That's the mission statement behind Riot's work on player behaviour. It was also the focus of a talk at the recent Games Developers Conference by Jeffrey Lin – Riot's lead designer of social systems. Teamwork and good behaviour can make or break online gaming experiences so we decided to interview Lin and find out how the player behaviour unit works and what else Riot has been working on.
Lin's background isn't in multiplayer gaming, surprisingly – it's in cognitive neuroscience. While working on a PhD (dealing with visual attention if you were curious) at the University of Washington, friends of Lin introduced him to Marc Merrill and Brandon Beck, the president and CEO of Riot respectively. "They had been playing games for 20 years – way back to the days when they were playing text-based MUDs or Everquest – and they all knew, they had this gut feeling that players want to have a sportsman-like time. They want to be playing with other players that are friendly and positive," he tells Red Bull.
With that in mind Riot was looking for a hardcore gamer with a background in social or cognitive psychology who could work on features to improve team player behaviour in League of Legends. That's how Lin ended up at the company and it's what led to the creation of a whole team dedicated to exploring how people play together.
"When I first joined we grabbed a bunch of people all around Riot and did some research," says Lin. "We really wanted to figure out why were players being frustrated in games. What were the common barriers to having a positive match in online games? What were the pain points players were having? One of the key insights we found early on was people or players weren't inherently bad. What's happening is every single player has their bad days."
That's a paradigm shift: most people don't play unsportingly all the time, they only do sometimes because they're in a bad mood every now and again. Players bring frustrations and tensions from the outside world into the game and even though it might not be common behaviour in a particular player, because it happens to most people once in a while you'll likely run up against examples of it pretty often in a game as popular as League of Legends.
Once you acknowledge that, Lin recognised, you have to change the way you mete out sentences. This realisation is why Lin believes punishing players is often not effective. "If you look at the community at large and split them into demographics based on their behaviour there is a small demographic, say one percent of players, who are persistently negative and you'll need systems in place to help those players reform or improve their behaviours, but the majority of negative experiences come from neutral and positive players having a bad day so we need systems like honour and behaviour alerts to shape their behaviour."
In other words, you need a carrot and not a stick. Where a punishment would come across as harsh and out of context, pointing out to players that they're letting their usually high standards of conduct slide usually results in a change of attitude. Incentivising the good behaviour with an Honour stat which could be affected by conduct in any match also serves to reinforce that good behaviour.
The interesting thing about incentivisation is that it doesn't help to give players character skins or other types of reward with extrinsic value. It messes with the motivation behind the good behaviour and you end up doing things for the reward rather than for the intrinsic value of being a good player or team-mate. An Honour stat is a way of reflecting good conduct while trying not to move motivation away from being a consistently good team player above all else.
Motivations behind behaviour in online games can be hard to unpick so Lin and his team use a wide variety of data and research methods when tackling questions. They change depending on what's being studied and can involve massive piles of data from games taking place all over the world or smaller, more detailed studies involving players working with the team in person.
"If we're looking at frustration we could look at stuff like eye-tracking," Lin gives as an example. "If you look at a player's eye movements, their eye-tracking and their pupil dilation you can use that as a correlate of how frustrated or how stressed out they are."
In terms of less up-close and personal data, in-game chat is the go-to option at the moment. "What we're showing is through a lot of these experiments we can help players improve the language they use and we see dramatic effects in the kind of vocabulary they use and their phrasing."
You might remember Riot's Teamwork OP video from a little while back which gives the statistics on how better teamwork has a positive effect on win rate. "If you look at chat in games in League of Legends at the end of a game you'll sometimes see the term GG which means good game or great game. Over the course of the Teamwork OP video and restricted chat mode we saw about a 2.5 to three percent increase in that phrase being used at the end of matches."
Lin equates the use of GG with a handshake at the end of a match in traditional sports like football or basketball. We ask whether there's a similar relationship with the term GLHF (good luck, have fun) which sometimes appears at the beginning of a match and can set a friendly tone.
"We haven't done the beginning of a match research yet but that's really interesting to note. I imagine the positive players in the community use that phrase a lot more and it probably has a more positive effect going into the game as well."
But when you've got a game like League of Legends where killing enemies and aggressively expanding your reach into their territory is part and parcel of the experience how do you maintain 'good' aggression while diminishing 'bad'?
Lin's answer is to find ways to vary matches and experiences. "In a competitive PvP game one of the things you could do through game design is introduce other ways of mastery or progression. Don't make winning the number one and only goal for players. If you do that you reduce the overbearing weight of winning but also keep it a competitive and still fun game."
Team Builder is one of the ways Lin is hoping to achieve this and so far he says that it's helping players explore different team compositions and break with the current meta. "We're seeing players try things like, let's send two players into the jungle and have them be super aggressive, constantly attacking the other teams territory. That's a super popular strategy in Team Builder."
Thinking about ways to seed positive conduct and set examples of good behaviour, we come to the relationship between professional players and the wider community.
"Pro players undoubtedly have a lot of social influence. If we look at a pro player and they're a very positive, super sportsmanlike player, that has a great weight in the perception of their local communities. They have a lot of influence - but how do we magnify and enhance that influence? We haven't started digging into that space yet."
In terms of exemplary behaviour, Lin calls out Carlos 'Ocelote' Rodriguez. "He was famously known in the past for using really negative language on players - probably one of the worst of any pro players. But he had a couple of turning points where he dramatically changed his behaviour one-eighty and probably now is one of the best examples of being a positive player on his team."
"He talks about one moment where he was at a professional venue in Europe and a player mentioned to him that he flew all the way over there just to watch him play that game. He talks about how that one moment made him realise 'Hey, why am I being a jerk because it's telling everyone else that being a jerk is okay.'"
How players behave online and the motivations behind that behaviour are growing fields of research, so can Riot see a future where different studios and research labs team up to form a broader team game research hub?
"If you look at Riot and other game studios they're all starting to have a team focused on player behaviour research," says Lin. "It's not about the positive or negative side, it's about researching what it means to play with other people online.
"It's a little bit too early to tell what's going to happen five or ten years down the road but I'm willing to bet almost every studio is going to have a team dedicated to that space and at least researching how to design games that are more fun from a social aspect and not just a gameplay aspect."
In the meantime, don't be too surprised if the Xbox Live team adopt a similar tactic for dealing with wayward Call of Duty players. Here's hoping.