36. Richard Garfield Interview: Variance as a Design Choice [Epilogue]

Sike! One more, nerds. Richard Garfield's back. We chat about the role variance plays in his game design. It's cool.  

Read more on Richard's latest in-progress video game, Chaos Agents.

★ Support this podcast on Patreon ★

Creators & Guests

Host

Zach Armstrong

KeyForge Podcast Producer & Host since 2020

What is KeyForge Public Radio?

KeyForge Public Radio is a strategy and news podcast helping create informed KeyForge players, hosted by Zach Armstrong. Visit the website to see all strategy episodes, and join us on the journey to unlock the Unique Card Game. To support KeyForge Public Radio, join our Patreon or rock some merch!

[00:00:00] Psych one more nerds. Dear listener, welcome back to one more episode of KeyForge Public Radio. We are excited to be back because, we have a special guest on today that you may have heard of elsewhere and on Call of Discovery, of course. We have Richard Garfield, creator of KeyForge and a lot of other games, enjoying a great career as a game designer.

So, Richard, thank you so much for, joining me from, around the globe today. I appreciate having you on again.

It's a pleasure to be here. Always fun to talk about games. ready to, ready to do that.

Fantastic. we are going to be talking about Listener a little bit more than just KeyForge today. KeyForge will be a part of some of the conversations, but we're going to talk [00:01:00] about, I'm going to ask, be asking Richard here a couple of questions related to variance in games, and if you're familiar with everything from KeyForge RoboRally to some of the more recent work you're gonna know that that variance in many different ways shows up in many people's games, but especially in the games Richard collaborates on here.

So, uh, one of the one of the games, Richard, that in doing some research on KeyForge and some other things I found Spectromancer, that when, with three donkeys, you all helped develop and then publish back in 2008, And, KeyForge fans are gonna be a bit familiar with one of the core mechanics of Spectromancer, I believe, in that if I'm correct, each deck was randomly assembled from a known pool, though I think there was some customization in the campaign mode.

Is that correct?

Yeah, it's been a little while since I looked at it, but my memory of it is that there was a fixed deck, and each person [00:02:00] was dealt some cards from that deck. So in particular, if you had a particular card, you knew your opponent didn't have it. And those lasted you for the entire game. Now in Spectromancer, each player did choose one specialty, and that was just theirs alone, they would get a handful of cards from that specialty.

Yeah. So, uh, before KeyForge, what, what did it look like when you pursued variance in your game design? I know the original version of Magic, nobody, the, in the vision, nobody was really expected to have access to all the cards, at least in the, you know, kind of 1. 0 vision. So what did variance look like in, in that age before KeyForge?

Variance comes in a lot of different forms, and one of the reasons that my games tend to be high variance, because I'm not, it, it, I should say that I love low variance games as well, like Go, and, and, and Chess, and And one of the reasons that [00:03:00] I am drawn to high variance games is because those are games which they're just not as common, and it's something I like, and so when I'm want to play a particular type of game and it doesn't exist I tinker with it.

So, uh, one place that variance comes about is So, one of the reasons people remove variance is because what they're trying to do is trim off bad game experiences. And,

uh, and, and so, they'll, and you know, that, that that could be, so if you, if you began chess with random pieces, for example, and whatever was generating your pieces gave you all pawns, that wouldn't be a good experience.

And so, as a designer, You sort of trim those cases out, and the more you trim the less variance you have, and the more of a curated game you have, until you get [00:04:00] all the way to the place in chess where you have a fixed army, and, and so, my personal goal Preference, I would rather play a few bad games in order to have more variance in general.

So, so I draw that line closer to the have all pawns side than another designer might. Just because when I play, I don't mind. I think it's worth having a game or two that Isn't great in order to get the extra space you get of possibilities in your game. And so, and so I always draw that line, I err on the side of too much variety, which means that some of the games aren't, aren't, there'll be blowouts in one way or the other.

And that's, you know, that's a cost of that approach and something I accept.

Yeah, for sure, for sure. You're talking about being a bit more high variance in a number of things, and I think anybody who has enjoyed a raucous round of [00:05:00] RoboRally will be familiar with some of the more, Kind of joyful extremes of especially hilarious situations one can laugh at with all the flipping of the directions in RoboRally.

So, how do you think about KeyForge as a milestone in the games that you've contributed to? Sure.

Bringing up Roborally does give me a, a, a, a, a good idea. Well, an idea that I think is good. It's, it's an idea. It's to illustrate the this idea of drawing the line for variance. Roborally does have the high variance, as you said, and all sorts of crazy things can happen.

It's something I've always loved. And from the earliest days players have complained about getting dealt terrible hands, which is okay, I don't mind it if you get a terrible hand and you suffer for a turn or two it gives you an opportunity to I like being able to work with work my way through challenging situations.

[00:06:00] But When you get dealt all turns, that's not great, and and if you get that a couple times in a row, which is certainly possible, it's, it's pretty bad. And that's one of the things which led me to redesigning in recent years with the Hasbro version in like, 2009, and most recently with the Renegade version we, we made it so each player had their own deck which was just, you know, something I hadn't really thought about back when I designed Roborelli, because that was even before Magic, where the idea of people having their own deck was a big thing

and that it still allows you to get I think you can probably even get hands of all terms, but it's much, much less likely.

And if you do, you're going to get a lot of movement on the next term. And so that's an example of a designer my younger self putting the line in one place and then when given the opportunity to redesign, shifting it over, but still having a lot of variance.

Absolutely! Yeah, wow.

[00:07:00] So with KeyForge it, it kind of, it, KeyForge is, A high variance game in a different way, because as high variance as a game like RoboRally is, it is, at its highest level, a reasonably fair game. That is you you may get terrible hands, but there's an equal possibility the other players will get terrible hands, and so, it's fair.

And you may get an unfair situation in one playthrough, but, you play a bunch, it'll all average out. And that works for some people, works for me, but, you know, some people like a little bit more consistency. With KeyForge, it sort of, is a little different in that the decks themselves are unfair.

They're not designed to be unfair, but they're designed to be very varied. And that carries with it this this possibility of being unfair, and just like with my game design, I tend to err on the side of I'd rather have more [00:08:00] unfairness and more variety than I would cultivate everything so everything is fair.

First of all, I don't, it's really difficult with the game of KeyForge, like, even if you started to say, let's make the, make this a fair game, you, you cannot do it. It's just, there's Too much going on and no matter what, what, how restricted you make the deck design, people are going to complain that you did it wrong and so, and so I, I just tend to jump in with both feet and and to sort of help people work with that give them tools to uh, Um, uh, to moderate their own environment for the experience they like with handicaps, the chain system.

But then also with my own design and designs from the community that I encourage systems like where you each play one you each play the deck once and then you swap the decks. So when people want that sort of fairness, there are, you can take unfair experiences and Always make them fair.

With Roborally, [00:09:00] you might get one unfair game, but you play a league, it's all going to wash out. Same thing with poker. You're going to get an unfair season, but if you're a good poker player, you're going to win in the long run. With KeyForge, you've got these decks which are unfair, but But the variety is crazy and it really excites people.

You can make that fair if that's what you want.

Yeah, absolutely. I think the that taking ownership of one's play experience is a pretty big, a pretty big thing that people have talked about in KeyForge because there is so much of that With the design of KeyForge, there is so much for people to, if they're in a complaining mood, to grab onto and say, Oh, well, are some decks stronger than others?

And we'll say, yeah. And their reaction to that answer is often the dividing line between who may enjoy the game and who, you know, may not enjoy the game. Where if they said, oh, some are fairer than others, like, like, can I take the strong ones and put them against each other? Can I go try to beat the strong ones with some ones considered?

Week, right? The [00:10:00] infinite puzzle is something I like to call, you know, a set of KeyForge matchups. Even some of my oldest decks will have new life when there's a new set out. And there's a new kind of unfairness to, to try out against, and sometimes the good decks have gotten worse over time, and then sometimes new things come out, and all of a sudden, my now six year old deck is very good again.

Just depending on, on the matchups and that variance, which is a lot of fun.

Yeah, I know. I love that. And, and that, that is that question, that reaction to that, that statement that, that some decks are more powerful than others. You, you're right. That is a a touchstone for players. There's an illustration of that in development that we had, which was when we had a particularly powerful deck some groups would, would because the groups were playing with the same decks some groups would fight over that deck but other groups, like my group and a few others, would fight to play against that deck because Because, I don't [00:11:00] know, winning with a powerful deck isn't that exciting, but beating a powerful deck, that's very, that's thrilling, and and, and it's all good, good gameplay, so, uh, and, and, and, and when you think about a game like like Hearthstone or, you know, Magic one of the, Which, you know, I think Hearthstone is amazing.

I've enjoyed it a lot. I think Magic's terrific too. And I, I, I mean, that's, I'm the designer of Magic but it's worth pointing out because, because KeyForge was designed in many ways to Bring back some of the lost magic of magic. That doesn't mean that I don't like where magic went. But but one of the things I miss when I play magic and Hearthstone is when I drafted, when, when I'm drafting a deck or playing a deck and I have to get rid of cards, which I like, but just aren't powerful enough to make the cut.

I hate that. I would rather play with a deck, which is fun, exciting, varied, and then and then you know, then the play experience is [00:12:00] more challenging for me. And it's, it always perplexes me that so many people just really don't get that. Like, and they're the same people who would certainly understand in a computer game, you don't always It's sort of analogous to me to yeah.

Saying, well, I, this new shooters come out. I'm gonna put it all the way on easy and play on the easiest because then I, I, I own everything. I can understand the appeal of that, but I can't understand that the, that the, like the personalities of the people are switched here because the, the most hardcore board game players are like that.

They want to play the most powerful decks. They don't want the challenge of playing sort of more varied positions.

Yeah, and there are several other dynamics that came out of decks being unique, not just, you know, here's a set of strong decks, here's a set of weaker decks, or whatever, that enriched those exact challenges, right? Like, I remember I had two decks right off the bat before getting some others, and playing playing the deck that was [00:13:00] would often lose the matchups.

I remember playing again and again with a friend until, and I remember, it's, I remember the victory of that weaker deck against the quote unquote stronger deck after five games in a row losing more than I remember most of the victory with that other deck. And another with those decks being unique and with, you know, Selene doing her work, the, oh gosh, staccatic, wait, stochastic, exotic, well, I got the first two, but the, the naming algorithm.

Oh, right, right.

Uh, Selene was the name generator, and it's Stochastic, Exotic, yeah

Well, we're gonna have to

List exploiting?

List Exploiting, and yeah Name Engine or something, yeah,

Right, well, there's a podcast in, I think, 2020, early 2022, where that's laid out, so we'll have to go check that.

yeah.

But the one fun thing with strong versus weak decks is with KeyForge, with matchups and new [00:14:00] sets coming out, the definition of strong and weak would change. Over time. And I also remembered when I was still understanding and growing in my understanding of the paradigms of KeyForge and how draw the three houses, no resource costs and drawing up to end of turn, all mechanically affected things, I remember going back to a deck I wrote off as middling and playing it much more much later when I had a deeper understanding of the game and it turned out that the limit on that deck's performance was not that deck and its 36 cards, but was in fact my understanding of that deck.

I had not been good enough for that deck, which was a really special kind of card game moment, especially when it has a name like the Rani of Bambagom, and it has a, you know, a bit of personhood to it.

yeah, that's a that brings us back to Spectromancer, actually cause there was a, a phenomenon I discovered. I had sort of understood that it was probably true, but it was really illustrated in in Spectromancer that there's a danger [00:15:00] of balancing things too much. And, and the other, the, the thing which came up in Spectromancer was after it was out for a little while, we We looked at the win records for the different classes, because there was a lot of people complaining that the Priest class was weak.

And this was to be expected, because Priest class was the free class, so everybody assumed it would be weak. But, the reason it was worth testing is because is because We did, we of course, we did not choose it to be weak, we chose it to be varied because it showcased a lot of mechanics in the game. So we did a study, and the really interesting thing was that was that for beginners The priest class was weak.

It had like a 45 percent win rate. So, it was within bounds for what I'd accept, but it was weak. But for experts, it was actually one of the favored ones, which is, like, it had about a 55%. And the reason was, is because [00:16:00] having variety, having flexibility, That's a skill testing thing. And so, of course, beginners are going to do worse with that than advanced players.

Similarly, there's a deck, the Necromancer, which people thought was broken, but it actually had a very low rate, and it was very good against beginners, but in moderate players, it was sort of even, and then experienced players, it was weak because you knew how to play around it. And this phenomena you see in In essentially every game if there's any depth to the game, balancing it for just the highest tier players will, will make it so the game is not as fun for the low tier players, and balancing it just for the low tier players makes it so it's not as fun for the high tier players.

And so finding a way to do both or a middle ground is really important.

Yeah. Absolutely. Absolutely. Now, with some of the games in the oeuvre here, well, the games themselves aren't overly simplified. They have kind of more you know, [00:17:00] family pointing art direction. Everything from like King of Tokyo, Bunny Kingdom. Half truths which is a pretty, has a pretty entertaining, a pretty entertaining setup, even Mindbug.

So how how do you approach, how do you approach the inclusion of variance in games which are perhaps more directed at a, at a family setting? I hesitate to use the word lighter, but often more directed at a, at a family or more casual group setting. What role does it play when there's maybe perhaps a different player in mind, or a different set of players at the other end?

Hmm.

well, high variance games are, are for family for broader family Party, Kids Deez, and, and non gamers High Variance is almost universally loved by that audience. They don't, they don't have the visceral dislike of, oh my gosh, dice, I hate dice, that sort of reaction. And [00:18:00] there's, there's good reasons for them to like it because because often times even if there is skill.

Anybody can win, or at least anybody can have some small victories along the way, which is, you know, something I aim to have. So, so I find that I, I don't have to work so hard with those audiences but with when I'm trying for games which are more gamer oriented, there are people who are difficult to sell high variance games to and when I say sell, I mean, sell the idea of and, and one of the things I, and, and they're not for everybody.

So I don't want to convince people who want more deterministic experiences to to play these games. But in order to advertise that. It's high variance. I usually like to have lighter, more cartoony, a little bit silly art, which is why you get a game like well, and it's, so, so that's why with KeyForge, for instance, you had the House of Martians

Hmm.

I was really worried that, Fantasy [00:19:00] Flight was going to take it and make it all serious and so, and, and so, and, and I, I really wanted to see sort of cartoony art with a sense of humor.

I didn't mind if there were serious areas in the world, but I didn't want people to look at it and say, oh, this is a serious game, then sit down and say, this is garbage. I've, you know, I just got owned by luck in the serious game, but if you make it cartoony and silly and it's got Martians in it. You don't, you're more playful sitting down, and that opens you up to this idea, and then later on you can find out, oh, there's a lot going on actually.

Yeah,

Oh, yeah. Absolutely. I think the Martians pretty solidly did that. Among other art directions, the Martians said, hey, don't Like, yeah, this game can be played at a high level, but don't take yourself too seriously. You know, you're, you're using little green men.

exactly.

And many of the other houses did a successful merging of two wacky archetypes to, to bring themselves together in some very silly way, like, Steampunk, you know, Steampunk [00:20:00] Giants and Brobnar and, and such.

So all very fun. All very fun. This Getting towards the end of the questions here. This is a bit more of a almost more of a kind of, not quite a full business question, but a lot of the designs you do or collaborate on are you were brought on as kind of like the design partner, or design originator, or a big part of the development, and you know, they get to stick your name on the front of the box for nerds like me, and those who followed the details of your projects will know that you are like a co owner of Popularium which I believe I have correct, which is making Chaos Agents.

There was a lovely little Gen Con TV teaser a few months back. So, what does it, what does it mean for you to be a co owner of a studio making a game that is, is following this thread of variance? What does it mean to be an owner in that situation instead of the, the designer for the first few sets?

I, well, it should be pointed out, [00:21:00] I am the designer for Chaos Agents. I do have a lot of I do have a lot of people helping me. But I would say being an owner there just it's really not much different than a regular game because I I Most of my games I get paid a royalty on and so it's just a matter of I'm willing to stake.

Whatever work I'm doing on this idea and games are a high, high games themselves are a high variance business and and they're not all going to be hits. And in fact, most of the time I do better taking A fixed amount of money than I would taking a royalty, but but I believe in each of the games and and I believe in it here and so, and so, by doing that, by taking the royalty or taking ownership in the business it just ensures that I'm, I'm, I'm being very serious about where I draw the lines for the game.

Thank you. example with the variance. And I don't actually think [00:22:00] that's necessary for me. It's an, I don't pay too much attention to the business end of my work. I I'm so focused on You know, just trying to make the best games possible and make it appeal to as many people as possible and to explore something which which is new.

What are some of the core ideas behind Chaos Agents that are new and that that are just exciting enough to, to put your stake, put your stake on like you're doing?

It is, it is an auto battler, I'm sure they must have said that and, uh, and,

and, And, and there are a number of other elements which I think are really exciting to it, but let's look at Autobattler first.

Autobattler is this is this amazing genre of digital game, which with you can, you can trace it back to before auto chess, but auto chess is really what started to define what an Autobattler was. And it really hit this [00:23:00] chord for a lot of players because it, it, it seems like this perfect meshing of what it means to play a digital game and what you love about analog games.

I feel like when I play these games that I'm not playing a game that I'd rather play around a table. And at the same time, I get a lot of the feedback. There's a feeling that I get when I'm playing a game around the table where I understand what's going on and there's a lot of mechanics which feel sort of gamey rather than simulating.

Sure,

And so, ever since auto chess came out, I, along with many other designers have been and taking that to heart and either working on board games, which are inspired from that or. Digital games which go in different directions and that Chaos Agents is, is one of those.

Fantastic. And, dear listener, if you don't know Chaos Agents, there have been a few articles, I'll link those in the show notes so you can get more of the details on that so we don't have to put poor Richard at risk here of the ire of the co workers. final [00:24:00] question here before we, before we sign off, Richard.

What do you enjoy most about getting to design games?

I like the feeling of, of exploration. That word comes up a lot when I'm talking about game design and gameplay. I, I feel like I'm exploring new territory and discovering new things. There's nothing I like more in game design than coming up with a mechanic or a system and not being sure how it's going to play out and then playing it and discovering something new.

You know, some and sometimes whatever that is doesn't work out, but I almost always learn something new. And And I still view myself as a student of games and because I don't think games are ever going to stop teaching me.

Agreed. And same here. There's a functionally infinite number of games to go play and design and experience all inside of them, for sure. For sure. Well, Richard, thank you so much for [00:25:00] joining me today. I appreciate having you on KeyForge Public Radio. And yeah. Yeah. Anything else for the KeyForge fans before we sign off?

Yeah thanks for the interest. Play games, play lots of games.

Woo! Awesome. You heard him. Alright. Thanks, everybody.

​