How statistical misuse is hurting League of Legends’ ranked mode

Dynamic queue was meant to be the future of League of Legends.

Its inception was innocent enough: Riot Games saw a negative trend in its playerbase and determined adjustments were necessary. The mode’s launch, however, was bogged down by a handful of flaws Riot has failed to correct in the five months since launch.

Riot’s attempts to discuss the game’s controversial new queue drew constant flack from the community. It’s understandable that a seven year old ranked system needed an overhaul. Riot committed to the growth of dynamic queue and accepted any change would be met with angst from the community. Riot ignored early concerns with the reassurance the new system would prove superior.

In a Riot Games roundtable that attempted to clarify the goals and struggles associated with dynamic queue, William “Scarra” Li was asked of his experience with the controversial system.

“It’s awful,” Scarra said. He then described an ideal system that included fair matchmaking, reasonable queue times and ideal role enforcement. “On all three levels, the matchmaking system has failed.”

Why is Riot so adamant about a system some of the game’s top players think is fundamentally broken? It’s all down to the numbers.

Scarra captured the competitive community’s concerns, yet the panel of Rioters — which included Chris “Lobster” Cobb and Chris “Pwyff” Tom and  Eric “Socrates” Kenna — failed to provide a meaningful solution. Instead, they assured everyone dynamic queue is better than solo queue. They pointed to a dip in harassment and a rise in ranked play as justification. Those numbers seem to make a strong case for dynamic queue, yet there are too many lurking variables Riot cannot ignore. Of course a system that lets friends play together will decrease harassment.

If Riot believes harassment numbers make a strong case for dynamic queue, it should release how often players report their queue partners. If players are encouraged to play with their friends, then that would explain a false dip in harassment. If harassment is significantly lower in games with no premades, then at least Riot’s point is valid. If Riot is content reducing interactions between random players, then we must ask: Is that the most important goal of a matchmaking system? Should players be content with Riot when the company is convinced longer queue times, less meaningful ranks and weaker matchmaking is a fair tradeoff for an unsubstantiated drop in harassment? There is no perfect answer to those questions, yet Riot’s refusal to budge on the issue demonstrates that it is focused more on twisting the numbers than receiving feedback from the community.

British Prime Minister Benjamin Disraeli is often attributed with saying, “There are three kinds of lies: lies, damned lies, and statistics.” Riot’s cases for dynamic queue lives up to that notion — a weak argument bolstered by misleading statistics.

If there is any question where the community stands on the issue, simply spend a few minutes on the League of Legends subreddit. Multiple posts on the front page mock Riot’s response, and many users went as far to downvote any Riot response into the ground. League players felt betrayed by Riot’s decision to can the return of solo queue, and this is their platform to lash out.

The negative outbursts have done nothing to curb Riot’s campaign for dynamic queue. They trust their numbers more than feedback, which is a dangerous platform for a company to take. “We understand your concerns, but our numbers say this is better,” is the kind of arrogance that forces players to seriously reconsider their gaming allegiance. The positive buzz about Overwatch only amplifies the risk. If Riot can’t find a way to make ranked play meaningful again, then the issues in high MMR will trickle down to all facets of the game.

Although Riot flirted with a limit on high MMR dynamic queue, perhaps it could cap dynamic queue at Diamond. It would force Master and Challenger players to play a separate solo/duo queue to climb the rankings. That would solve the high-elo issues with queue times and premades, but it would not address the lack of a Ranked 5s game mode. There’s no easy answer, but ignoring community feedback — both from professionals and regular players — with a “holier than thou” perspective could stunt the game’s growth.

Riot’s confident in its numbers. If the developer is correct, the outcry comes from a vocal minority that will fall in line soon enough. It’s up to the players to vote with their wallets and prove to Riot that their feedback is truly worthwhile.

Cover image courtesy of  Chris Yunker/flickr.