Lyte talks Tribunal, Chat Restriction, "Call vs Pick", and more.

Posted on at 3:34 AM by Moobeat
Lyte, Riot's resident Lead Designer of Social Systems, has taken to the forums to discuss and update on a few of the player behavior systems.

Continue reading to check out what Lyte has to say about the Tribunal, chat restriction, "call vs pick order", and other player behavior related issues.

Lyte started out his barrage of player behavior posts by responding to a thread asking him what he is up to and what he has planned.
"1. How is the Tribunal doing, and what's in mind for the future? 
a. We're at a point in the Tribunal where we're consistently identifying then reforming or removing the most toxic players in the game. It has a high accuracy, and an extremely low false punish rate.

Regarding Groupthink, I've talked about this issue here: 
However, the Tribunal is not the complete player behavior solution. It's just one piece of the toolkit. In the past year, we've been conducting a lot of research on the player behavior space. What are the biggest problems? What systems could we add to the community player behavior toolkit to help enact change in this online culture? We know that speed and clarity play critical roles in shaping behavior, and speed is a current weakness of the Tribunal. We're exploring whether we can do anything to improve the speed of the Tribunal, or improve the speed of feedback for negative behavior in general. 
In the future, we've decided to make a pretty big shift and focus on features that focus on improving the experiences for the vast majority of our players--the neutral and positive players in League of Legends.

2. What are your plans for the Honor Initiative? 
a. Player behavior systems are experiments in progress. We've been iterating on the Tribunal and implementing new features and tweaks for over 12 months! The Honor Initiative is the 2nd piece of the community player behavior toolkit, and is still in its early stages. We recently used Honor to help identify players that were positive during the Snowdown Showdown and awarded them a surprise Santa Baron icon. We're going to take the lessons we learned from this, and potentially plan more events around our positive players in the future; for example, a fun event might be to give players a goodie for being awesome at identifying and honoring positive players. 
Similarly to the first question, we're also exploring ways to provide faster feedback for positive behavior in general.

3. How are you? 
a. Bet you didn't expect me to answer your post! I'm doing well.

4. Anything you would like to announce? 
a. We've been experimenting a lot with account restrictions on the PBE. We agree with players that bans are not the complete solution--some players are deterred by bans, but some also create new accounts and just shift their toxicity onto a smurf account. The first experiment we tried was a 'restricted chat mode' []. We're intrigued by the potential it showed on the PBE, and are currently figuring out how to test a large-scale version of this on live servers.

Secondly, I've been seeing a bunch of threads about pick order versus call order in Champion Select. We have been running a lot of research on Champion Select and agree, Champion Select is currently not the ideal environment for teamwork. Sometime tomorrow I hope to grab statuskwoh and davin and see if we can have a discussion with players about Champion Select. I'd like to talk about some of the current problems with Champion Select and why the problems are difficult to solve; however, this is a problem I hope the player behavior team can take on in the future."

He then responded to someone questioning the effectiveness of the chat restrictions currently on the PBE, saying:
"We're curious about many of these questions too. Our goal is to not restrict or inhibit players from positive communication; however, how many truly toxic players use chat in a positive way? Will the new ping system be enough for these players to communicate their intentions, even without chat at all? 
Experiments on the player behavior team are just that--experiments. Human behavior is complex, and it's very difficult to predict every outcome. Sometimes we just have to try, iterate, and try again."
When asked why Riot doesn't promote the mute future as a way to deal with toxic players, he responds:

"The mute feature is an interesting discussion point. I consider this to be a 'reactive' feature, meaning that you can only use it in response to a human behavior. If someone verbally abuses you or uses a racial slur, you can simply mute the player; however, damage has already been done. 
Some players can criticize the playerbase for not being hardcore, but what does being hardcore have to do with anything? I've never heard or seen a correlation that being a hardcore, competitive player means being a jerkwad. I've been a hardcore gamer all my life and have played at an extremely high level (and currently entered Platinum in League!) and I still don't feel like spewing f*ggot or c*nt adds any value to competitive play or teamplay.

One philosophy on the player behavior team is to create preventative features--features that can prevent toxic behaviors. At the end of the day, we can promote Mute and make it a better feature; however, we want to do better for players."
Lyte also responded to someone who suggest a "toxic meter" that let players know just how bad they are being and how close to being punished they are, sharing an insight on why players might be toxic or "rage":
"A toxic meter in the profile somewhere is an intriguing idea, but one with some pretty large concerns. One initial data point that players might not know is that the vast majority of players in League are neutral or positive. if the vast majority (and I'm talking 95%+) of players are not innately toxic, what's the point of having a meter that only moves for a very small % of players? We could change the resolution of the meter and apply some fancy math to make a meter still work for the entire playerbase, but our latest research points us in a different direction. 
Some of you are probably wondering, "Lyte, if the vast majority of players aren't toxic, why do my games seem so bad?" Unfortunately, the problem here is context. All players have bad days. A player could have a bad day at work and get raged at by his boss for no reason. They go home and play some League and play a terrible game (because they are distracted by what happened at work!) and his teammates rage at him all game. He snaps. He rages back. But importantly, he probably doesn't rage in the next 50, or even 100 games. 
This really shows you the power of context, and how context outside the game can affect behaviors in-game. With millions of gamers with unique lives and situations... a lot of gamers are having their 'bad day' in any given game. We've found that this is the source of our problems.

The Tribunal is great at identifying the systematically toxic players--the players who are just toxic in a huge chunk of their games. It's not great at solving the above problem. This is the problem I've been working with my team on."
As a follow up to this, he also said:
"The majority of this playerbase is neutral or positive. We need to find ways to highlight this behavior. Spotlight it. Talk about it. We need players to look around and realize that most of the players around them are awesome people."

No comments

Post a Comment