's the vampire place?

Speaker 2:

it's romania. I think it's romania. It was amazing. There was one guy who was like running a hamster wheel.

Speaker 1:

So it's like a combination music show and like live stage performance.

Speaker 2:

Yes, they spent a lot on the choreography. The choreography is like very purposeful, like sometimes they like to tell a story, other times they like to do weird camera shots. They have control of the camera, which is really interesting, so they can do a lot of really weird things when it comes to the unique camera angles, and they cut very quickly. So it feels like michael bay is directing this, like michael bay is just always off camera, yelling like more explosions and like they always give it to you on the show, eric, you should watch the movie eurovision.

Speaker 3:

It's a good one. I've seen clips from it, fire nice, crazy swedish accent I haven't watched the whole thing.

Speaker 1:

Oh, it's Icelandic.

Speaker 2:

I think it's Icelandic.

Speaker 1:

I've watched it twice.

Speaker 3:

Oh, it's good man, I liked it so much.

Speaker 2:

And you know what they tried to do this in the United States. They tried to do a 50 states one. Did that go well?

Speaker 3:

Yeah, Isn't that just American Idol?

Speaker 2:

Yeah pretty much, which sucks. We're going to be talking about Eric's new article. He is the winning party right now in a sub stack bet, I believe, between Chris and Eric. He is publishing regularly and this is his latest banging piece. It's actually really fun to talk about. We have a lot of really cool graphs in there calculating like expected win probability and when he should forfeit, which is really cool. And then I'd like to rant a little bit on transmedia and what we think the incremental value of transmedia is. It's something that I've been thinking about a lot, with the rise and the fallout numbers, which have been like through the roof. Before we begin, let's talk about what we've been playing.

Speaker 1:

Got a selection of good things on sale stranger.

Speaker 3:

Phil, you never go first. What are you playing?

Speaker 2:

So I'm actually I got Eric Pilled. I've been playing a lot of Soulslike, so I was playing Bealto, the roguelike poker game, which is an unbelievable amount of fun. It's addictive. I didn't realize how satisfying those loops were. I got Hades 2 in Early Access and I've also been playing Date, the Diver, and all of these games would be infinitely better if they were free to play. Mobile games that were monetized at higher Pooh values, just let it be known.

Speaker 1:

You can take.

Speaker 3:

Phil, out of Scopely if you can't take the Scopely out of Phil. Would they be more interesting for the player?

Speaker 2:

They can be harvested.

Speaker 2:

They can be harvested. Their loops can be transplanted. I just look at them and I'm like, wow, this is really good shit. Only if there was a monetization Cool, and there is. But I would say, like, their ability to create loops and create theory crafting is super interesting, and I've been playing a lot of squad busters as well. I'm trying to write up a super new launch later this month and they have a lot of roguelike elements, but they aren't clicking. And I think these games, their roguelike elements, do click. And so why do they click? What makes them click? To me, it's really about theory crafting and in-round progression. But yeah, that's what I've been up to. I've been converted and Bialto I will give a shout out to that. That deserves a free-to-play mobile launch, but it is also a shit ton of fun.

Speaker 2:

Poker roguelike, yes. So what happens is that the basic idea is that each suit or each hand scores a certain amount of points, but you can trigger multis on that so you can level up your multi, so you can say, okay, it might be 30 base points for a flush, but my multi when I get a flush is going to be two rather than one, and so, okay, now you can start to form a strategy. Some of the abilities let you replace cards so you can stuff your cards. You can remove spades and you could add hearts, which would give you a higher probability of having flushes. So that means that you can form, like on the fly, poker strategies, which is really freaking cool. Like it works really well and like some of the abilities are very auto chessy, it's like there's a shop in between rounds how do you spell it?

Speaker 3:

is it with an r or? With a bellatro b-a-l-r-t yeah, bellatro oh, bellatra, okay, that makes a lot more sense. I was seeing a board game and I was like it doesn't seem right.

Speaker 1:

Stefanfeld game yeah, it's really cool how they just like poker. Everyone knows poker right and like the idea of forming poker hands. But this idea of where you can manipulate the rng and change the payout of different hands and all these layer, all these mechanics on top of this simple base that everyone understands super cool. Anecdotally, I've heard a lot of people who aren't like traditional gamers, like play the spire didn't appeal to them. The whole damage and armor and health and stuff like for us it comes naturally. For them it's like a whole new system to learn. But they understand, understand poker hands like that and they're like oh, and the roguelike part of the loop really helps them quickly.

Speaker 2:

It almost exaggerates natural features of poker to a certain degree. That's what it does. Each run is exaggerating. Your strategy is exaggerating. Oh wow, I'm going to go with a flush strategy. I'm going to build everything around flushes and that's really freaking cool. I wish more games should do this. Find a retentive core, a classic retentive core, and reimagine chesscom. Like Chris, you play a bunch of chess. Like they should think about what is roguelike chess look like.

Speaker 3:

I don't want to say there is a roguelike chess game, that's at least in development. There's auto chess, there's a ton of new chess games coming out, and I know that chesscom is working with Supercell. These days they have a chess game, a Supercell chess game what's that? One called Clash Mini.

Speaker 1:

Chesscom is working with Supercell.

Speaker 2:

They did a crossover for Clash Royale and it actually drove a lot of installs.

Speaker 1:

There's chess pieces for Clash Royale. I forget what the crossover was like, I think they did brand marketing assets.

Speaker 2:

I. I forget what the cross-border was like. I don't even know what it looked like.

Speaker 3:

I think they did like brand. There was like brand marketing assets. Dude, I'm just trying to learn how to play the game. I don't have time for these other extra fucking game modes. They're trying to sell me on Chess 960, which is, I guess, a randomized chess board every time, which is suicide.

Speaker 2:

So yeah, like Roblox or Backbaby, I'm all for them. How about you guys?

Speaker 1:

I was playing a little untitled game with my daughter, so it's from like a couple years ago, but you play as a goose and you just steal things and harass people. Oh.

Speaker 2:

I actually remember this game, so you're just a little shit.

Speaker 1:

Yeah, exactly, there's this gardener and you like turn on the hose and steal his carrots and shit and try to steal his keys and you're just this obnoxious goose running around. She loves it. She like squeals and laughs every time she steals a carrot from the gardener. Great, mechanically it's a stealth game, right, it's like you're sneaking around, you're creating distractions and then, while their back is turned, you do something. But you know, instead of having that like dark edgy, like Batman, ninja, assassin's Creed, you know, vibe, it's like much more colorful and goofy. And yeah, I don't know if kid, I highly recommend it. Is it mobile, free to play?

Speaker 1:

I don't know if it's mobile, definitely not free to play it's co-op right yeah, there's a co-op mode it's really easy on co-op, but it's great for the kid, what's the win condition? What's up? You go through a series of levels. They're not framed as levels, but here's the gardener and you got to steal his keys to open the gate. And then you get to the town and there's a little boy and you gotta steal his lunch or whatever like so it's basically like linear quest based.

Speaker 2:

Is there any vertical progression? Does like my goose get more tacky. So it's there's. It's just interactive. Good old fanny or storytelling. I guess, what is that?

Speaker 1:

what Mario, what's?

Speaker 2:

that I don't even doesn't classify as modern gaming to me. There's no gamified mechanics, right. It's just puzzles, right. Yeah, what is a puzzle, quest-based puzzles.

Speaker 1:

Or Mario levels are just. You never upgrade your character right, you just go to the level.

Speaker 3:

I just hate geese way too much to play as a protagonist goose.

Speaker 1:

Don't you want to know what it feels like to be on the other side?

Speaker 3:

I want to play Duck Hunter, but with geese instead.

Speaker 2:

Those. I don't play duck hunter, but with geese instead. Those geese are all over the place pooping on everything.

Speaker 3:

You should talk to Canadians, man.

Speaker 2:

They're not gonna show any mercy because, like this or hate them, I'll hate them. They're great, okay.

Speaker 3:

Oh, that's what I thought. For some reason I letter Kenny made me think that Canadians like geese the hate geese.

Speaker 1:

Actually I think it's the first. A shared hatred, yeah, but all of them, you sure the country people geese are obnoxious. They're super aggressive and angry.

Speaker 3:

Oh yeah, they're fucking horrible. They're demon qualities. Almost attacked me the other day.

Speaker 1:

I was outside did it spread its wings? Yeah, exactly, you can do that in the game and, like you can intimidate people by spreading your wings, I won't be playing that game chris, what are you going to be playing?

Speaker 2:

what have you been playing?

Speaker 3:

I haven't been playing a ton of stuff. I've been playing a lot of chess. Of course, I actually stopped playing chess because I started tilting. So I played three or four games Tilting.

Speaker 2:

In chess.

Speaker 3:

Yes, are you getting mad?

Speaker 2:

What do you do? You throw a rock at the wall.

Speaker 3:

I have a philosophy that I stopped abiding by and it's never, ever resigned, especially at my rating level, because you can have people blunder like they could be up like 10 points of material and they'll blunder a loss, and I've won actually more than two games in scenarios where I was completely losing.

Speaker 3:

But anyway, I started tilting and so I would lose two or three games in a row and then I would start resigning because I was like I need a win, so I'm going to resign so I can get to the next game quicker, and I think I dropped like 40 points of rating in the span of an hour. So I stopped and then started playing slower games. So now I play hour-long games, which means I play less because I don't know Like it takes a. You got to find an hour to sit down and play a chess game, started playing. I finished with vampire survivors, got bored of that. The lights and the action and the, the colors and the smells got old after a while and starts in in typical like playing a game 10 years after it's over.

Speaker 2:

I started playing overwatch recently oh no, you did not dude. Let's front, let's up. I've been going back to overwatch too. I've been going back to Overwatch 2. I've been going back to it this week. Yeah, eric actually messaged me and tried to shame me. His wife worked at fucking Blizzard and he's shaming me for playing the game. What did you say? You asked me if I was like back on the pipe or something. My god, the game's trash. I'm sure it's going downhill. From a kpi perspective I can tell you.

Speaker 3:

So here's my problem. Here's my ftu e I I played through the, I played through the like first tutorials or whatever, and I actually enjoyed it quite a bit and I was like, okay, this is fun and I use their recommended, like they're the player that agent 76, I think, or soldier 76 or whatever, just like the most boring guy in the world who just like shooting a gun hey, and he took that personally it was a good first character.

Speaker 3:

I really liked him. But after I finished that first tutorial section, I got on to try and like actually play a match, and so one of two things happened. The first thing is I'm looking at each of the different types what ammo it's tank, attack and and support and I see an eight minute timer to play attack and I'm like I don't want to wait eight minutes to play attack, so I'll go with tank. Oh, I see a five minute timer for tank. I don't want to fucking wait five minutes to play a game. And then I see less than one minute for support. So I'm going to play support because I want to play a game. I don't want to sit around for eight minutes waiting for my fucking match to start.

Speaker 3:

And support is not easy, especially when you're first starting in any game. I've never, ever played a game where support wasn't the hardest to play types of characters. So it's the most difficult to play character, but it's the easiest way to get into a game. And I have a screenshot Zero kills on my team and the other team had 50, 60 kills. I don't even know what the. It was an absolute slaughter. It was just like walking into a fucking ray of sun or like a supernova. It was a horrible.

Speaker 2:

What happens when?

Speaker 3:

you walk into the sun, you get fucking obliterated, you get blasted, yeah, so I don't really know.

Speaker 1:

It's like where you list.

Speaker 3:

Actually, it is getting warmer out, but no but I watch. Fallout and it's like getting like getting you know nuked anyway. So I you know nuked anyway. So I wasn't super happy with that. I actually haven't played in a few days, but all my buddies are playing it. Uh, so I'll probably jump back into it, but it was a terrible ftu. If I hadn't, if I didn't have friends who played it, I probably wouldn't. I'd probably just don't download it.

Speaker 1:

I think your description actually really hits in a point, poignant point, which is that the game is super fun. The characters are really colorful and it just feels good to play all the sound effects animations and everything.

Speaker 1:

But a lot of the structure they built around the game really ruins the experience they tried to. If they could have kept like a lighthearted, like Team Fortress 2 kind of vibe where people are just hopping in and out of servers, play whatever you want, people are being goofy and it doesn't matter, I think the game would be doing a lot better. But I think the game would be doing a lot better. But they put too much of that competitive structure on. And as soon as you put that competitive structure on, like you said, everyone wants to play the damage roles. But the optimal team comp doesn't involve damage and if you put too much stick on winning, everyone starts arguing over hey, someone needs to play tank, someone needs to play support. And then they add this queue system where you have to wait eight minutes to play if you want to play as Soldier 76, and then the whole thing gets a lot shittier and the skill-based matchmaking throwing you, just totally throwing you under the bus in that first game but hold on here.

Speaker 2:

I think this is an interesting debate. The I did we need, don't we need a change explanation, like we could get a level explanation, like that might be a level explanation, but ultimately when this product launched it was a monster success, like Overwatch 1 was a monster success. They spawned Overwatch League, which is a complete flaming disaster right now, but like it did do really well and it's declined and it had all those things at launch. So that doesn't give us the decline explanation, though.

Speaker 1:

I think they leaned too hard into esports. So first of all, the game probably would have to gradually decline anyway, but I think they leaned way too hard on the competitive side to the esports, to the ranking systems, right, and that's what why is that a bad choice? Okay, it creates the game itself. Isn't that greater than eSports? Let's be honest, everyone wants to play damage, no one wants to play support, but strategically, the game necessitates that you play support, and so you've got any difference, you have to huh.

Speaker 2:

Was this any different in tf2? No one to play.

Speaker 1:

Medic right? Yes, but they didn't.

Speaker 2:

I remember all the you remember, like the voice command Medic.

Speaker 1:

Yeah, everyone wants to play Sniper and like Gout and nobody wants to play like Medic and the Fundies characters. But the thing is, tf2 doesn't push you into the esports aspect, right? There aren't all these rank points. You're keeping track of waving their dick about how they're diamond and everyone else is bronzed, right? They don't have the Overwatch League where they're highlighting oh, here's what competitive play looks like and here's what you should aspire to be. And so when the game, when your moment to moment gameplay, you want to play with a particular play style, but to win and win the metagame of I want to raise my rank points you have to play a way that's less fun, then the game is making itself less fun by emphasizing esports so does this have to do with the structure?

Speaker 3:

I'm assuming the reason that I have to wait forever is that there's always two tank, two damage and two. The teams are five or six, five it used to be so.

Speaker 1:

It used to be two of eight, and, yeah, there's always damage for long. Yeah, now they did two, two, one, and that's why the tank queue is long is because there's only one tank and and so all the people who want to play tank have to wait for that one.

Speaker 3:

Is it? That's causing the problem, couldn't they have? Hey, if you just want to be a bunch of damage, like we'll put, throw a team of five damage together, a bunch of random shit you can do open. You can do open, yeah.

Speaker 2:

You can opt into one of two matchmaking pool, which will ensure that it has a certain team comp, and another one is free floating where you can. Okay, so the free floating.

Speaker 3:

I don't think I've unlocked yet, and that's maybe something that one could think about. That should be the first thing that I unlock. I shouldn't have to wait to unlock because I'm sitting here thinking, oh shit, I don't wait eight minutes to play the character I want to play. And I'm not against playing support, but I'm like it's going to take me a while before I'm ready to play a support character.

Speaker 2:

I think what you said, eric, is very interesting. Let me push back a little bit. It is a MOBA. It is a first person shooter MOBA. That to me seems naturally competitive. Moba is already competitive. First person shooters are already competitive. One plus one at least equals two, maybe equals three in terms of competitiveness. Instead, what I might offer is like the problem is that there's no real progression, like there was like weekly, cosmetic based progression, and that was it. Like there was that you can't customize your weapon. There's nothing gameplay affecting that like you really you're earning. They had characters at one point and and that's what I've struggled with in this the franchise and they also just didn't ship enough widgets.

Speaker 2:

I think there's a supply side story here, but I don't know. It's too competitive. Tf2 is a server-based game with, like what 32 players, 64 players, I don't like. Do you really have drop-in, drop-out with Overwatch?

Speaker 1:

Yeah, why not?

Speaker 2:

If you're running a team comp, you want strategy. And to have strategy, you want consistency, right.

Speaker 3:

I want consistent communication.

Speaker 1:

I just throw up the communication window. Play who you want to play.

Speaker 2:

Would you increase the size of a match? Cavit six if someone leaves, it opens up a slot and quick play for someone else to jump in play wherever you want. Why don't you just do ugc and see what's successful in mode design? That's also an option.

Speaker 1:

I. Ugc is a whole can of worms, though you got to build the tools for that and yeah, that's fair, okay, interesting.

Speaker 2:

Sorry, chris, I've been playing a lot of overwatch as well and I've been trying to think about just this game, that just it just went off a cliff I, I suck, I am terrible I'll hop on with you guys.

Speaker 3:

Yeah, what I was just saying is I need to get better before I play with you guys, because I'm terrible he said he played two, two games, one with soldier 76 and I played. I probably played like 20 matches but I still haven but I still haven't gotten past the fucking intro or whatever. I think.

Speaker 1:

The game itself looks really good. All the characters are super charismatic.

Speaker 3:

It's really a pleasurable experience, Like shooting and hitting enemies. The sound effects are really nice.

Speaker 2:

The core mechanics of each of the characters is just so refined and feels like there's something to master. It's truly inspirational. To be honest with you, the core design is so good.

Speaker 3:

I don't know why it needs the progression. There's no progression in Fortnite. Why does Fortnite work Like they're a?

Speaker 2:

monetization. It's kind of Battle Pass. At least, To be fair, they did add a Battle Pass to. Is that all cosmetic progression, though? Yes, that is also true, that is also true, but there's no roguelike elements with inside of a match for Overwatch either. There's no progression whatsoever except your ultimate ability changes, and that is weird to me. It's almost very arena shooter. Although there's no pickup items either. There's pickup consumables, but that's just health. There's nothing else.

Speaker 1:

I feel like Phil's always trying to push vertical progression. On games, is there a single game you think should not?

Speaker 2:

have vertical progression. I would say some CCGs could not, it'd be too difficult. There was a Marvel Snap talk that I posted that was from the director of product about the fact that they basically had to put together a meta in six months with the trade-offs they made. And he said they investigated that and I can see why that's unscalable.

Speaker 1:

Well, you don't like more cards?

Speaker 2:

True, but that's unscalable. Well, you don't like more cards? True, but that's horizontal, true. So, yeah, there's some games that that shouldn't have vertical progression. It's just I just, it's just seems so clear to me in the data, like when you add these things, like numbers go up, conversion up. That's not why suicide squad bombed. I don't know if you followed any of that. Okay, we won't go into that.

Speaker 1:

But they had some loot controversy, but they added power score and shit yeah, following my thesis of they should have made overwatch less competitive focus, I do think vertical progression worked better in less competitive games anyway, that's the.

Speaker 3:

It's one of the first questions I asked my buddies when I was playing is okay, can I buy like a better character or is this all just unlocked? They were like it's all cosmetic, because the tough part is you're inevitably going to get into the play to win or, yeah, pay to win thing, which like traditional gamers fucking hate, and I know like a bunch of projects that are trying to monetize with vertical progression, but it's basically impossible without getting that hatred and, as we saw with hell divers, like review bombing is it's warfare let me just to illustrate just for a moment because I'm writing the piece on squad busters the difference between hd gamers and mobile.

Speaker 2:

So right now, squad buster is about to go live. This is supercell's new game. It's going to Just to illustrate just for a moment because I'm writing the piece on Squadbusters the difference between HD gamers and mobile. So right now, squadbusters is about to go live. This is Supercell's new game. It's going to go live at the end of May, and so they have implemented a consumable monetization system.

Speaker 2:

So what you do is you spend real world money and you buy gold. Gold has a variety of purposes, but one of the things it does is it buys keys. So when you go into a match, you unlock chests. Chest contain abilities, and so what can happen is that you can activate one of these keys a cooldown schedule and when you activate the key, the chest immediately transforms into one of the rare chests that the key corresponds to, and you don't have to pay the gold price for the chest in round, so you progress faster in round. Gold price for the chest in round, so you progress faster in round, so you're more likely to win. So the advantage is immediate and there's no reason not to like spending them, like always spending them will always make you better off. There's never a chance when spending a key makes you worse off. It always makes you better off, and this is about to go live in a product to millions of people, and that's how close monetization is to payment. Is these keys, if that makes any sense.

Speaker 3:

And keys. Are they a cold card currency or can they be purchased? The?

Speaker 2:

keys are consumable. So when you go into the round and you're looking at a chest, you open up your battle bag, you select one of the keys. When the key activates, the chest instantly becomes free and transforms to that chest type associated with the key. And if you don't get the basic chest Correct, you would go through the normal chest cycle of rarities. This lets you get a special one that's even more rare.

Speaker 1:

But it's all within round.

Speaker 2:

It's all within round.

Speaker 1:

And what does, let's say, doing better in the round get.

Speaker 2:

So it preserves your win streak when you finish in the top five. An extended win streak generates more additional chances to level up a chest, which gives you character shards and character shards. Ultimately, let you evolve characters, and evolving characters gives them stat boosts and new abilities.

Speaker 3:

So your thesis is like mobile gamers are much more tolerant to pay for progression. Like almost every Supercell game I played has pay for progression.

Speaker 2:

Just radically different social norms.

Speaker 3:

Radically different social norms, radically different organizing tactics, radically'd be met with a lot of negativity versus if you were to just rebuild Overwatch in a mobile setting, then you, like I, don't think League. Does League have pay for progression, Eric?

Speaker 1:

No, they used to have a little bit. They pretend like it didn't exist.

Speaker 2:

I remember the Apex Apex monetization designer. He did a lot of League pay to progress designs.

Speaker 3:

And then you've got Brawl Stars, which is like a MOBA, and it does have pay for progression, so it's just like a mini version of the HD game, and then you slap monetization on it.

Speaker 1:

I think Phil's totally right. It's Western HD. Gamers hate pay for power, but elsewhere it's a lot more palatable.

Speaker 2:

It's marvelous to watch. It's truly interesting. It's a case study that communications professors have never investigated this and it's really bothered me because I think, in all seriousness, gamers are the forerunners of a lot of internet organizational tactics. So review bombing on Steam. That's a new weapon that, like online internet masses have added to their arsenal about how do they curtail online public opinion. I'm actually serious here. Guys like I. I think they're like the early r gamer organization, the reddit, the reddit hive mind like they were able to drive a lot of early cancel culture tactics. If you remember the battle over the fanboys and the different platform holders, they were able to harass people on the internet very quickly. Like ratioing, like that idea of bombing. Something which is like overwhelming negative sentiment is a very internet gaming driven phenomena that, like other communities adopted Sorry, that was random, but that's something I really want to see someone do is like how they came with these organizing tactics. Okay, should we talk about our topics? Doubling cube, this is us. We're doubling. Chris, do you want to double with me?

Speaker 1:

Yeah, Marvel Snap. Let's talk about the doubling cube.

Speaker 2:

Hold on, we should intro you. You published a sub stack piece which you're going to link to in the show notes. Yeah, absolutely To Eric and Chris's sub stack, which will be in the show notes.

Speaker 1:

Yeah, so Marvel Snap has this doubling cube. It's the map, as they call it in the game. The best way to explain this actually is it's just poker wagers. So when you go into a marble stat match, you always have to wager one, let's call it a chip one chip to play and the winner of the round takes all their opponent's chip. At any point in the game you can map to double your wager to two chip and you can raise, and then your opponent can either choose to call to match your wager, and so they also have to put in a second chip, or they can forfeit. You take their one chip and they lose immediately. Now the opponent can also double you and then at the end of the game it automatically doubles the wager on the very last round. But yeah, it's poker, right, you have a wager, you can raise and try to encourage your opponent to forfeit.

Speaker 1:

If you think you have a strong and naturally this adds all sorts of poker, like dynamics, right, where there's, where there's betting, there's bluffing, there's assessing the game state Like oh, I have a good hand, I should raise. Oh, I have a weak hand, I shouldn't raise. I have a really bad hand, I should forfeit and give them the pot. But, importantly, it simplifies the decision a lot Like in poker. There's a continuous number of chips. You can bet right, and so you right, and so I want to bet five or seven or fourteen, like decision. Honestly complicated, more complicated than it needs to be to create these emergent gameplay effects. And marvel snap just simplifies, it's just double. Do you want to double or not? And any poker player will tell you that. Not the quantity of chips that matters, it's the ratio, right, it's the pot. Odd, like when you don't say I'm going to raise five dollars, you say should I double the pot or not? Um, and effectively what doubling does is saying hey, I think I have higher than 50 chance, I think my hand, my situation is better than yours and I would rather. I want to win more. I want to double the stakes.

Speaker 1:

And on the receiver's end, their decision is essentially how bad is my hand? Is there a chance? There's a chance that I win, like a pretty good chance. Then, okay, maybe I pay up because I've already got some money in the pot. But if, oh, there's absolutely no way I'm going to win, I'm just going to forfeit and shove my chips, give them the pot, and this creates a lot of assessing the board state. As you're playing, you're not just thinking how do I win, but also what are my chances winning right now. Am I doing well, am I doing poorly? How poorly am I doing? Which adds a whole new layer to the game. Marvel Snap on base is a very simple card game. You just put numbers on piles and try to make your number bigger than theirs, but this adds a really nice bluffing and betting mechanic on top of that.

Speaker 2:

Could you go over why the decision to forfeit or the decision to double is not mirrored when you cross the plane of zero win probability, like why is it when the expected value is negative that it's not the mirrored distribution or the mirrored outcome when the expected value is positive, where I should always double, yeah.

Speaker 1:

So let's say you and I are playing and we both put $1 in, You've got a good end, right? I think my odds of winning are 60%. So Phil doubles his bet right. So now on my hand I have-.

Speaker 2:

Higher expected value for me.

Speaker 1:

Awesome, exactly, yeah. If Phil has a 60% chance of winning, if he thinks he's more likely to win, he'd rather Advance my position, winning be bigger, yep. On my end, I have two options. I can forfeit, I lose that dollar, or I can match the wager and if my chances of winning, let's say, are like 49%, like I probably should just play because it's pretty close to break even, if I have a terrible hand, I'm almost certainly going to lose. I don't want to lose that extra dollar, right, I'd rather just hold onto that extra dollar. And there's a break even point at 20% chance where, if my odds of winning are less than 25%, I should just forfeit. Right, I just forfeit the $1. It's already lost, don't worry about it. But don't put more money into the pot that I'm likely to lose. Whereas if I'm above the 25% chance hey, I do have a. There's already a big pot in there I could put money in at the odd chance of winning.

Speaker 2:

So does this result, independent of like how big the pot is?

Speaker 1:

It depends on the ratio of the pot to the raise. As poker players, we'll call this pot odds. The pot's already there. You're not getting that money back unless you win, and so the rager is just your marginal cost to continue to play as opposed and the probability win is going to change, or I guess that indifference point is going.

Speaker 3:

Not the indifference point, but the switching point is going to change depending on the ratio.

Speaker 1:

Exactly, yeah, you can imagine if Phil went all in and put like a million dollars in the pot right Then the $2 in there is basically irrelevant.

Speaker 2:

So I'm just thinking of what would be the Nash equilibrium here, because we're saying that this is what you do as a player, this is how you should play, and I couldn't gain anything by pressing my odds even more, though that wouldn't change the situation Like knowing what you're going to do, like knowing your forfeiting schedule or your doubling schedule. Me knowing that, as a player, does not make me better off in any way.

Speaker 1:

No, not in the perfect information case. When the hidden information case, where you might have hidden information, you might think that, hey, I have a strong state, but my opponent might have a powerful card in hand and they might be purposefully holding it back.

Speaker 2:

So here's what I'd love to know Should we just put this doubling cube in every single game that has ranked?

Speaker 1:

I think there's a whole bunch of positive effects created and I do think more games should use this. I do think there are some caveats, though. First is that it works best in a turn based game. I imagine it'd be weird to like pause a real time game, like you're playing like Street Fighter, and all of a sudden you pause to double the wager, right. That kind of messes with the flow.

Speaker 1:

And in some games where poker players value this skill of assessing risk, assessing your willingness to go in or not, in some games that you might not care about, that right In Tetris or something right you might just say that's not a skill we care about in this game. This game is all about solving kind of puzzle, so it works best in turn-based games and I think it works really great in strategy games. Like I mentioned, assessing the board state is a big part of this. You've got these cards on the board, these cards in their hand, and what do I think my odds of winning here are In a game you can imagine a game like Rocket League, where the board state is just the team's point, and so that assessing the board state component isn't as interesting or dynamic of a decision, whereas in a strategy game like League of Legends, where over the course of the game there's all sorts of resources accruing to either side and map control at all. A very complex board state those ones.

Speaker 3:

I think it worked much better I'm wondering how you would apply this to chess. There are definitely scenarios where you're in a losing position.

Speaker 2:

Sorry, it's almost the easiest to apply this chess I think, it worked great yeah yeah that's what I'm saying, like it would be a really fascinating application.

Speaker 3:

And and then it makes those swings where, let's say, you're in a losing position and then you all of a sudden secure a win. At that point you probably have gained many more points in ELO than you would have if you had, even if you had won early on or something like that, because one of the things I hate about chess is that if you play a really good game, you get the same rating boost as if you play a shitty game against a, if you play poorly or something like this, which, like that's how ELO works. It's not really a problem with chess, it's probably necessarily a problem, but it's some feature of ELO and this seems like it could potentially add to that.

Speaker 1:

Yeah, I think in games like chess, like you're effectively wagering ELO when you play Right and it can feel homogenous because every game your ELO wager is basically the same, and one nice thing about this creates more highs and lows in that regard. But on the flip side, it is a little tricky to integrate into existing games with existing skill rating systems. Marvel Snap was designed from the ground up to use these wagers as their ranking system. How would you reformulate ELO to allow people to double their wager? Is this a separate currency? Is their skill rating a separate? This a separate currency? Is their skill rating a separate rating, that like rubber bands to their ranked rating, or is this literally what they're being match made off of?

Speaker 1:

So there's a bunch of questions about how you switch from an elo system to a wagering like chip system. But with that aside, I do think chess as a game this works super great in and you mentioned forfeiting earlier right, there's this cultural thing where, like you said, oh, maybe I can play back from this position, maybe I can win. Right, if your goal is to maximize your elo, expected elo outcome, you're gonna stay for the entire game and you're gonna grind out that losing position even though it feels bad. Right, you like you want to leave because you're getting shit on, but you feel like you need to play it out. This gives a strategic out where, like you can assess the situation and say, oh, my opponent raised, I'm probably going to lose this. The right play is to forfeit, and then you can feel good about your correct assessment of the situation and your strategic retreat, as opposed to feeling, oh, I gave up because I don't have the resilience and mental willpower to play out the game.

Speaker 1:

It helps games conclude quicker.

Speaker 3:

Yeah, I think that's a big benefit, because marvel snap is always about speed and I was trying to get the average average turns down yeah, and in chess I'd rather a game that I know I've either lost or won finish in 15 minutes and I get fewer elo points than if I had taken it to its conclusion. But you'll have in games that'll go on for 20 moves and it's jesus christ. I just wish this was over. So I'd be to sacrifice a few ELA points in order to finish the game earlier.

Speaker 2:

Do you have any idea or any thoughts about how many stakes should be in the doubling cube? So right now you can do a maximum of eight. So I think it's important to clarify that the cube rewards points that increase your ranked level, which does have a reward track, but it's mostly about competitiveness. You can do eight, so why not 16? Should they increase the variance even more? Is there anything that should stop them?

Speaker 1:

So in early versions they actually tried that, and you can see screenshots from Ben Brode's talk, where it's 512. Oh my God, are you serious Doubling cubes? Yeah, there's no limit on doubling. You just keep doubling every turn.

Speaker 2:

But you figure, the number of turns is fixed, right, so there's a maxima.

Speaker 1:

Yeah, so it would be 2 to the 6th times 2 to the 6th Both players double every single turn.

Speaker 2:

Yeah, that's right, because it goes back and forth.

Speaker 1:

Yeah, so it just gets bonkers where 512 regular wins is the cash. So I think it makes sense to put a little ceiling on it. I think 8 or 16 seems pretty reasonable. But yeah, and then I think locking it to doubling specifically is good Because, like I said, in poker there's a continuous wager option and so in poker you might assess oh, I have a exactly this percent chance of winning, so the optimal bet, exactly 1.5 times the pot, or something. And this just simplifies that and lets people focus on this binary.

Speaker 2:

Yes, no, so the thing I was finding challenging is it felt like the decision rule for this is that I should always double every opportunity I get the chance when seasonal reset happens and there's uncertainty about what my true rank is, because I always have a greater than a 50% win probability when seasons are reset because I haven't reached my true equilibria yet, and so I always do it. And then when I get to equilibria and I'm starting to feel like I have closer to a 50% win rate, then I basically stopped doing doubling cubes and then it just basically becomes an error bar around my true rank range and that's the decision set for the feature and yeah, I think that's it.

Speaker 2:

Like the feature has been solved. I have nothing like. There's no more gamesmanship for me.

Speaker 1:

So you're right that the that macro strategy of if I'm better than my opponent, just double every time and also you're probably hitting bots, which could basically give you 100 percent win chance yeah, but there's still a lot of within round decision making around the cube Right, like you draw a bad hand early on and your opponent lets a play, makes a turn to play You've got nothing on the board, then they double right. Like you still have a decision there about do I retreat or not. The decision is simpler when you know you're better than your opponent and it gets more complex and even match up. Yeah, that makes a lot of sense, I guess.

Speaker 1:

One last commentary is a lot of games are real-time, team-based games, like the most popular ones are all like Counter-Strike, Rocket League, league of Legends and there are some. It is trickier to adapt this to a real-time game. I could imagine maybe in the base there's like an item you buy in League of Legends that activates the doubling cube. Or maybe there's an objective on the map, like there's like a big cube monster you have to kill to activate the doubling cube or something. There might be some creative in-world ways to energetic ways to do the doubling cube.

Speaker 1:

And then another tricky thing is the team component, where your team as a collective has to decide whether to raise, match the raise or to forfeit. And how do you do that? Is it a voting system? Is it majority rules? You know how to, so there's definitely some complexity there to be worked out, but I do think that this mechanic could be applied to almost every ranked game and I think that definitely adds a lot of spice on top. You don't have to change the core mechanics of the game very much to add this extra meta layer.

Speaker 3:

Yeah, a lot of video games aren't like asymmetric. A lot of early board games were like a lot of the fun of them, especially poker, was the asymmetry information. Asymmetry Seems like this adds in an ancient, like the title of your ancient inspiration. Oh, look at that. That was not intentional, but this is like an ancient form of game that could be added onto any video game, which is really interesting, and I love the idea of being able to forfeit like a public, like a forfeiting option, but your opponent knows that the team is debating on forfeiting. Like the opponent, the opposing team has three out of five votes to that could be such an interesting information game. I don't know if, like, traditional gamers care about that, though they want to come in and they want to shoot stuff.

Speaker 2:

The incentive this creates, the incentive to forfeit, though, which to me, is the remarkable part of the design. It actually isn't the stakes as much as I thought I would normally be attracted to this stakes as much as I thought I would normally be attracted to this.

Speaker 1:

The forfeiting is actually interesting. Yeah, you guys ever played a game of starcraft where you know the terrans can lift up their buildings and hide them in the corners of the map and, like, you've won the game, defeated their entire army? But they're just hiding all these buildings just stalling for time to frustrate. You know you snap that doubling cube.

Speaker 3:

Now they're forced to forfeit instead of just like wasting your time is there a way to use annie to also make this more interesting, like? In poker like oh, auntie a, a sorry auntie like every I don't know what it is in poker, every 15 minutes there's a raise or whatever so they kind of do that auntie introducing aunties.

Speaker 2:

There's an example oh yes, apex legends does this with their rank system there's a minimum. There's a minimum number of points you enter when you enter a Battle Royale match, which is your ranked points, and you get progressively more points based on where you rank, and so there's a cutoff period where you become point positive and there's a point also where you're negative.

Speaker 1:

Meaning you're anteing more than the average.

Speaker 2:

You're entering more than your entry, so your net position becomes negative so you have to it's, but it's basically like a cost yes, but and it also increases as the ranks go up. So as you go from the lowest rank to the highest rank, your minimum break-even position to come out ahead or to come out at the same rank, increases. Like you have a definition 10th or above and a fifth or above, it's something along those lines that's like draft and mtg arena.

Speaker 3:

It's like it costs 900 gold or something to, or 900 gems to join a draft, and then you have to win four matches in order to be positive, but you're negative for the first three that apex thing is interesting because it's like stakes-based matchmaking.

Speaker 1:

You don't necessarily have to put the players of the same skill level together, just make the good players wager more so that the bad players EV is still higher.

Speaker 2:

That was the innovation of that rank system when it came out. I never understood why more people didn't do that, because it solves the battle royale problem of matchmaking, which is just give them different bars to achieve.

Speaker 3:

So how does it work? They have, based on their rank, they have a certain amount of this currency and they're able to wager it. Yes, so you start with this. Yes, so I've got more in my piggy bank, so I'm going to naturally be put up against people with bigger piggy banks.

Speaker 2:

Yes, because you're going to have a higher rank Now. They don't have to do that. They can mix them together because it isn't zero sum objectives.

Speaker 3:

So how does the?

Speaker 2:

match work different. Everyone can have different bars. So if I'm in gold, so let's say I'm in gold, I'm in gold, which means that I have 10 000 points, because 10 000 points corresponds to me being in gold. And so to enter a gold match, I'm you're not entering gold match, you're entering a match. Well, at gold, I might have to pay a thousand points, that's, that's the entry fee. And so you might only have to pay 100 points, but you have less points because you're in bronze. And so the outcome for us to increase our ranked position, to get positive points out of it, to get more than we paid out, is different.

Speaker 3:

I have a higher bar and you have a lower bar, so you might be 10th or above and I'm fifth or above and because I have a bunch of golds in my queue, if I do win, I get a larger portion of points because I versed a bunch of golds. Yeah, I think. So there's like a bigger no no, you do not get.

Speaker 2:

You do not get additional bonuses for beating super high tier player. Unfortunately, the only end of round multipliers for your ranked points are how long you survived and how many kills you had. You basically your damage. Interesting, they're multipliers on one another. So if you get to the end of the game and your last squad surviving and you did a bunch of combat, you get these insane multipliers okay it's a fascinating system.

Speaker 3:

I don't understand how the currency thing, though. How does it fix the matchmaking?

Speaker 1:

so here's's, I think, a simplified version, right? Imagine you and I are evenly matched 50-50, right. So we both put $5 into the pot and then whoever wins takes it all. Right, the game says there must be, let's say, $10 in the pot. Now let's say you and I are unevenly matched. You're way better than me. Let's say it's 90-10 odds that you win.

Speaker 3:

So you put in nine dollars and I put in one dollar and then we're finding out.

Speaker 1:

So the pot size is still the same, it's just your wager is higher if you're more likely to win, but if you win, you still get those points. Yeah, but your upside is lower than mine. You have to do better in order to proportionally speaking.

Speaker 3:

Yeah, I don't know if that's how apex works, but that's how I assume. Okay. So when phil said it doesn't you don't necessarily get more points because it was a higher ranked opponent, it's because there's, like, this pot that's filled Could be 10 really talented people or 100 not talented people. So that amount doesn't change. But okay, that makes sense. Wow.

Speaker 2:

Did I totally misrepresent Apex's thing. Nope, that's completely right. The only thing is there's no pot. That's the other thing I want to mention. There's no pot. That's the other thing I want to mention. There's no point pot. That's ahead of time.

Speaker 1:

Isn't it implicit that the sum of the payouts is some amount?

Speaker 2:

because you can have a bunch of kills happen in a match and that would technically be a match that rewards more points than a match that has zero kills because, remember, there's also respawning, so you can do a bunch of damage and you can have multiple kills of the same character if they keep reviving, which technically moves out almost like the point. Supply curve might be one way to think about it. It's an incredibly clever design and solves a whole bunch of problems.

Speaker 3:

Eric, once you start working at Second Dinner, I want to know an estimate of how many players care about this. It's really hard to value, but yeah, I had the same thought.

Speaker 1:

The system's probably really unintuitive for most people, and so I imagine most new players don't even really engage with the system. I'm curious how accurate their assessments are, Like when they're forfeiting. Are they really forfeiting at 25 and under?

Speaker 2:

I think they're more likely to play on. I think they'll play on when they shouldn't.

Speaker 1:

That'd be my bet. People are biased towards putting in more.

Speaker 2:

Is there an explanation that's not bias? Is there like a more of a price theory explanation for this? Isn't it just basically that like marginal utility curves are I was going? To say losses are bigger than gains in a marginal utility curve.

Speaker 3:

There's a in micro theory. There's this parameter in macro theory as well, but where there's like this risk parameter, I forget what the Greek symbol like traditionally is for it, but it basically tells you how risk averse or how risk loving they are and they're going to. They're going to change their behavior quite dramatically and you can tell, based on the cutoff, the decisions that they made, not how irrational they are, but how risk averse or risk loving they are.

Speaker 2:

So that is a utility story. But aren't most people risk-averse? I guess the people who self-select into RNG-based games that are probably risk-loving. That's probably a fair assumption.

Speaker 3:

Yeah, there's actually a bit of literature on this. I think people are more risk-averse than you'd think.

Speaker 2:

Or, sorry, risk-loving. That wouldn't predict this behavior, though, right.

Speaker 3:

Sorry, they're more risk-loving than you would think. Now the behavioralist would tell you that they're dumb and they're irrational, but the information theorist would tell you that they're risk-loving.

Speaker 2:

Do you know? What would be really interesting, Eric, as an experiment, is to basically send a notification to people when their win probability is low and to see whether or not they'd take the forfeit.

Speaker 1:

Give them more information.

Speaker 2:

Basically give them a prompt to forfeit, because the other thing that could be happening here is that they're opting into it because it's the default track. I have to opt into forfeiting.

Speaker 1:

Yeah, that makes sense Default to forfeit as opposed to default to continue.

Speaker 2:

Exactly, you could have them auto forfeit and then you have to de-forfeit and continue playing.

Speaker 1:

I will say in poker one common reason cited is I want to see a flop or I want to see if you're bluffing or not, like people sometimes are paying for information or paying for a conclusion.

Speaker 2:

That's a good price.

Speaker 3:

theory explanation Do you think the stakes make the game more enjoyable. Because that's why so many people do casinos and betting and gambling is because there's stakes and it's just so much more exciting.

Speaker 1:

For me it just adds an extra element to the game. Like without it the game is super simple. It still is quite simple with it, but like it it adds a much. Yeah, the decision making is more interesting. I'm sure some people just love the. They just want to go in all in every time. On reddit some people are like I just snap every single game. Do we know if?

Speaker 2:

I'm that guy, I can game.

Speaker 3:

Do we know if they added the mechanic after the core game was designed or if it was one of the original gameplay?

Speaker 1:

It was the very first thing they did. Apparently, the very first prototype was they played Hearthstone with a doubling cube, so it's like really deep roots.

Speaker 2:

Speaking of IP, there have been a string of reports. Man, you had to be living under or inside a fridge. Is that what you do to survive a nuclear blast?

Speaker 2:

To have not heard them, and this is on the wake of the news that Fallout, the TV show on Amazon Prime, has now become the most watched Prime series. It also has reached I believe it was 60 million to 70 million viewers. I believe it was 60 million to 70 million viewers, and this is on top. This is as a proportion of Prime's monthly active viewers, which I think I read a report it was around 200 million. So you're doing about 40, 40%, maybe 30% of the subscriber base activation, which is basically bigger than Lord of the Rings.

Speaker 3:

It's probably cheaper to produce as well.

Speaker 2:

I think that those P&Ls are closer than anyone might imagine. I think it was an extremely expensive show and it was showrun by jonathan nolan and his wife, lisa joy, who did westworld, and they are. He is the brother of christopher nolan of batman fame. It's done a lot to bring back sales of the game. So there was also also an 80% sale that was on Valve when the when TV show was airing and sales of the game spike. Not only did sales of the game spike there's actually been not a ton of conversation but there was a lot of downloads and revenue increase in Fallout Shelter, which is their mobile free to play game. In fact, the revenue impact of Fallout Shelter may be the biggest out of all the products and they've also started to sell so many copies that engagement is coming back for both Fallout 76, which was the last release, and Fallout 4. So the question is okay, is this a part of the transmedia strategy? And I've always seen the transmedia strategy as like the rising tide that lifts all boats. So when you have a main media property, you get all these other properties which can grapple onto it and then basically experience increases in revenue as the tides start to increase, so like when Witcher came out the TV show on Netflix, the game saw revival in terms of player bases. When any one of these IPs comes out, you always see the game rise in sales. Now the problem with marking this as a big item on the P&L is that it's just not that much fucking money. This is. The whole problem is that a lot of these headlines are from low bases because no one is playing the game to very high percentages. Now in Fallout's case, I think the revenue impact was actually very, quite real. I estimated at about 50 million back of envelope. I might actually have to revise that upward because the game is still actually moving quite a bit of units at its full price when it's returned to full price, which actually makes me question whether or not the price too low during the sale, whether or not the price willingness to pay was highest when the transmedia show was going on. Why should you discount? Maybe you should actually increase price or at least hold steady.

Speaker 2:

But the marginal revenue, if this is the ceiling, if this is the ceiling, then the marginal revenue impact of this transmedia strategy is fairly muted. It's not very high. You're not going to experience all these crazy revenue uplifts because this is a best case scenario. Look at what it took to get maybe 50, 60 million in incremental revenue. The rumored budget for Fallout is 150 million, plus another 100 million for marketing. That gets you to 250 million. So you're saying, okay, I got a 50 million incremental uplift on 250 million spend. Now, that doesn't include the main revenue from the series, but that's still just a pretty small share. When you think about 50 million on 250 million, it's got to be a little bit higher. And remember, this is the ceiling.

Speaker 2:

This is like a perfect execution of an IP, of an excellent showrunner, of having a lot of games on market. There's a lot of things on a prime distribution network and so this is the best transmedia has to offer. I think the strategy is not very viable. In fact, like I would offer a different version of transmedia, a different definition of transmedia, I think in many cases we are just talking about franchise, management and licensing, which always makes sense. That's basically zero. Marginal cost revenue is for Bethesda just to go to Amazon and say, okay, write me a check for $50 million and I will give you the privilege to go out and spend more money to make a TV show against this IP. That's my position as a licensee is that I don't have to do any of the capital costs. I just get a nice check and Bethesda got a nice check here. The question is whether or not you can include this as a big line item in a transmedia P&L, and I just don't think you can do that based on this. The ceiling is way too low.

Speaker 2:

But I would say like the real kind of transmedia, I would argue, is that those things bring up and build an audience for the next product, like the reason that Fallout TV show was successful. It's because they built a fucking gaming IP. They built a fucking gaming IP, marketed it well, sold copies and built an audience for their brand. That's what transmedia is about. It's about the next project. It's about how you carry forward momentum. It's not about the rising tide lifts all boats. It's about how these things end up building distribution networks that catapult the next iteration of the IP, the next incarnation of it, to even grander heights.

Speaker 2:

That would be my argument. I think there's another angle where transmedia is like a loop where, like you consume one force in form of media and it pushes you to the next one. It propels you, like I read comic books and it propels me to watch the tv show. I read the tv show and it propels me to watch the movies, but I just think, by and large, we've gotten transmedia wrong and the economics are bankrupt. Okay, I'm fucking done. Now.

Speaker 1:

I did my rant I don't even know what the take on transmedia is. The consensus one is Because you said $50 million on $250. That's 20% margin.

Speaker 2:

That sounds pretty good, to me it is. But that's the ceiling. Okay, but Amazon put the bill for this right.

Speaker 1:

Amazon didn't make this, expecting Bethesda to make a bunch of money from it. That's true.

Speaker 2:

That's true. As a licensor, Bethesda has come out ahead. There's no doubt about it. All I would argue is that the transmedia line item in a P&L, if this is the ceiling, it's not a very strong line item at $50 million, Because we do care about how much LTV the brands add to the franchise. We care how much the uplift is in terms of just how much revenue Fallout can generate regardless of who gets the distribution rights.

Speaker 1:

So transmedia alone is not enough. Right, the show you're making making you still need to sell it to a network and make money on that sale. Right, like can we do?

Speaker 3:

can we define transmedia, because I assume that in order to be a transmedia firm, you need to be like. This just sounds like partnerships. This doesn't sound like transmedia. I don't even know if I know what transmedia is, even though I've had it pitched to me three times that's been part of.

Speaker 2:

The reaction to the piece has been like what is this? Just licensing?

Speaker 3:

I guess it basically is licensing unless you want it to mean something more it's not like a company comes out and says, hey, we've got an idea for a media franchise and we're going to create a video game, we're going to create a movie and we're going to create a book, we're going to ship all those things at the exact same time, because that seems like a terrible idea to me. It also seems not utilizing the beauty of specialization. Which video game companies are good at making video games, movie companies? I don't understand transmedia at all. Do you have a definition of transmedia, phil, that you use?

Speaker 2:

I think that is the exact issue is that there isn't a wide consensus on what this means and it sounds like it's just a sexy rebranding of how you build a franchise and how you be a licensor.

Speaker 1:

Is that a metaverse situation?

Speaker 3:

I don't even think it's that right, because a metaverse is still one company, it sounds like. Oh, I just meant as a buzzword that means something and everything. Yeah, as a business, like transmedia as a business, it sounds like it would basically be a bunch of communications consultants who try to make sure that, oh, we're going to get you in contact with these guys, they're going to make your movie and these guys are going to make your video game. That seems like the business.

Speaker 2:

But hold up. I do think we could be fuzzy on the details, but I do think one of the transmedia thesis planks is that when you go build an IP, you're going to have all these other IPs that are going to monetize that central event. So if you go out and you launch the Avengers, all of your other Marvel properties are going to experience uplifts, like there is a thesis here. The transmedia thesis is that rising tides lift all boats, and so you want to have all these different properties in a shared world. I would say what transmedia also needs to be, though, is it can't just be climactic universe, because then what's it? That'd be mono media? Right, because marvel is basically just told over movies and tv. There's no lore-based tv, lore-based video game, so I think, like for transmedia, you need to tell things over multiple media formats the strategy basically sounds like the marvel cinematic universe where it's.

Speaker 1:

it's not just one story, it's a whole series of things. They all intertwine with each other, they all boost each other and some of these mediums don't monetize very well, but build brand. And other mediums monetize better. Right, in Fallout's case, the mobile game monetized the best. But it's like Disneyland and Disney movies, right, like they sell the merch in the theme parks where they get a lot more money than the movies do At the end of the day, I think it's just a fancy way of rebranding partnerships.

Speaker 3:

It's a win Bethesda gets some extra cash for something that's been sitting around for many years. There's zero marginal costs at this stage, literally free to make more money. Prime gets this awesome IP out of the deal. I don't really understand why it's all of a sudden this like new phenomenon. I'm not like necessarily bearish on the idea, like on the power of, of multi-platforming or transmedia, like I think that 20%, like Eric said, is pretty good. I just don't really understand why it's. It sounds like a new thing, like it sounds like they're trying to make it a new thing and I just don't really know if it is a new thing.

Speaker 2:

I think now there's more media formats to pull this off on than there have been. Like it's never been easier, I would say, to monetize IP through different media formats. It's never been easier to launch a game, a TV show, a movie, a podcast. You can now have more things at your fingertips, and what was your only other form of monetization Maybe 50, 100 years ago? It was like you can do a book and then, I don't know, maybe you can do a poem. Like your arsenal is limited.

Speaker 1:

Yeah, imagine the Fallout podcast or the. It's like survival diaries, I don't know. So the show.

Speaker 2:

Marvel actually commissioned an original podcast series. I think it was aired on Spotify, Cool.

Speaker 3:

Was it like meta or was it content, like it was something that they were actually? It was content. It was like original content. It was like a story.

Speaker 2:

Wow, sounds cool. So, anyways, I'm just skeptical of the incremental uplift. That's it. I still think building franchises can be positive and, ultimately, I think the thing that's important about building franchises is how this TV show is going to affect the next Fallout game. It's not about how it uplifts the games right now. If this is the ceiling, I just don't think it's a very high ceiling. It's about how this TV show is going to sell units of the next game. That, to me, is the important part here, and that, to me, isn't transmedia. That's just franchise building.

Speaker 3:

If somebody came to me and was like if I was a VC? And somebody came to me and was, like pitch a company. That is transmedia. We have this awesome ip idea and we're going to build three different products around it. I would tell them, like absolutely not go to hell. Like you can't. It's nearly impossible to build a video game, nearly impossible to build a movie and nearly impossible to write a book. Like, trying to do all these things that are really hard to do all three at the same time sounds like a horrible disaster. So if that's what it is I want, out you don't have to do it with the same time.

Speaker 1:

Sorry, skylanders.

Speaker 2:

Is that transmedia? I guess when you launch a TV show?

Speaker 1:

it's a card game and a TV show.

Speaker 2:

Yeah, I guess that I know, but it's not really telling you story though, it's just, it's more franchise management, though, like the card game, doesn't tell us story. Oh, does that make sense, that's probably.

Speaker 1:

The legends tell a story.

Speaker 2:

It did. It tried to. By the way, I looked into that, there was no uplift during that.

Speaker 1:

Yeah, yeah, I was going to say one of my takeaways from your data was that Fallout, shelter, the low-cost, high-monetization-feeling, broad-appeal mobile game, did benefit the most from the Transmedia campaign. Yeah, which tells you like TV shows go wide, right, video games go narrow and deep. Tv shows go wide and you want a game that can capture the upside of something that a lot of people have access to and has a high spend. And, yeah, this is an interesting takeaway, it's the monetization side of the media, whereas, like League of Legends, for example, is a pretty niche hardcore game and so, like people watching Arcane aren't going to boot up League and like suddenly become hardcore League players. But up league and like suddenly become hardcore league players. But if there was like a like a match three league puzzle game, maybe they would bend in that.

Speaker 2:

I would say, like the struggle is like getting these Venn diagrams right. That's another problem I have with this theory is like the Venn diagrams of people who are into one media format and another, like it's, by definition, smaller than any audience that's into one of the media formats, like that, to me, is another shot across the bow at this theory. It's like those Venn diagrams are always going to be small. Okay, I got to run guys.

Speaker 1:

Let's talk soon.

People on this episode