MMO designer Raph Koster reveals ‘Trust Spectrum’ research for online games


If you’ll be at GDC this week (we will!), you’re in for a treat, as research from MMORPG designer Raph Koster will be on tap.

It’s new design framework aimed at co-op multiplayer game designers, conducted as part of Google’s Advanced Technology and Projects (ATAP) group under Aaron Cammarata. The team is calling it the Trust Spectrum, and as Koster explains on his blog, the idea was to study how trust impacts games and vice versa, specifically for the purpose of building games that make sense for the level of trust players have for each other – and then building games that actually push people along the trust spectrum in a way that makes sense.

What they found in digging through games and gamers of all stripes was that “virtually all games are actually played at all levels of this spectrum; meaning, you can play competitive games with friends or strangers, a bidding system or supply chain system may exist at any point on the spectrum.” Ultimately, the investigators were able to map features across a trust range to make predictions on everything from audience size to retention.

It’s a massive piece with so many implications for online games that’s definitely worth a read – more for our developer audience than players, but we suspect MMO players will recognize themselves (and their guilds) in the piece. We’ll be at GDC this week and gunning to make this talk, so stay tuned for more!


No posts to display

newest oldest most liked
Subscribe to:

That was a great read! I always learn some great new things when reading Raph’s work and I love seeing this sort of “behind the scenes” work that goes into designing games.

Kevin McCaughey

Very interesting and just the type of thing we should see more of in MMO-land. Games should be designed with reference to research like this rather than developers forcing their own flawed systems on people. We would have more successful games on the back of it, and that is what we all want to see.

IronSalamander8 .

This is some very interesting stuff. The class thing was especially intriguing. I don’t mind classless systems but I tend to prefer some kind of classes myself, even if more open than some others.

Nemui Byakko

This is very good and profound article. Trust is a highly underrated factor that determines the development of any society.


Very interesting read for game designers. Half way through I was thinking “is this going somewhere ?” but the last part was extremely informative.
The part that puts high risk with close friends I am skeptical about. Because that doesn’t seem to explain a high risk game like Everquest. Possibly there are other factors in play there or maybe I just missed something in the article.
But anyways, I feel I learned something.

Sana Tan

Thank you this is very interesting!

Kickstarter Donor
Patreon Donor
Loyal Patron
Ashfyn Ninegold

It’s rather sad to read all those trust words and how one builds trusts in games, only to realize that there are a number of players who take advantage of trust building with every intention of screwing the other-trusting player. Usually, the first thing any new MMOer learns is not to trust other players, who will give them wrong information, overcharge them for mats, steal from them in trades, take their time to accomplish selfish goals, and manipulate them in guilds.

As in most other things in life, people have to want to be trusted or trust in order for them to go along with the designers attempts to create trust in a game. Since many gamers considers MMOs competitive on every level, from AH price gouging to kill-stealing to loot rolling, I can’t see this going far.


Yeah, and nowadays when PvP survival sandboxes and the like have become so prominent, trust is even harder to form between folks, and in general it feels like the social aspects of gaming has dwindled quite a bit in favor of quick, easy, and little/no communication needed.


I dunno, I like PUBG, trust noone, kill on sight everyone lol…


The full article is a quite interesting, if overly long, read. It kinda aligns with my experience; as a player I never, ever, trust any other player unless it’s a real life friend, and as a consequence I tend to dislike any activity that falls in the “medium trust” column or above unless I’m playing couch co-op with friends.

Funny thing, if you apply the article’s findings to raiding, you reach the conclusion that raiding is never going to reach mainstream appeal; some of the common features of raiding are even listed as the mistakes made by the research team when building game prototypes to identify which features work well together.


I have some respect for people who create useless graphics to explain away the made-up either confirming or creating speculation on non-issues. Respect in that they have a good hustle in which they make money for doing next to nothing. Kudos.