Riot Games, the company behind League of Legends, takes their community super cereals. I wrote out, “super cereals” because I’m sure it’ll delight whoever needs to edit this piece, but according to Riot’s latest research, there’s a 25% chance it doesn’t mark me as a hilarious wit but rather a unbearable git. From paying employees to resign to tribunals for offending players, Riot have made an art out of tackling toxicity and their latest study has found that in-game trolling can translate to bad workplace attitudes and behaviour.

Thinking there’d be a link between Rioters’ toxic in-game interactions and workplace attitudes, Riot reviewed the people they’d let go over the past year and compared it to their in-game activity. They found that a quarter of those had corresponding bad behaviour in-game, with the most common traits being passive-aggressiveness and using perceived authority (as Rioters) to belittle or bully others. 25% isn’t negligible but I certainly wouldn’t call it the reliable link they were hypothesising would exist. Given that there’s only a 30% chance of a new player staying on if they encounter toxicity in their first game, however, 25% is 100% too many percents for Riot.

More interesting is what else the study reveals about Riot’s corporate culture and hiring process. If you work at Riot, you need to play League of Legends, and Riot has on record your personal in-game handle to track behaviour and chat logs. Having approached 30 junior employees about their poor behaviour in-game as a result of the study (apparently with very positive results), they’re now experimenting with a system that flags potential toxic players with a mark and sample chat logs for recruiters to peruse before they’re hired.

Not everyone seems thrilled with their approach. Renaud Charpentier, who designed and produced Total War Battles: Shogun at Creative Assembly, commented on the story at, as seen below.


A tad dramatic, but what do you think? Is Riot right to monitor this info about their employees? Do you think 25% is a good enough indicator to institute change? Let us know in the comments.

Source: Re:Works,

More stuff like this: