I’ve already talked about the two talks I went to on Saturday that were about Mass Effect 2; here are my notes on the rest of the day.

10:30 am: Motivating Casual Players: Non-Traditional Character Progression and Player Retention, by Laralyn McWilliams.

Another fabulous talk, marred only by the less-than-fabulously-legible quality of my handwritten notes. (The speaker plans to post slides soon, fortunately.) The talk was about Free Realms, a MMORPG for kids. It was designed by people who had been working on EverQuest, so they had a decade of experience with what an MMORPG should look like; they took that experience and designed a stripped-down, kid-friendly version.

Which was, in her words, a total fail. As they discovered, casual players approach games differently from core players: among other things, they

  • Have other interests, only devoting a small amount of time for games.
  • Don’t have a lot of experience with a broad spectrum of games, they’re not always expert with controls.
  • Will stop playing at the first barrier.

(She gave more positive characterizations of casual players later.) Fortunately, they do have a play style which her team managed to alter the game to match; that play style turned out to be consistent across ages and genders, interestingly.

She then turned to a discussion of progression. First, some broad characteristics:

  • Progression isn’t the same as accumulating XP; more generally, progression means something different for casual players than it does for core players.
  • They should reward all playing styles fairly.
  • They wanted to respect player’s time, effort, and money.
  • Progression should be fun: there should be a reward for winning, there should be a sense of accomplishment even if you don’t win.
  • Progression is tied to self-worth; players should develop skills and feel special and/or smart because of that.
  • Progression is good for revenue: you want players to want to play again!

Next, she delved into three specific types of progression, namely character progression, player progression, and social progression.

Character progression involves accumulating experience points, using those to level up and gain skills and abilities, unlocking items as you level up, and it persists from one game session to the next. At launch, they had a class system; players could change classes as they wished, and leveled up separately. The reward for leveling up was “stars”; in combat jobs, you could spend these on your abilities. (I also have something in my notes here about “25% minigames, 75% quests, based on later notes, I guess that’s what sorts of actions were required to level up?) There were also quest rewards and loot drops. (Unlike traditional MMOs, though, everybody gets a loot drop.)

This didn’t work well: it turns out that casual players didn’t care about levels. Fewer than 10% of players were above level 10 in any job; what’s more, even when players accumulated stars, they didn’t spend them. Oops.

So they changed the system. They replaced the levels with progress bars. They got rid of spending stars entirely. They made level locks much clearer. And they rewarded people for having fun, not for doing extra work: you can pull up a map and click on whatever it is that you want to do, and you can get rewards for doing minigames.

The result was that players would level up without really knowing that they did so; as she put it, “if they don’t care, I don’t care”.

In the first incarnation, they had a lot of stats, covering everything; you could, for example, have stats that improved your ability at match 3 games. This turned out not to make a lot of sense to people; they got rid of non-combat stats, and significantly simplified what remained. The combat wearables had one number, a “power rating”; weapons had a power rating plus two abilities. (As she mentioned later, showing players that their abilities improved by a few percent each level didn’t help—it’s fine to have that behind the scenes, but if something isn’t making a big difference, then don’t bother telling players about it.)

As to rewards: before, players didn’t know about them. They changed that so each quest / game would show you the possible rewards in advance, and you’d spin a wheel after winning it.

Next, social progression. This can express itself in terms of hosting events, being in guilds, being in chats. Free Realms launched without a lot of this; and the Child Online Protection Act meant that free chat was impossible. But what few social spaces they had turned out to be quite popular; in fact, when they looked at the data, the realized that the few social spaces were more popular than all of the minigames combined!

So they formalized this by adding “party zones”, and looked for ways to help people show off. Before, people could customize themselves through their choice of weapons, through their job clothes, through just-for-fun clothes, and through toys and tricks; it turned out that, of these four, the most popular was the toys and tricks, then the fun clothes, with the more “pratical” gear running a distant third. (She used a banana suit as an example of a popular costume.)

Another issue where socialization came up was in terms of formal grouping, e.g. joining a party in order to do an activity together. It turns out that players wren’t doing this at all; their mental models seemed to be that, if somebody was on their friends list, then they should already be grouped, and if players were standing near each other, then they assumed they were grouped. The designers reacted to this by adding lobbies to games, and by handling all the grouping and un-grouping behind players’ backs.

Finally, player progression. Examples here are achievements (which she related to high scores in an arcade: those are your initials on the high-score list, not your in-game character’s initials), and a profile page showing all of your characters. They also had a way you could design a trading card out of one of your in-game characters, turn it into a physical object, and give it to a friend!

They discovered that people really wanted to express themselves as real people, not as the characters they played in the game; this is really important, much more so than character progression. In fact, it’s important to the extent that the two ended up merging: casual players overwhelmingly have their avatar be the same gender as themselves, with a similar skin tone and features. (In contrast, in EverQuest 2, only half of the players create a character with the same gender as themselves.) Basically, people are creating avatars, not characters.

After this, she took a slightly more theoretical turn, discussing the psychology of progression. They designed progession by thinking about characters needs, the interactions that they would have, and the rewards they would receive.

They could segment needs by market:

  • Core players want personal skill and information.
  • Casual players want wealth, social status, and individuality.
  • Both want relationships, ownership, entertainment, and a sense of progress.

They didn’t segment interactions by market. But they did segment rewards:

  • Core players want a leaderboard, knowledge, character leveling up, and useful items.
  • Casual players want money, profile popularity, appearance, and socialization.
  • Both want friendship and fun.

And then she went to mix and match them: e.g. somebody wants to improve their relationship, they trade with an NPC, and get rewarded by an improved appearance.

To sum:

Casual players don’t:

  • Care about stats.
  • Care about equipment. (I think, I might not be reading my handwriting correctly.)
  • Make traditional comparisons (e.g. levels) with other players.
  • Want to work hard for a single reward.

And casual players do:

  • Value differences in appearance.
  • Enjoy tasks that unlock new content.
  • Want to feel special and unique.
  • Like frequent smaller rewards.
  • Understand large changes in effectiveness.

3:00pm: Metaphysics of Game Design, by Will “Phaedrus” Wright.

The identity of the speaker wasn’t published in advance; glad to hear that Will Wright is also a fan of Zen and the Art of Motorcycle Maintenance. I won’t even attempt to summarize the talk here: basically, it was a microtalk given by a single speaker going at full blast for an hour straight.

And then the formal part of the conference was over. Michael Abbott and I plotted about the future of the Vintage Game Club at the Samovar Tea Lounge; many thanks to Jorge for introducing me to it! And I’m quite excited about the future of the VGC; I hope that we’ll have something to announce in a week or so. After which I had one last GDC dinner, with Michael, Ben, Nels, Wes, and Matthew. (I hope I’m not forgetting anybody; GDC is starting to blur…)

Honestly, the conference started out a bit subdued to me. But I went to some really excellent talks over the last couple of days, and had many pleasant conversations; I’m still starting to figure out what it all meant to me (and may do one last post putting it together), but it’s left me just wanting to sit down, think, write, and read, and that’s all to the good. To everybody whom I met, it was great seeing you; and I hope to see even more people next year!

Post Revisions: