The 2016 ranked system and the death of ranked 5s

Understanding the shift in the ranked system in the context of the rise, fall, fall, and fall of the ranked 5s system and competitive play.

Riot yesterday announced a large number of changes to the very infrastructure of the league system for season 6. There were some major surprises including the announcement for the alpha of a new client and an official chat app, and some lesser surprises (ranked team builder in general), but one point stands out as no-one having seen coming – a complete restructuring of the ranked queuing system.

The full details are available in the Riot post. In short, players will be able to queue up with any amount of friends for the traditionally ‘solo’ ranked queue. Much has been made of the impact this will have on solo queue, particularly with the existence of 4-stacks. The author has his own opinions on that, but this article isn’t about the effect on solo queue. Rather, this article is about a quite different trend – the death of ranked 5s.

Ranked 5s before teams

Ranked 5s in its current form came about with the creation of the leagues system at the start of season 2 – prior to that, while there existed an arranged 5×5 queue, the teams system didn’t exist and Elo was tracked individually.

During season 1, ranked 5s were extremely popular among EU teams and players, with numerous players (and, by extension, teams like Fnatic, SK, ALTERNATE, aAa, exGameburg Team (future MYM), etc.) on EU registering upwards of 300 games in said queue. ForellenLord, the iconic high Elo player of the era, registered 852 games on solo queue and 860 on ranked 5s in season 1; yellowpete, who finished #1 on both queues, played 432 and 351.

This wasn’t quite replicated in NA to the same degree – TSM and CLG both played around 150 games against quantities pushing into the four digits in solo queue, and the number of players with proportional ranked 5s games is tiny (three fun historical notes: ManyReason, then of Team OP and later of the Xenics org in OGN, was #1 in ranked 5s without being in the top 225 in solo queue; teammates a Lilac and Cornsalad were second and third; #4 was mancloud, who played 683 ranked 5s games) – but it’s fair to say that the queue was still well-used by top-level teams.

Season 1, however, was where ranked 5s hit its peak. Season 2, in theory, should have meant renewed strength for the ladder – after all, it provided a very public way to make your name as a team, and gave as fair and open a model as possible for Riot and other tournaments to use in assessing what teams to invite to their own online and offline events. Yet, in the event, participation declined massively.

Season 2 and gamesmanship

While there were the complicating factors of the pro scene and online tournaments reaching peak prevalency in that season, as well as the pro scene itself becoming sufficiently developed as to allow for things like the regular scheduling of scrimmages between top teams, the new system quickly proved itself to have a fundamental flaw compared to solo queue: the point of having too much to lose. Even in solo queue, Season 2 was the year of the smurf, as people realised that a) their solo queue rank was important to their perception in both the pro scene and the larger community, and b) the Elo system allowed them to gain rank quickly and then maintain it indefinitely. A majority of top players on the NA ladder had played over a thousand games in season 1; the highest in season 2 was Burbs with 630.

In ranked 5s, teams quickly realised that the Elo system meant that they could get to the top of the ladder extremely quickly on a good winning streak, and that they would then lose so much more in terms of said rank from a loss than a win should they be matched up with a low-level team that why even bother? The top 5 teams on the NA ladder (CLG, Curse, Dignitas, and two others) played an average of 17.6 games between them, with a combined record of 84-4. Teams were somewhat more active – and more prone to losses – on EU, but even then, most had between 15 and 40 games played.

This was not the situation that Riot had wanted. While the implementation of teams itself – being as obvious a step as it is – didn’t necessarily say anything about their intentions from the ladder, it’s hard to think that there wasn’t a hope that it would strengthen the ranked 5s system and encourage teams to take it seriously. Consider Riot’s decision to change the Elo system to leagues in season 3. That decision can be seen as a clear incentive in itself against the ‘elo-parking’ prominent in both solo queue and ranked 5s, but there’s an interesting tidbit to be found in Riot Yegg’s explanation of the changes back in September 2012:

Will I have the opportunity to go pro if I reach the top of the ranked 3v3 or solo/duo Challenger tier?
Although there are no League of Legends Championship Series events for 3v3 or solo/duo competitors, reaching the Challenger tier in these rankings will help you find other top notch summoners to play with. If you’re interested in making a run at the Championship Series, try messaging some fellow Challenger tier competitors who aren’t attached to a current 5v5 ranked team. You’ll probably find you have plenty of potential comrades to help you pursue your dream.

It was accompanied by this image, explaining the qualification process for the upcoming League of Legends Championship Serieses in Europe and North America:

The system and the philosophy hinted at by Riot have largely remained the same since the season 3 changes. Riot has tried repeatedly to protect and promote ranked 5s as a central, near-exclusive hub of amateur play – it should in theory provide the best arbiter of competition if everyone’s on it, it provides a clear integrated infrastructure from the lowest levels through to the LCS, it’s integrated in-client and without third parties and thus provides the best possible fan access, and so on.

If one wants evidence of this, they need look no further than the LCS promotion system itself. The four qualifying members from partner events mentioned in the graphic lasted for the spring 2013 qualifiers and summer 2013 promotions. For spring 2014, there were Riot-hosted qualifiers throughout the summer drawing from an online qualifier between the top ranked 5s teams – the system was hence already taken in-house. For summer 2014, qualification took place through the participants in the newly-formed Challenger Series – another organised online tournament of ranked 5s teams with offline ‘finals’. The same rough format was followed for the Expansion Tournament in 2015, and for the 2015 Challenger Series and Promotion Tournaments. Riot has wasted very little time in building up a theoretically robust strcuture that all ultimately comes from ranked 5s play, to the exclusion of any other competition.

Giving up the dream and scandal in 2015

There was a small hint that Riot would be looking to go in a different direction from 2016 onwards. A new format for Challenger Series qualification was hinted at a while back, and we’ve known for the past month that the ESL Meisterschaft (the top-level LoL competition in Germany) would provide its winner with a qualifier spot this season – the first time an external tournament has done so since early 2013.

So, why now? In short, 2015 has been the year to fully blow open every possible defect with the use of the ranked 5s ladder as representative of pro play; if season 4 marked a watershed in ‘solving’ the game in general, season 5 has been the watershed in ‘solving’ how to game the system. While the Expansion Tournament ladder rush largely passed without serious issue (bar the usual complaints about the inferiority of EUNE, and the inauspicious shock collapse of PLATINNRUSH from #1 seed to out in the dying hours of play), the 2015 Spring Challenger Series qualifiers brought a bit attention on account of Origen.

Origen picked up considerable attention as they expectedly stomped their way up the ladder on the way to Challenger Series qualifiers. However, it didn’t escape notice that – in spite of being fairly inarguably the best active team on the ladder at the time – they only finished 9th, taking the #6 seed. While part of this was due to decay being entirely disabled in all ranked ladders for preseason (and thus removing ANY competitive incentive for top teams to play – most teams on EUW were essentially totally inactive prior to qualifiers), it was noted at the time that teams near the top of the ladder were, according to (reportedly) Origen management, dodging games with them.

While in the event they were able to put sufficient distance between themselves and the rest of those who hadn’t in practice already secured their spots in the previous ladder rush, it did bring back the realisation that there might be something intrinsically wrong with how the ranked 5s systems worked. This would be brought up, in far more high-profile fashion, in the summer, with another ‘super-team’ climb – this time under the stewardship of Wickd.

Like xPeke and Origen, Wickd and ‘team brun sovs’ (now under the Denial eSports banner) were an attempt by a veteran player to put together an all-star team and make it into Challenger Series through the ranked 5s ladder. However, while xPeke’s team had plenty of time and operated when there was no real ‘ladder rush’ for qualifiers (since most spots were determined already), Wickd started playing with his team mere days before the ladder lock, with competition already raging at the top of the ladder.

That, combined with Wickd’s willingness to be far more vocal than Origen or their management about ladder issues, meant that this time the community noticed. Two Reddit threads (1, 2) on the issue got over 2,000 upvotes, and Wickd and others spoke openly on a number of problems surrounding the ladderlock including teams dodging, teams smurfing and queue-sniping to ensure miserly gains or significant losses for rivals, and even drophacking of games (although it’s important to add that no accusation was made of the involved parties themselves engaging as such). What had briefly looked like a great week for ranked 5s visibility (some teams were streaming their games, while spectator streams of high-elo ranked 5s games were also modestly successful) became a massive embarrassment for all involved.

Wickd’s team did eventually make it into qualifiers and then the CS, but the episode did a lot to undermine confidence. To add insult to injury, the NA ladder race – which had looked relatively clean – was itself brought into some disrepute when Escalate Legion Gaming, the #7 team on the NA ladder, were disqualified for using a ringer via account sharing in their own ladder rush.

The ladders on both NA and EUW currently stand as a particular embarrassment. Most of the teams from the ladder rush in May have entirely decayed out – part of that is due to the characteristic roster swapping of Challenger, but there are multiple active teams that have simply not even bothered to climb or to maintain their position in spite of how little effort that would take (one of the few prominent Challenger teams to do so in NA, CLG Black, has reached rank #12 in 33 games). The point where in NA, 400 LP would be good for #4 – by way of comparison, said theoretical team would have been #18 a year ago. EUW has the eye-sore of a single team – Enigma eSports – occupying three of the top five spots – something that Wickd weighed in on shortly before the changes were announced:

This isn’t all that there is to Riot’s decision, of course. Ranked 5s has other lower-level problems with progression and matchmaking, and rolling the ability to play 5-mans into the main queue makes those problems go away – yes, it also creates new problems, but at least those problems are in one place. But the relation between competitive is important to consider, and it goes a long way to explaining why Riot are happy to see the mode dwindle away.

It will be interesting to see in the coming months what this will mean for Riot’s relationship with third-party tournaments, and in general what it means for the sub-Challenger Series scene that – in EU in particular – has gotten weaker and weaker over the past year or so with the end of the likes of the Black Monster Cup and dwindling participation in many ESL events.