2014 has come and gone, taking with it the creation of seven new featured game modes for League of Legends! Looking back, we had fun making each and every mode, tackling the unique challenges each one presented, and hope you guys had just as much fun playing them. This feels like a good time to look back at 2014’s Featured Game Modes, what we learned from some of them, and how we arrived at a variety of design decisions along the way. But first...
This is something we're continuously defining for ourselves as Featured Gameplay Modes evolve. We want to learn from previous modes to help us make even cooler new things, but it's not as simple as just which mode was played the most. What does “the most” even mean? Is it the highest single spike of concurrent players? The game you played for the longest? Modes also need to be unique in some way and not just replace or overwrite themselves and other permanent modes. As we’ve mentioned before, we have a few design pillars that help guide the creation of each mode, but we’d like you to discover new or interesting ways to engage with your favourite champions in League of Legends. We undoubtedly also want you to have FUN! :D
Then there are modes which produce amazingly cool stuff from you guys in the community that we can't even measure with numbers. How do you measure fun? Or cool streams, videos, comics, fanart (the list goes on)? Sometimes it’s even nice for the mode to help you train in skills (team fights, skill shots) that are transferrable over to SR, but should that be a core pillar? We’ve said before how we like that each Featured Gameplay Mode feels like a wild west of meta for their duration. Anything goes initially and you guys theorycraft cool ideas about the most OP comps and champions for each mode, and surprise us every time. There's always a few sleeper champions that appear who we could NEVER forecast. So healthy champion pick diversity is also good, but is it something we should weigh the success of a mode on? These are all just some of the things we consider when both designing and then weighing an individual mode’s success.
Featured Game Modes are designed from the ground up to be short-term engagement experiences. This transience is what gives us the creative space try new things and see what works, listen to your feedback, then improve a mode before re-releasing it. Trying to build a long-term sustainable game mode would actually constrain us from doing things like Doom Bots, URF, Legend of the Poro King, Ascension, etc.
The main design concern (which is personally the most important as it = less fun) is that Featured Gameplay Modes are not designed to be long term engagement experiences. They’re designed to make a big splash, with the knowledge they will be removed shortly thereafter. Adding permanency would constrain the design wiggle room we have to make each mode as unique as possible. As we’ve since learned, it would also add additional balance and maintenance costs that we’ll share some examples of later in this retro. For now though, let’s get into the modes!
2014 started with a Featured Game Mode meta changeup. With the original Hexakill, we wanted to try exploring a new space that modes hadn’t really touched yet at that point. The end result was the closest we’d ever brushed up against the ‘regular Summoner’s Rift meta’, but with the twist of an extra player. Our goal was to still preserve the feel of Summoner’s Rift, but change up how you interacted with your champion and played within the map’s meta.
Originally planned as an April Fool’s Day joke intended only to last one day, we were blown away by the response Ultra Rapid Fire received. Based on the response from players around the world in multiple regions and languages, we extended the length of the mode (twice). Who doesn’t like making more plays?! URF also taught us a fantastic lesson about the novelty of game modes over time and the cost of ongoing maintenance.
Featured Gameplay Modes have been shown to taper off in popularity sharply after a short period of time. People often forget that URF actually broke our golden rule of not touching Featured Gameplay Modes post-launch, but to try and keep it from becoming stale in the face of wildly toxic champion play patterns, we did anyway. And yet, despite our consecutive on the run tweaks to keep URF treading water balance-wise, we still saw the same declining engagement and burnout that we see with other modes. That doesn’t mean players don’t have a blast with each mode like this before they fade away. However, being designed as a can of whoopass in the first place ultimately means that the flame burns twice as bright and half as long. And that’s okay!
The original One For All was one of more heavily played modes we made in 2013. Before it was even released though, some enterprising players hacked their clients and were hosting ad-hoc games of One For All on Howling Abyss. After the mode’s first outing on Summoner’s Rift, many players heavily requested that we bring the mode back on Howling Abyss to replicate their ad-hoc games, so we got to work on the resurrection. In the end, One For All: Mirror Mode may have lost a lot of the strategic choice of Summoner’s Rift (naturally absent on Howling Abyss where your only option is to “push”). By homogenising both teams into the same champion, we also lost the nuance in gameplay between two champion’s kits. Engagement with the mode proved out that this maybe wasn’t the best experiment, but that’s what learning is for!
We also learned a lesson about the cost of resurrecting a mode. Each time we bring a mode back, along with any improvements or additions we might want to make, time has to be spent QA'ing it against the latest patch release (one of the costs of having a game that evolves constantly). For example: The original One For All required about 80 champions to be hand edited to work with the mode's mechanics. When we wanted to bring it back as One For All: Mirror Mode, about 35 champions had changed since the last time (yay champ updates and improvements!) which required another whole round of editing, plus entirely new champions had been released. If we ever made a mode permanent, we’d need to weigh the impact this would have on the ability to create new modes VS maintaining existing modes.
The map change isn’t always necessary though. Sometimes smoothing over the experience and some bug fixes could generate the original experience we were looking for. URF would be nice simply to not have 9 champions disabled. It always depends on the mode.
The idea of building some ‘absurdly challenging’ AI bots had come up before, so when the AI team was updating what are now the intermediate bots, we saw a chance to do a cool collaboration with them. The Doom Bots of Doom became the first PvE mode we’d ever done, and it they were pretty fun to make. Unlike other modes though, the Doom Bots were extremely content heavy, as we hand-crafted 15 alternate kits (not mentioning the ones that got thrown out). It ended up being super rewarding though, watching player’s reactions the first time they see Doom Lux ult or try and 1v1 a Doom Galio or Doom Malzahar. As a bit of trivia, the Doom Bot with the highest winrate on 5 bombs actually turned out to be Doom Amumu. Too many tears.
The goal of Ascension was to highlight the Shurima event and bring gameplay to some of the reigning lore themes like “power corrupts”. We wanted a scrappy, team deathmatch kind of feel right from the start, and it took some crazy prototypes before we landed on the right mix. In response to players asking for more ways to express their skill level visually, we also experimented with the Perfect Ascension icon; a true challenge mode icon rewarding exceptional play. For those of you who did earn it, it’s one of the rarest summoner icon’s in all of league.
The making of Ascension has been discussed before in quite a bit of detail with a dev blog, which can be found here.
After the first outing of Hexakill and the warm reception it received, we felt like it was a good candidate to bring back. As with One For All: Mirror Mode though, we don’t want to just resurrect the exact same mode from last time. We wanted to turn up the volume on the design goals of the original Hexakill (being to shake up how you think about and execute on established meta on a given map). That led us to port the mode over to the Twisted Treeline. This is a good example of how we want to improve modes each time before we re-release them.
It depends on the mode, but it’s always heavily informed by player feedback. In the case of Hexakill, we wanted that exaggeration of the original design intention. Hexakill on SR just wasn’t different enough; going from 5 to only 6 players didn’t really make you change the way you played that much. Twisted Treeline felt much higher impact because we were essentially doubling the player count on the map.
Don’t worry, he lives on... inside of YOU! *little tear* Legend of the Poro King was seeking to change the way players thought about combat on the Howling Abyss. That we were able to introduce everyone to the grandest most splendiferous King of all the Poro’s himself was just an added bonus.
I want to call out the Poro Toss summoner spell here as well, which gave an across-the-board bump in viability to some historically unpopular picks. Even with picking your champs, engage from the Poro Toss summoner spell was able to wing clip the power of some notoriously strong Howling Abyss champs. It was an overall healthier experience, seeing good champion diversity and lots of big plays.