The Art of LiveOps

The Story of Game Narrative w/ Tom Abernathy: The Art of LiveOps S2E5

September 08, 2020 James Gwertzman and Crystin Cox Season 2 Episode 5
The Art of LiveOps
The Story of Game Narrative w/ Tom Abernathy: The Art of LiveOps S2E5
Show Notes Transcript

Game narrative is far more than just words and cinematic sequences. In this week’s episode, award-winning narrative director Tom Abernathy lays out for us how narrative functions within games, and why it’s so much more important, and omnipresent, than most people realize. 

Support the show

James Gwertzman (00:00:05):

Hello. I'm James Gwertzman.

Crystin Cox (00:00:06):

I'm Crystin Cox. Welcome to The Art of LiveOps podcast.

James Gwertzman (00:00:15):

Hey, Crystin.

Crystin Cox (00:00:15):

Hey, James. How's it going?

James Gwertzman (00:00:17):

Good. I realized we have a colleague, a former colleague of yours on today's show, Tom Abernathy.

Tom Abernathy (00:00:23):

Things have changed a bit partially because we wanted to do a better job of monetizing the game, which our player data told us meant keeping players in the game for longer, because the longer they stay, the more money they will spend. We had to start thinking about what you can do to keep folks playing. I think it is natural that to most people, certainly to most designers, the first thing that you think is not narrative content. I tell people that narrative as a discipline, basically, it's not about words or text. It's about context, and it's about meaning.

Crystin Cox (00:00:56):

See, Tom and I were together at ArenaNet, where he's the narrative director, and he's still narrative director at ArenaNet. He's the franchise director for Guild Wars. Tom comes from a background in ... it's very narrative focused. He was a writer in Hollywood, a screenwriter. He's written several games, but it's a very much a narrative, narrative designer.

James Gwertzman (00:01:22):

Now we're going to talk about how narrative applies to LiveOps, which I think is fun, because I think this new for him too. He's not necessarily been doing LiveOps his whole career.

Crystin Cox (00:01:30):

Exactly. He came on to ArenaNet, three years ago, and it's his first time working on a live game. But I really think he has a great perspective because he is so dedicated to storytelling. He's really dove in at ArenaNet to figuring out where narrative fits inside of a live game.

James Gwertzman (00:01:52):

Wow. Okay, cool. That's a subject near and dear to my heart. That'll be exciting. Awesome. Let's go chat with him.

Tom Abernathy (00:02:03):

Yeah. My name is Tom Abernathy. I am the studio narrative director at ArenaNet, and also the franchise narrative director, by extension for all things Guild Wars at ArenaNet. I've been doing that for a little over three years. I have a very long and checkered history, but that's amazingly Reddit, or excuse me relevant. It's just again a notification that said Reddit that came over my computer. Amazingly relevant to what I do. I started out very young as a child actor, professionally on stage and on screen. I did my first professional play when I was 11. My first movie when I was 13.

Tom Abernathy (00:02:42):

By the time I got to college, I figured out that I didn't want to be an actor. I wanted to be a director. I transitioned to being a theater director for a while after that, ended up going to film school. Along the way began to pick up writing as an extra thing because the film school that I went to, they know that ... there are not going to be roles for 50 people twice a year who get dumped on the streets of LA from the film school to play as director. You better have some other skill sets.

Tom Abernathy (00:03:14):

I took a screenwriting class because it was the mid '90s and we all wanted to be Quentin Tarantino and Spike Lee and discovered that I really like screenwriting. I'd never really enjoyed fiction writing. The process on is too slow for me or something mentally, I couldn't stay in it. But screenwriting really suited me. I spent some time in LA after that as a screenwriter. But almost literally, even from that point, from late in my film school career, I was trying ... I was also knocking on doors trying to get into games. Because I had a friend who moved out, he got a job as a production assistant at Activision and he rented the apartment over mine and we would play PS1 games every night after our long days. We would pitch to each other about how terrible the story was in them, because it was 1997 or eight. They were just starting to try to do that.

Tom Abernathy (00:04:03):

He'd say, "Yeah, I really wish we could get you in over at Activision studios because I think you can make our stuff better." I said, "I think that'd be great. I would love to try that." I spent the next five or six years doing little jobs, working some QA, writing a manual here, writing a few cutscenes there, trying to get somebody to give me a larger opportunity. Finally, was fortunate enough that that door opened for me with a game called Destroy All Humans, which unexpectedly ended up being a pretty big hit. It got talked about a lot for the writing, which is the thing that did not happen in 2005.

Tom Abernathy (00:04:41):

That was very unusual. That was the thing that opened the door for me into games. I was at pandemic. Oh, that word has different common connotations now, but I love Pandemic Studios. I work on Destroy All Humans one and two there, and the Sabbath Tour. Then I went to Microsoft, worked on a bunch of different stuff. The most high profile, which is Halo Reach. Then went to Riot, worked for a year on League of Legends. I also wrote on the Division and did some stuff for Man in the High Castle for Amazon TV. Then I came to ArenaNet and found the role that I am now in which I have enjoyed immensely.

James Gwertzman (00:05:28):


Crystin Cox (00:05:29):

Awesome. But yeah, this is ... I think you're our first narrative person on the show.

James Gwertzman (00:05:35):

I bet.

Crystin Cox (00:05:35):

He was purely ... No. Who's really focused on narrative. I think that's going to be a lot of interesting conversation around what narrative looks like when you're running a live game.

James Gwertzman (00:05:48):

Yeah, I'm really excited to learn more about narrative, especially in the content. I'm assuming, based on the context that narrative is used often in live events and creating live events that that can steer the narrative one direction or another, or can have interesting implications for your ongoing narrative stream.

Tom Abernathy (00:06:07):

Yeah. That's absolutely true. I mean, there's a number of different kinds of content that we generate for the game. Some of it is what we call Golden Path Content, which is fairly linear, and is story intensive in the sense that if you want to follow the story, you need to do it pretty much in the order that it's supposed to be done. Occasionally, there's some minor branching or choice options for various things. But honestly, most of that is pretty superficial. Historically, or at least, I'll say when I first had came to ArenaNet, which is a little over three years ago, that was the thing that most everybody thought of as the major story delivery venue.

Tom Abernathy (00:06:53):

Things have changed a bit. Partially because we wanted to do a better job of monetizing the game, which our player data told us meant keeping players in the game for longer, because the longer they stay, the more money they will spend. Of course, the game is free to play of course. To keep them in the space longer, we had to start thinking about what you can do to keep folks playing. I think it is natural that to most people, certainly to most designers, the first thing that you think is not narrative content. Certainly we weren't doing anything to incentivize people to stay in the game using narrative content. We haven't been giving rewards if you replay. We haven't doing anything.

Tom Abernathy (00:07:47):

Design went to work, designing a whole lot more stuff to do in the open world, repeatable events, things that you could grind on a little more, if that's what you wanted to do. Because we certainly have a decent number of players who enjoy that, although I think Guild Wars 2 is a game that part of its appeal to me, and I think part of it what sets it apart historically is that it's not a game that focuses on grinding in the same way that World of Warcraft does, for example. How to put this? It's an experience that ... this way I would think of it. I tell people, that narrative as a discipline basically is about ... it's not about words or text, which is a thing a lot of people misunderstand. It's about two things. Primarily, it's about context and it's about meaning.

Tom Abernathy (00:08:38):

What I mean by that, basically, is that narrative creates the contextualization in partnership with other disciplines, the contextualization of the experience, and then minds and articulates and expresses and hopefully makes a way for the player to share in the meaning that hopefully arises from that context and that experience. The example that I like to give is if ... as a game designer, I come to you and I show you this little thing that I've made in a gray box where you have a sphere and you can bounce the sphere off the floor, you can throw the sphere to another player. Every once in a while, if you want, you can go to one end of the gray box or the other and put it through a horizontal ring that's been set up on the wall of the gray box, that's a pretty fun for, I don't know, 15 minutes maybe. Then probably you'll get a little bored with it and you'll leave.

Tom Abernathy (00:09:39):

If on the other hand, I then tell you that your name is LeBron James and there is point seven seconds left in game seven of the NBA Finals, and Staples Center is going crazy with 17,000 people screaming their heads off and you just got fouled, and you get to shoot two foul shots, and the Lakers down by one. Suddenly, the context makes it a whole different experience. The meaning, the emotional value of that experience is radically transformed. Parenthetically, will say, as a big baseball fan, that this is the thing that I've been feeling in spades lately since we're all shut up inside because there's no baseball and I should be able to watch baseball now and I can't. The fact that that part of the springtime and summertime of my life is not happening right now hits me in a big way.

James Gwertzman (00:10:32):

Let me jump in there. In that sense then, and this may be super obvious, but I'm going to state the obvious. Narrative then in games is very different than narrative, and say film or books or traditional linear storytelling, because what you're saying is storytelling is about the context and motivation, and more of almost the ... it's almost more like a soundtrack as opposed and an emotional motivation as opposed to, necessarily the thing itself is that going too far?

Tom Abernathy (00:11:01):

I think those are the irreducible parts of it, put it that way. Those are the things that it's most important that narrative and games accomplish. It can accomplish a lot of other things, including as you suggest, stuff that's more akin to what you would find in a traditional more linear storytelling medium. We can do a lot of those things too, at least under certain conditions. But the thing that we are vital for is contextualizing the experience, and making it matter to the player so they love the game, and they want to keep coming back to the game, and keep playing it, and hopefully pay us a little money so that we can make more games for them.

Tom Abernathy (00:11:38):

To bring it back to the question of how does that fit into a LiveOps context then, as I said, we have our golden path stuff, which is the most similar thing to traditional storytelling media. You've got a fairly linear plot. You've got cinematics every once in a while that will play that are the closest thing we have to what you'd get in a movie or TV show. But we also need and want to contextualize this stuff that's not part of that, because if everybody's doing their jobs, and we are managing to figure out ways to make people come back over and over again and spend more hours hanging out in the game and doing stuff that they can do that's for various reasons, from to get various rewards or whatever, or just do it with different friends, then it only enhances our ability to get players to want to do that. If we can also make sure that those events and experiences are meaningful for them as well.

Tom Abernathy (00:12:40):

It's a little bit more terra incognita for me. It's a thing that most narrative folks have not had a ton of experience having to do. We were a little wobbly out of the gate. I think everybody was. I think we all wanted to happen. We weren't sure exactly what the best way to do was. We had to experiment some. But I think that during this season of content, we don't really call it that anymore, but what we're doing right now is called the Icebrood Saga. But it's the same ideas, our previous seasonal content. We have had to come up with ways to contextualize the open world meta events that are going to recur every two hours or whatever, and not just let them be activities. I mean, be very easy just to go, "Yeah, that's the thing you're going to do, have fun."

Tom Abernathy (00:13:34):

Really good example, I think of how we figured out a way to get some meaning out of something like that. We had a writer named Samantha Wallschlaeger who is no longer with us, but who's terrific. For Episode One of the Icebrood Saga design had these battles that they wanted to happen in the open world and that would deal with recur, and that was just basically given to us. We have these battles. You're going to fight these people every two hours or whatever. It was on Samantha, and to a slightly lesser extent the rest of our group, to figure out ... because we wanted to contextualize it, and we wanted it to be meaningful. How do you do that with a thing that repeats over and over again? I mean, is it Groundhog Day? Are you stuck in a time loop? There are things you can ... ideas you can come up with to sort of try to make some sense of it. They feel a little gamey, perhaps, until you get into the details.

Tom Abernathy (00:14:35):

I got to give Samantha a lot of credit, because what she did was in collaboration with the specific designer, she decided to make the people that you would fight ... she decide to make basically deposit that are NPCs that you're fighting with, each of the antagonists that you're fighting against is someone meaningful to them. We do this for a mount in GW2. The player character is more of a player character than in some games, but less of a full on character than in others. A lot of times the meaning and the emotional aspect will accrue significantly to one or the other characters and you get it by the transitive property because you're their friend, or whatever.

Tom Abernathy (00:15:26):

In the case of our character, Bram, for example, they had him fighting what he imagined in his head to be his father, who he has a lot of issues. Bram has a lot of parent issues in general. But this is one we haven't dealt much with. He's fighting in his head. He is in this hallucinogenic state, if you will, they all are. He imagines that he sees his father in front of him trying to kill him and so he has to fight his father in that's very traumatic. Our character ...

James Gwertzman (00:15:59):

He has to fight his father, you mean you the player have to fight the father?

Crystin Cox (00:16:03):

Yes. Both of you have to fight his father. Our character, [Margaery Delacour 00:16:07] who is this very cool film noir detective type and who has the spirit of her dead sister, Belinda, in her sword. Don't ask me why. I can't explain that

James Gwertzman (00:16:20):

As we all do.

Crystin Cox (00:16:23):

But suddenly she sees Belinda before her seemingly alive attacking her and you and she have to buy Belinda. The experience as Samantha wrote, it ends up by the end being so traumatic for Marjorie, who's a big fan favorite. That she literally ... basically when it's over says, "You guys go on ahead. I got to take a minute and recover from that because I'm feeling wrecked," basically. I just thought Samantha did a remarkable job of being creative and coming up with some ingenious ways to move character progression forward in the context of an activity that otherwise would have had no real meaning whatsoever.

Crystin Cox (00:17:05):

There's been a lot that this season for us. It's a challenge, but it's a cool challenge. It's exciting because I think I probably speak for a lot of game writers when I say that it's the game part of it. That is the most interesting part of this job, in others, I went to film school in LA and I worked in Hollywood for a while as a writer, that's the background that I came from. But I also love games a ton, and saw them as a very vital storytelling medium with great potential even in the mid late '90s, which is why I spent six years banging on doors trying to get somebody let me write a game.

Crystin Cox (00:17:43):

The things that were always the most interesting to me about writing games, once I started actually doing it, are the things that aren't the stuff you get to do anywhere else. They're the challenges that happen because you're dealing with nonlinear content, or you're dealing with stuff that always that's interactive, certainly. That just right there. It's a different ballgame than anything you do if you're writing a screenplay or tell a player a stage play, other stuff that's meant basically dramatic writing for actors to speak. There are tools in your toolkit that ... they're taken out that you can't use in this context. Then there are other things that you can do that you can't do anywhere else. That aspect of it is always really cool. But bringing all of that, finding those opportunities in a LiveOps setting, that's not the linear golden path is my favorite thing we've had to do this season.

James Gwertzman (00:18:30):

This is fascinating. I've got ... I'm sure Crys I'm bubbling with questions, because this is an area we haven't really explored at all yet in 20 plus episodes now, which is this notion of "Why?" as opposed to "What?" I've got one more and then Crys, I want to give you the chance to jump because I'm sure you also have a lot of ideas. One of the themes we've explored in previous podcasts is this notion that when you're designing for LiveOps, it's not making movies, it's creating a nightclub, and that you, the designer, elect a nightclub impresario. You're creating opportunities for players to have meaningful interactions with each other. I think a lot about nightclubs and some of them are just ... you go there and you dance and it's boring. Other legendary nightclubs, like the famous Studio 54 have so much more context and who's there and the personalities and it becomes larger than life, because there is context. There's a backstory to the club. I love that you riff on that idea a little bit, because you're nodding already on the on our video cam here.

Tom Abernathy (00:19:27):

I actually think that's a great metaphor. Actually, it had not occurred to me. But yeah, because especially with a game like ours, which is an MMO RPG, and it's fairly heavy on the RPG side, but there's certainly lots of players in the environment, particularly if you're there in the first few days after we drop a release. It is a big party. I mean, it's funny that you said that. But I'm thinking about some of my earliest experiences in the game. One story that I love to tell people is that I was walking around doing the personal story as a Sylvari, and at one point I supposed to find the Pale Tree, which is the source of Sylvari stuff, I won't go into it.

Tom Abernathy (00:20:11):

It's complicated. But I was supposed to find the Pale Tree. I go to the place on my little mini map where the Pale Tree is supposed to be. I don't see a tree anywhere. But there's tons of people around me. I'm looking around, looking around, and I can't ... I spend 15 minutes looking for it. I can't find anywhere. Finally I just typed in the chat box. I did a little wave emote and said, "Hey, anybody, I'm looking for the Pale Tree. Can anybody help me?" Literally, within 10 seconds, six people approached me and started offering help to do what I needed to do. There was this huge sense.

Tom Abernathy (00:20:43):

Anyway, there were so many people bounding all over the place and doing stuff. I never really played an MMO before Guild Wars 2, it's just hadn't been my thing. To go into that environment and see all those actual people, obviously, I'm seeing their avatars, but each one of them represents an actual person somewhere. Who's in the environment a game and isn't following a story, isn't doing a meta event at the moment, isn't doing ... just is hanging out and talking to people and do it like you're right, virtual cosmos and Manhattan's would have been really nice under those circumstances. I think it's a great metaphor.

Tom Abernathy (00:21:15):

It helps keep us narrative folks on the straight and narrow regarding linearity because it is so easy for us to fall into a trap of feeling like we have to tell a story that is linear. In fact, there are lots of ways to tell stories. The human mind did not evolve to take advantage of some of them, but that doesn't mean that we can't adapt and do that. For example, we had a prologue episode for this season, that was called Bounded by Blood in which one of the char imperators, char or race in the game that you can play, one of their leaders calls all the char back to the Blood Legion Homelands for a big powwow plus Woodstock basically and Burning Man. There's literally a heavy metal concert, a char, heavy metal concert with a band called Metal Legion.

Tom Abernathy (00:22:25):

We knew going in with this season that one of the things that was going to be different for us was we're going to have a smaller footprint in terms of the number of lines that we could write and record a VO. We were going to have to shrink that overhead a little bit, because we wanted ... we were going to put out episodes more frequently. That'd be a little smaller. At first that felt an imposition because it was like, "Well, how do we tell stories that are as interesting or as robust as the ones we have been doing with less time, and fewer lines to spend basically?" Then once again, our writers Novera King, principally, who's the narrative lead on GW2, Living World, I should say, started getting creative and coming up with the idea that you can give people, you can move the story forward without people having to experience things in a certain sequence.

Tom Abernathy (00:23:28):

That may seem pretty obvious and there's a sense in which it is obvious. I referred earlier to the ways in which when you, as a writer, move from linear media into games, you come to realize that you have to give up some of the tools in your toolkit. Sequence is one of the big ones. Because it's not uncommon in a game that you've got to make things elastic enough that a player can experience them in whatever order. We do it. We don't usually do it in a way that's carrying a lot of water from a storytelling point of view, just because it's hard. Again, that's not how the human brain evolved. We're not used to telling stories in a nonlinear fashion.

Tom Abernathy (00:24:12):

But what happened was, I think, as Novera thought about the story that was going on in this season, which is that one of our elder dragons, Jormag, is whispering to people in the world of the game because Jormag wants to wake up and Jormag wants an army of people there to do what Jormag once, once Jormag is awake. I'm saying Jormag all the time because Jormag is a nonconforming dragon gender-wise. Jormag is fascinating character. I really love them.

James Gwertzman (00:24:55):

Keep going. I want to know what's going to happen with how do you tell a story in the ...

Tom Abernathy (00:25:00):

Jormag can whisper into people's minds, basically. The way we decided Jormag operates is it's not a Jormag makes you want to do a thing. Jormag persuades you to want the same thing that Jormag wants. It's very soft skill. What happens in the episode is as you're moving around the world, whether you're doing instance content, which there isn't a ton of, or you're in the open world in various contexts doing things, you're hearing these whispers sometimes, and you're certainly seeing other people who clearly are hearing them and don't know what they are. There's this growing sense over the episode which starts out very fun and rollicking. Yeah. We're all going to have a giant Burning Man party.

Tom Abernathy (00:25:51):

There's this growing sense over the episode of dread, because something is not right here. There is something that is corrupt at the heart of all of this and it's not at all clear what it is. But the whole world is telling you that bit by bit. It doesn't matter ... The thing that works great about it is it doesn't matter who you hear in what order because it all ... it's story content that accrues rather than unfolds in a little ...

James Gwertzman (00:26:16):

Intentional fragments. It's intentional fragments and intentional little snippets.

Crystin Cox (00:26:22):

Exactly, exactly. I think, in that sense, I felt it was the most successful episode that we've done since I've been here. We've done some that I'm super proud of. But that, in other words, adapting to ... having our storytelling techniques to suit the specific very game specific and GW2 specific needs in terms of structure, in terms of the mechanism. The delivery system has been a big challenge, but it's a fun challenge. I've been amazed at how our writers and our designers have come up together content ... have come up together with ways to do it.

James Gwertzman (00:27:03):

I think that really touches on something that I'm fascinated by it. I've been fascinated by, basically, since working in MMOs is how a community of people can get engaged in something for many, many, many, many years. You're talking about this specific episode, and I saw lots of chatter about that episode when it dropped. You're just out in the MMO community of people who play MMOs. I think that really feeds into a place that I don't think we've spent enough time talking about. You've touched on it where narrative such important part of context. But having something for the community to talk about is such a huge part of being able to stay connected to something for a long time.

James Gwertzman (00:27:52):

I actually think that MMOs have had it going on for a long time. I mean, gosh, I can remember having our long conversations about Everquest, which in some ways seems very silly. It's not about even that. There's a very light lore layer there. But I think that it was been doing it for a long time, but you're actually starting to also see film and television come around to, man, what really matters is the fan community and what they're engaged with over a long period of time.

Tom Abernathy (00:28:24):

I think that's right. Within the context of making an MMO, without going to details, I'm sure that anybody who's listening understands that any other entertainment business, A, you need to make a profit, but, B, sometimes those margins can be relatively thin. Particularly when you have a product that is designed to be free to play at its root, and we give you ways to give us money for things that you want at various points. Nonetheless, it's figuring out how to stay afloat is a thing. It is easy, I think, from a systems design perspective to come up with some systems that will drive activities and events that people will enjoy grinding on a bit and all that kind of stuff. There's some people who that's really all they care about, which is fine. It's harder to ... You can measure though that kind of stuff using gameplay metrics and data that you collect as players are playing the game. You can measure how long they're spending in the game, how long they're spending doing what, what things they achieve, all that kind of stuff.

Tom Abernathy (00:29:42):

At this point, it is harder to quantify the value that narrative brings to that experience. I don't think it's impossible, but I just think that not very many people have yet tried to design user research questionnaires, or metrics that would necessarily track that in some meaningful way. It's a little fuzzy. Sometimes we have debates and there are people who are skeptical of exactly what the kind of value is that narrative brings to the experience. My feeling as you know, Crystin, and it pairs up with what you're saying is that ... and this is anecdotal, but boy, I just I've seen it all over the place that even players that I know who play the game primarily just for the grindy stuff.

Tom Abernathy (00:30:39):

They just like doing raids with their friends or whatever. They still talk about their player character. They still buy stuff for their player character. They still get online and talk about Bram and how he's grown, or about Caithe and her relationship to Aurene and she reformed is she no longer a bad person in some of the ways that she seemed to be earlier on or what's ... I mean, what's up with Marjory and Kas? How come we haven't seen them? Wow, now they're back. But they seem they've been through a lot of stuff. They seem rocky. What's going on there? That stuff matters to people for whom it's not even the point. It helps create, as I said before, it helps create the emotional connection between the player and the game no matter what it is that you love doing in the game. I think the vast majority of players, that's true.

James Gwertzman (00:31:24):

I think, in the early social mobile days, I think we saw a lot of games that would take systems that were designed to perhaps create monetization or retention, and they would put them out there so nakedly, and so transparently, that I think is a real turnoff for players. It's login every day. If you log in seven days in a row, you get this thing. There's no attempt to dress it up at all. It's here's a little check box and seven days in a row and you get the thing. It's like, "Wow, that's a pretty naked retention mechanism," as opposed to everyday the spell grows and if you were ... you've ever bought pot, at least The Farmville is like, "Okay, you got to water this thing every day. You've got to do something for a reason." That motivation, I think, is what sells it ultimately. I think context and narrative as motivation to your earlier point, I think really does help dress up a lot of these systems that otherwise are just too transparent.

Crystin Cox (00:32:17):

I think that among game developers, it's not surprising that each of us is ... that we care most about the thing that we love and that we do. Narrative people are drawn to game narrative because they like narrative. They like the context and emotion and the motivation. Designers, a lot of them are drawn because they love doing stuff and making stuff to do and making it fun and exciting and getting that adrenaline loop going. People obviously who are artists are attracted because they like pretty things and making beautiful stuff and environments and clothes and all that stuff, all those things.

Crystin Cox (00:32:53):

I think it's understandable that sometimes folks from other disciplines don't necessarily perceive the value of narrative to a lot of players in the way that we who do it do and in the way that I think it's a lot more obviously easy to understand the value of, say, environmental art. Because if you don't have that, then you're in a gray box and there's ... I mean, there's no context at that point physically. But I do. I've always believed strongly ... and this came from me as a player. All of the reasons that I wanted to get into game narrative, and worked so hard to get into game narrative even at a time when there was virtually no such thing was because I knew as a player, that I latched on emotionally to some things in games and to some games in the same way that I did as a film student with some movies. If television had been better than it was in the mid '90s that I would have probably done with television.

Crystin Cox (00:33:58):

Just for example, when I was in film school I would play Twisted Metal at night. That's a game that's got virtually no narrative context whatsoever. But it's got all these giant nutty vehicles in a arena, and you're just doing a monster truck, Bumper Car thing, and it was so much fun to imagine, just to imagine if you're playing this vehicle, who's in that vehicle? What's their life like? What environment do they come from all that kind of stuff?

James Gwertzman (00:34:27):

What's the motivation?

Tom Abernathy (00:34:30):

Yeah. Yeah, exactly. We look for meaning. We know that about the human mind or the human mind looks for patterns. It looks for meaning. It does not want to believe that chaos is just chaos. It wants to believe that things mean something. We feel, I think, if we're doing it right, that we're the ones principally on any game dev team, for whom that's the job is figuring out what is all mean? Why am I doing this stuff? Designers are the "what," environment artists are the are in the "where," and designers is the "how." As well we're the "why."

Tom Abernathy (00:35:07):

I just fiercely think ... I don't ... again, it's hard to come up with metrics. You'd have to really do some intensive user research to come up with some ways to prove this. But I've seen that done to an extent. What I'm saying is the thing that I feel intuitively, but also I have seen it borne out in some data, that the vast majority of players regardless of the kind of players they are, this isn't the way we categorize different players. They all want meaning out of the experience of one sort or another. That's our job.

Crystin Cox (00:35:42):

I think about this a lot now, because I spent a lot of time talking to traditionally AAA teams as they try to make a transition into LiveOps. One of the ways that I often differentiate to them about, what's the transition you are actually making? Is in AAA, it's very developer-centric, It's like we're going to bring to you our point of view, primarily. In LiveOps, it tends to be more player-centric. We think a lot about what is the player want, and how can we meet those needs? I'd love to hear you talk a little bit about that in narrative because it's certainly not the only place where that's ever been true. In narrative, there's some long traditions of that being true narrative. But I think it's a place in games that can be challenging for narrative to wrestle with their relationship with the player, who is actually ... usually millions of players with lots of different desires?

Tom Abernathy (00:36:37):

Right. Yes. Again, I mean, this is one of those things that is specific to games. You don't get that challenge in any other medium, really, at least, not much. The reason is to challenge, just to quickly illustrate for folks. It's probably pretty intuitive. But I remember where [inaudible 00:37:00] article about this actually. I remember when Mitchell Hurwitz announced that he was going to be doing a fourth season of Arrested Development for Netflix. It was a big announcement with lots of fanfare and he said they're going to be 15 episodes. You're going to be able to go through them in any sequence that you want. It'll be great because you watch them in one sequence, and then you'll see a joke paid off that was planted, but you'll see another a payoff that you haven't seen the plant yet, and then you see the plant later, and you'll get how it works. You watch them again, in different orders to see all the jokes and all that kind of stuff. I thought he was crazy and didn't understand what it was he was saying.

Tom Abernathy (00:37:42):

Indeed, that turned out to be true, because about nine months later, a few months before the series actually dropped, he once again emerged into the media and said, "Okay, first of all, they're going to be 12 episodes, not 15, and second of all you know how I said you could go through in any order you want, forget that. There's going to be an order we want you to go through." I just laughed as a fan of that show for sure and someone who's excited about new Arrested Development at the time. I laughed though, because, I mean, I'm a writer, I'm not a mathematician. But I know how to do factorial multiplication. If you have 15 episodes, and you're telling people you can then go through them in any sequence they want, that's 15 times 14 times 13 times 12, et cetera, et cetera. It's over 3 trillion possible ways to go through the content.

Tom Abernathy (00:38:36):

One of the challenges that exists for us and it's much more ... you feel it more keenly in an MMO context as you say, than you do in a AAA ... even nowadays, contexts like God of War, Spider Man or something like that control, is that you want people to feel that they are having a bespoke experience, but there is only so much you can do to create a bespoke experience for them. Because otherwise, you're creating a ton of content that a lot of players will never see, which of course all game devs hate, and you're just setting yourself up to have to create so much content.

Tom Abernathy (00:39:19):

What you're talking about, I think, is the answer that dev teams have hit on to try to solve that problem, which is I'm going to create, like you said James, I'm going to create [inaudible 00:39:30], I'm going to create a nightclub in which everybody hangs out and they can ... I'm going to put stuff in for them to do. You can dance. You can go do coke in the bathroom. You can go say hi to Michael Jackson and Brooke Shields over here. I'm going to put some in the world. Again, if you're doing it right, all of these things, the cumulative effect of them, and the cumulative impact will be that I grow to love the world and hopefully the people in it and the stuff that I get to do. But it's not telling a story in the way that human beings have done for thousands and thousands of years. It can't be that in an MMO setting, because that's far too limiting.

Tom Abernathy (00:40:12):

I think there are some things you can do to incentivize replaying somewhat linear narrative content. But I don't even think that's the most interesting space. I think some of that is good, because some players really love it. It is the easiest way for us to move plot down the line. But I've always felt and have some user data improve, actually, from Microsoft that plot was a bit overrated as a value in games. When I was at Microsoft, some of the UR folks there did some work that underlined that and validated that intuition pretty strongly that people don't really ... players don't ... if you ask them what the story is of the game they played, they don't really tell you the plot of the game so much as they tell you their user experience moving through it.

James Gwertzman (00:41:03):

I was just going to say that's a segue. I've been watching a lot of ... I've been catching up on Westworld recently. I'm now fully caught up. Of course Westworld has a lot of interesting analogs, especially for people making MMO RPG because it's basically a giant, real world MMO RPG. I think one of the interesting things about that show is the external narrative in the park, you never really see. You get hints of it. But it's clear this is a backdrop. All the really interesting narrative of the show, I was getting a little meta, the narrative of the show is all inner character narrative. It's sort of the character arcs and what's happening to them, within these people's heads, the park itself is it's backdrop. Actually, that's not a terrible analogy here, which is what you're creating these opportunities for players to have experiences where they're creating a narrative in their heads where they can tell go back and tell their friends, "Oh my God, last night, I did this thing and this thing happened. It was amazing. Then it did another thing."

James Gwertzman (00:41:54):

They're creating a narrative. Every player is going to have a slightly different narrative. They're telling of their experience. What's actually happening in the game is really just opportunities for those internal narratives to happen. Multiplayer games are especially ... League of Legends, I frankly not a League of Legends player, full disclosure. I go to these tournaments and I watch people playing on these big screens, and it's complete [inaudible 00:42:15] to me. I have no clue what's going on. But all of a sudden, everyone will gasp and something obviously happened and I have no clue what that is. But it's like in chess. It's a series of ... it's a structure that it creates opportunities for those kinds of moments. I think that's actually really fascinating when the game comes together to allow a person to build a narrative for themselves in a broader context.

Tom Abernathy (00:42:36):

Yeah. I think that's exactly right. The League of Legends example is terrific. I was at Riot for a year and worked on that game back in 2013. The thing I was brought in, because they didn't want storytelling in the game, it's not a game that lends itself to storytelling, but they want it to figure out how to leverage the thing that was the big value for them, which is the champions in the game. Because the Microsoft user research that Deb Henderson did that I was referring to, as it found that players don't remember plot very strongly.

Tom Abernathy (00:43:08):

What it also found was that ... what does stick with them is characters, NPCs, who they love, or in the case of League of Legends, PCs, who they love, because all champions in League of Legends are very specific. I mean, they're not blank avatars. They're extremely specific and 1,000 different ways. The champion creation, the development and creation process at Riot is the one thing for sure, when I was there that I would say, unquestionably, they have so nailed down. It's a wonderful assembly line that works five times out of six to come up with an amazing character for players to play.

Tom Abernathy (00:43:46):

You're exactly right to say with League of Legends, even for players who are mostly all about the jungling or whatever they love the champion that they play for that. They love the backstory that goes with it. They love the lines that ... it's characters that matter. I think that is the saving grace for games since we give up so much in terms of linear story. What we give up though, we get back in terms of people that you meet. Or in some cases, like League you get to be. You get to play act as, whether it's a character you create or a character somebody else created, who's very defined. The social aspect is the thing that drives MMOs on so many levels and being able to take advantage of that, figuring out ways to take advantage of that in an MMO context is part of the fun of working in it, for sure. Huge part.

Crystin Cox (00:44:43):

I will say though, I just want to jump in and I'll make this point that it can be very easy when you're designing a game to think, man, it really matters is the stories players are going to tell themselves what they did. I'm just going to put a bunch of stuff in the game, and then step two, question mark, step three, profit, is a very simple, a very common thing to happen. Especially when I talk to devs who are thinking about putting a lot of social content in the game, there's a lot of step one, social, step two, question mark, step three, fantastic, amazing game. I will say, especially having worked with some great narrative designers, yes, it is a lot about the stories the players tell themselves, but man, does it take a lot of work to generate that experience.

Tom Abernathy (00:45:35):

Yeah. No. Absolutely. You're right. Again, I mean, if it was easy for players to do on their own, then we would just be creating kits that allowed them to do that, basically. There are some games that play like that in some ways. But I think that's exactly right. I mean, why have professional trained, experienced narrative people working on your LiveOps game to begin with? Well, it's because doing things with other people is what the experience is all about. Doing things as groups frequently is what the experience is all about. But the thing you are doing matters. This is the point that I agree with you, I think, gets lost too often, particularly in a LiveOps setting is we just need to design some grindy systems, toss them out there, everything will be great. No. That's not how it works. Stuff has to mean things. To most people, it really does.

Tom Abernathy (00:46:30):

Allowing the player to be motivated by the fact that Marjory is freaking out thinking she's seeing her dead sister there attacking her and she has to kill her. That's the thing that players who've been playing Guild Wars 2 since the beginning are just going to ... I mean, oh my God, the emotional impact of that is enormous. I remember you were still there, Crystin, when season four started, and we put out the episode where it starts off this light Pirates of the Caribbean romp with [inaudible 00:46:59] the sly and all that stuff. You get yourself thrown in jail so that you can bust [Zaim 00:47:05] out of jail and you're busting him out.

Tom Abernathy (00:47:09):

Then all of a sudden you hear from the other room where the warden confiscated your comm., you hear Taimi, who's a character in the game in the Sora Small, very smart young. You hear Taimi on the comm. going "Commander, commander, are you there? I need you commander." You're like, "What the fuck?" As you listen because you can't get to your comm. yet, you hear her basically saying she's been kidnapped and she doesn't know where she is and she's really scared. Then she starts screaming because she's seeing something horrible happen. Then the villain of the first three episodes, Joko, the Lich King of Elona comes on the comm. and says, "Hi commander, have I gotten to you yet basically, come and get her if you can find her."

Tom Abernathy (00:47:59):

We played that all very straight. We didn't play an arch or anything, because what we wanted was, in a larger sense, we wanted to signal to GW2 players that we were from now on going to be willing to take some tonal risks that the game had not ever done before. We were going to put some characters that they really loved. Taimi is very high on that list. In actual Jeopardy, not fake Jeopardy, literally, the stakes are very high here. Many players, I remember that day, when that episode came out within two hours, there are people on our forums and on Reddit, saying, "Man when I heard Taimi's voice, and I heard that Joka was killing people in awakening them right in front of her basically creating ... killing people, creating zombies out of them right in front of her.

Tom Abernathy (00:48:48):

I couldn't see. I couldn't think. I was blinded by my need to find Taimi and save her." I remember who was, but somebody came up to me that day and said, "This feel a little validating for you." I said, "I'm not going to lie." It kind of does, because people were nervous a little bit and I was. I didn't know if Guild Wars 2 players would go there with us, but they really did. We've just built on that ever since. The thing that is my overwhelming conviction, that powers those kinds of choices is that people just want you to give them characters to care about and give them a role in those people's lives, and everything else has details.

Crystin Cox (00:49:32):


Tom Abernathy (00:49:32):

You know.

Crystin Cox (00:49:32):

I think that that moment, just to dig into it a little bit is really illustrative of that point, which is, there's a lot at the beginning of that episode, which as you said, is very light and very fun. There's so much work behind the scenes to get the player to a place, even though it doesn't feel like you're really doing much. But to get the player to a place where that hits, where that moment isn't incredibly impactful and that I think is the thing that is so challenging about it. But, I think, I would love to see more LiveOps games really dig into it because it really can be so impactful, especially when the player feels like they have that autonomy. It doesn't feel it's being told to them. It feels it's something they're experiencing and living through.

Tom Abernathy (00:50:20):

Yeah, absolutely. We find that over and over again. Again, we'd like ... We look for ways to keep us honest, to make sure that we're not backsliding too much, that we're not depending, leaning too heavily on traditional ways of storytelling that games can give us other things to do with. For a great illustration of that, I think is the ending of our fifth episode last year, all or nothing, which is an episode in which you and 1,000 other people with factions and your favorite dragon, Aurene, who you've raised from an egg are all basically getting ready, girding up for this fight where you're going to lure this elder dragon called Kralkatorrik who is insane, basically, and wreaking havoc not only on Tyria, but in the midst, which is the ... that's a different place than Finn here. I won't go into it.

Tom Abernathy (00:51:22):

But you're luring him into this cave, Kralkatorrik, and you're fighting him, and we had been telling people for a long time because Glint, Aurene's mother was a dragon of prophecy, that what we call Glint's legacy was that her progeny needed to kill Kralkatorrik. Aurene got her gift of prophecy and can see all the different ways in which a thing potentially can play out, at least up until the moment of her death. There's cinematic in episode four, I think, where she does just that. What she sees is that no matter how the vision, no matter how the battle plays out she's going to die at the end of every single scenario she dies. There is no Tony Stark 6,000,006 or whatever it was. She's going to die always in the same way, always in the same pose.

Tom Abernathy (00:52:23):

We told players that over and over and over again that she foresaw that, that there was a thing that was going to happen. But of course they didn't believe it. Then you play the episode and at the end she dies. Our assumption, well, I mean you know cut to the chase, she dies. But our assumption was that we're going to have to do it as a cinematic, because the idea, we wanted you to ... there's been a huge ... you've been knocked way back into this cave and your days, you're hurt, you're disoriented and we figured all of this stuff we wanted to do with everybody's reaction to Aurene being dead was the thing that's going to have to be a cinematic.

Tom Abernathy (00:53:07):

We have an extremely talented designer named Cameron Rich who thank goodness alone among all of the designers and narrative folks thinking about this sequence, thought to himself, "You know what, I think that I can do this in game, limiting some things, limiting the camera movement a little bit, limiting the speed that the player can move at. But I think we can do this in game. Of course, it will be better if you experience it, instead of just watch it, because everything in games is better if you experience it instead of just watch it." He mocked the thing up in about two and a half days and called us over and showed it to us. It was recognizably what you see in the game now. I mean, obviously, that version is a whole lot more polished. But it was amazing.

Tom Abernathy (00:53:52):

I even had one or two narrative people who are standing there who like, "No, it's not going to work as well." I was like, "No, no, he's absolutely right." Because whatever we lose in control over the presentation, we more than make up for in the players emotional response being jacked up to 11 because they just experienced it for themselves. They had to limp all the way through that 50 yard cavern back into the space where Aurene and everybody else's, and hearing people crying, hearing people upset not knowing what it means. But again, that sense of growing dread as you get closer. Doing that yourself is different than watching it happen to somebody on screen. It was just brilliant. It's a lesson that I now always keep near the top of my back pocket to always say if there's any way we can do this in game, we should do that, it's going to be better, it's going to have a bigger impact on people.

James Gwertzman (00:54:45):

Having just talked about two really, sounds a phenomenal narrative moments that worked really well. This might be a good segue into our favorite part of the show, which is our LiveOps train-wreck failure part of the show. We asked you to talk about times when it didn't work well. I guess in the context, since we're talking about narrative in LiveOps here, maybe can you tell us some examples where you perhaps tried to apply narrative to a LiveOps scenario and it just fell flat and it other had the unintended consequence or it just didn't work at all?

Tom Abernathy (00:55:11):

That's a really good question. I would love to be able to give you an example of that, I'm not sure I'm going to be able to come up with one simply because I don't think we were doing it very well, until this season, I think last season, and coming out of Path of Fire, which was our expansion pack before season four, which I arrived in the middle of. I think we were thinking of the storytelling in the game as a thing, which was much more akin to a television series and the way that that content will normally unfold. It is the new circumstances that we all as a large team wanted to pursue. We wanted, like I said earlier, to incentivize players to spend more time in the game.

Tom Abernathy (00:56:10):

Our focus shifted a bit and in a more proactive way than I've ever done before really started thinking about different kinds of things that we can give the players to do, that still will carry narrative value and progress our character arcs at least. But that are less linear ... We're approaching even less linearly than we used to. I have no doubt there are ways in which well ... okay, what I'll say is I do think we've done things like that I question for me as a player whether I would find them that interesting. I mean, we had some Joko Open World Meta events after episode three last season that were basically ... I think the idea was ... the contextualization was somebody is pretending to be Joko who does a really good Joko impression, I guess. He's getting on I kid you not.

Tom Abernathy (00:57:12):

There was a microphone in a little room somewhere in the world and the idea was the janitor or somebody was getting on the microphone and going over the PA system pretending to be Joko and convincing everybody to do all these things. The gameplay part was fine. I don't particularly feel that contextualization made much sense or was very effective. He had some good lines there. He had some decent Joko lines, but not really, because real Joko would be much funnier, and more stinging. I don't know. The whole thing didn't quite rise to the level to me of what I'd like to see.

Tom Abernathy (00:57:50):

Of course, my perennial example actually is Dolyaks. Dolyaks are a thing in Guild Wars 2 ... I mean, they're sort of yak-esque beasts, I guess, of burden, oxen or something. There are all activities at times that you will do with Delyaks, hurting them, checking out their feces, feeding ... I know, I get it, but at the same time like, man, isn't there some way to do something better than that for this? I mean, I know it's a FedEx quest or whatever it is. There's certain very common kinds of activities. We're all familiar with it, that people will create gameplay experiences out of. But I don't know that we've always done a stronger job contextualizing them as I think we could. I'd say as a general note, that's what I'd like to do better.

James Gwertzman (00:58:43):

Actually, I realize I'd several more question for you, which is an observant jumping Crystin. Especially Asian MMOs, I'm always struck when I read their live event descriptions in the really wide gap between what I'm calling the reality of the event and fiction of events. The reality of like live event for this event, the probability of this one drop goes from 1% to 3%, and the risk of this failure case goes from 18% to 22%, and we're adding a new drop. It's these very technical descriptions of minor changes or tweaking for the duration of the event. But then in the fiction is ... it's the Night Moon Saga. Dragons are rising from the east and all these sort of very flowing pros for it essentially changing three probabilities and adding one item. I'm curious about if you have any thoughts about that gap between ... and apparently it works because players are not complaining. The gap between the fiction you cloak around the thing which itself might be in some cases, often very prosaic?

Tom Abernathy (00:59:45):

Yeah. Yeah. I mean, I don't know for sure. But I would guess that a lot of the players who are playing an event like that know they're doing it for grindy reasons, know they're doing it for the stuff they may get or whatever. They don't necessarily in that moment, care as much about the contextualization of the actual event they're doing. I don't know that for sure. But I wouldn't be surprised if that were the case.

Tom Abernathy (01:00:07):

I think part of it is that we don't get to work very closely with either designers or particularly folks on the reward side in even doc ... right now, up to documentation, the notes to go out with that stuff. I think that if I got to make the decision, and we would need more writers to be able to do it to have the bandwidth to do it, but I'd love to have a writer who is basically embedded with the team that comes up with all items and rewards and somebody embedded with the systems designers working on events. Because we work most closely with content designers who are frequently doing content that has a fair amount of narrative and they are people themselves who have a lot of affinity for that.

Tom Abernathy (01:01:00):

But those other folks that are generating the stuff that you're talking about, we don't have a lot of interaction with them, not nearly as much as I wish we did. I do feel like if we were to, it would ... we could subtly transform the presentation of all of that stuff into a thing that always maintains the fiction, in a way. That's what I'd love to see, frankly, is a thing that never just goes, "Yeah, we're a game." Not that anybody ever knows you're not or doesn't know that you are game rather, but maintaining that illusion of immersion. I think we could do that more effectively, particularly in the areas that you're talking about, and I would ... it would make me happy if we could do that. But I think to some degree, it's also a bandwidth issue.

Crystin Cox (01:01:59):

Yeah, which I mean I think it brings up one of the harder questions about all this. You touched on it through this whole conversation, which is the cost, narrative can be very expensive. I think we're past the point in the industry where people are like, "Ah, we'll just get someone on the team to write all the ... to write the lines or whatever." But even in those situations that you're talking about, there is still some of that because you can't really have a high quality narrative designer touch literally everything.

Tom Abernathy (01:02:33):

Right. That's right. I will tell you, honestly, I don't know that we're as much past all of that as I wish we were. Again, I think that's mainly because aside from whatever it matters to devs as players, it's the case for the value of narrative. Well, I think everybody ... I think you are right. I think most people nowadays would acknowledge that there is value to narrative as a discipline in games, but I think there's still a lot of disagreement about what exactly that value is and how much therefore it should be underwritten. That's why to be perfectly honest, I've become a lot more interested in the truly LiveOpy side of what we do because it's become clear to me that we need that aspect of the game to really thrive in order to have the resources to do, to bring to the experience the things that we want to bring to it and that we can bring to it. You always, of course, just have to work within the means that you have.

Tom Abernathy (01:03:46):

But I'm a big fan right now of helping systems designers. I haven't come up with good contextualization for the stuff that the systems designers are coming up with, so that it will help make us more money, so that maybe I can get another head at some point down the road or whatever. But also so that what they're doing feels it means something, it matters, it fits in the context of the world and the stuff that's going on and the characters that we know and love. I know, as a player to me, everything is more exciting when you do that. I mean, I think about it and I can name names, but I won't. But I can think about a couple of games, a couple of MMOs that I play sometimes that have some superficial trappings that are very much I'm very much into. But that make no effort whatsoever, or a one percent effort, I will say. You said, there are lines. There are four or five lines. They're not good. They're not well-written. They're not particularly necessarily even well-performed. They're certainly not interesting, but they may have five lines.

Tom Abernathy (01:04:49):

But other than that, there's no contextualization. You're just going to a place to do a thing. As a player, and I recognize not everybody's like this. But as a player just going to a place to do a thing does not motivate me. You're not going to keep me in the game very long if that's all that's going on. You can give me a reason. If you're connected to somebody in the game that I care about, now we're talking something that's going to keep me on my couch for a long time. I like to tell people, Alyx Vance, in Half-Life, I would have done anything for that woman. I swear to God. I was so in love with her. I'm scared to play the new game because I'm scared of what might happen. I know from personal experience how strong the emotional attachments you get for good well-written, well-performed NPCs are and I want to leverage that.

James Gwertzman (01:05:37):


Crystin Cox (01:05:38):

Thank you so much for doing this. This is really interesting. I'm super happy we got to have you on. We haven't talked really at all about narrative. I think it's a big topic, and something that's on a lot of people's minds for LiveOps.

Tom Abernathy (01:05:52):

Yep. Yep. Well, I appreciate you asked me to do it. I was really excited, actually, when I saw your email because I figured probably that you had a lot of people talking about a lot of other things.

James Gwertzman (01:06:02):

Data Analytics and ...

Tom Abernathy (01:06:03):

Yeah. Yeah. Yeah. But I mean, we were about the "why." We create the soul of the thing. I think that matters. It has been one of the really true most enjoyable aspects of my professional life to learn over the past three years. I did a little bit on League of Legends. But again, we were trying to storytelling that game, in the game per se. But for the last few years working on Guild Wars 2, learning about LiveOps, learning about what it's like to make MMOs and then to develop content that drops on a regular cadence and all of that stuff, I mean, it's opened up a whole another side of how you can go about doing narrative in games, and I've learned so much from so many incredible people. I appreciate the opportunity to come on and just think out loud about it.

Crystin Cox (01:06:50):

Yeah. Yeah. That's great. Oh, man. Yeah, thank you so much again. Thanks for listening to the Art of LiveOps podcast.

James Gwertzman (01:07:02):

If you liked what you heard, remember to rate, review, and subscribe so others can find us.

Crystin Cox (01:07:08):

And visit for more information on solutions for all your LiveOps needs.

James Gwertzman (01:07:13):

Thanks for tuning in.