When we originally sat down to build Hextech we played with a few loot systems in other games to get an idea of what was out there. We noticed a pattern of hard monetization gates — in a lot of cases, the game would give you a chest, but ultimately you had to pay to open it. As a player this felt pretty crappy, so early on we knew we wanted players to be able to earn Hextech content without having to pay.

On the flipside, giving away millions of skins each month can quickly limit our ability to keep investing in better servers, balancing the game, and developing other features that help the game for all players (regardless of spend). From that we came to key fragments as a way to balance giving out free content at a pace that made sense for all players and felt rewarding whether they played a little, or a lot.

Now that the system has been in players’ hands for the better part of a year, one issue we’ve seen is that no matter how many fragments and chests players get, they say “I’m not getting enough.” Some of that can probably be solved with a clearer feedback system, but some of it might be unavoidable. Who doesn’t want more free stuff?

The basic rules for key fragment drops are:

1. Key frags are only dropped from wins.
2. Key fragment drop rates work off of a per player cyclical reset timer, similar to chests. Higher drop rates are slightly front-loaded and gradually slow based on the total number of key fragments you’ve received in a cycle until the reset.
3. It’s possible to earn 4 keys (12 fragments) in a 4 week period. This matches the number of chests players can get.
4. There is a bad-luck protection rule that puts a maximum number of wins between key frag drops.
5. Players get a small drop chance bonus if they were in a premade.

While this ruleset is functional, there is definitely room for improvement. We are looking at some possibilities for where we can take this system next, but aren’t ready to share all the details yet.

As always we appreciate all the feedback you’ve given around this and hope you’ve all gotten a ton of great loot over the last year.

Getting advice from players who are extremely good at playing League of Legends is really valuable when it comes to working on the game. We’ve got a number of Masters, Challenger or ex-Pro players at Riot who are constantly involved in the balance and testing process as a result. However, there are a lot of other skills that are also really important in addition to being good at LoL (in many cases more important), so our game designers are spread across a range of different divisions.

We’ll take a look at a team-by-team breakdown below for the teams that work on balance-impacting content. Before that, though, it’d be good to do a quick runthrough of some of the skills we look for. This isn’t a completely comprehensive list (I’m sure I’m forgetting a bunch of things), but hopefully captures the core concepts:

Design craft:

Analytical ability and problem-solving

• Being able to assess a situation or element, break down what’s created it, propose different approaches that could be taken and their likely effects, etc.

Creativity and thematic ability

• Understanding what themes resonate with players and fit the LoL IP. Finding new ways to put a spin on broad archetypes. Recognizing the right matches for thematics and particular gameplay mechanics that are intuitive rather than forced

Useability/accessibility and feel

• Making player interactions with the game (using an ability, navigating the shop, natural

Communication and Leadership:

Vision building

• Ability to get others bought on and excited about an approach

Mentoring

• Helping others grow, learn new skills

Written and verbal communication

• Concise, able to generate excitement, clear, right method for the circumstances and so on
• Ability to communicate well with other disciplines (this one is particularly important; e.g. forging strong relationships between art and game design).

Management

• Managing others (performance, etc.) professionally

Other stuff:

Technical skills

• Scripting abilities, statistical analysis, writing code, etc.

Organizational skills and reliability

• Being able to juggle multiple projects, deliver work on time, solve resourcing conflicts, etc.

So, looking at the individual teams then:

Live Gameplay:

• The team responsible for quick changes to the game. Tends to be focused primarily on game health and balance
• Values understanding of the current and past states of the game very highly
• In ranked: Gold through to Diamond, average high Plat/low Diamond

Game Systems:

• Systems focuses primarily on long-term work, working on the foundational rules that LoL runs by. Stuff like map layout, how runes/masteries work, how XP/gold are generated, etc.
• Systems works in more conceptual/abstract spaces on average than the other teams. Design principles and theoretical analysis are especially important here
• Bronze up to Gold

Playtest:

• A team whose primarily responsibility is to test upcoming changes throughout the day and offer feedback and analysis on them
• This is where ability to play at a really high level is particularly important in order to validate in-game how things will play out in really high level play.
• Mid-Diamond upwards, with some former Pro players (LCS). Some players who move in and out of Challenger occasionally

Champion:

• Champion team makes new champions.
• Thematic understanding, feel/useability tend to be particularly important here, along with game health understanding (strengths/weaknesses of kits in particular).
• Pretty even spread from Silver up to Diamond

Champ Up:

• Updates old champions, whether full VGUs, like Warwick recently, or class updates like the assassins
• Similar set of skills to Champion, though with a heavier focus on game health
• Silver up to Diamond, clustering in Plat

Hope that’s been informative and offers some useful context on the range of different skills we look for and how the different teams compare. People who are awesome at everything are certainly ideal, realistically though there’s always some degree of tradeoff of different skills for any given role we’re looking to fill.

We’re still a couple patches away from detailing the specifics of the 10-ban system outside the pro scene. Increasing the number of bans has a different set of goals for an organized team of players versus individual players on a team. For now, here’s a look inside a few of the design thoughts around a 10-ban system for all draft queues. 

Agency – Giving every player a ban should give everyone more control over their game experience. While hard lane counters aren’t too common in League, this means you can grab Tryndamere and not worry about Teemo ruining your day.

Fairness – Your order in champ select is essentially a dice roll. We want to make sure that no matter where you sit in the champ select order, there’s less potential of feeling kinda disadvantaged. For instance, being second pick on blue side is arguably the weakest position in the current system – you don’t get a ban, and you’re fourth pick overall, so you aren’t especially likely to get a power pick or a counter pick.

Time – We’re not aiming directly for a time reduction in champ select, but we’re always looking for new ways to shave seconds off anything from getting into game. Adding four more bans to the system in the same format could stack up to two extra minutes between clicking ready and hitting the loading screen, but that’s not the direction we want to go. We could counteract this by shortening the ban window, but stacking even more time pressure when your proposed ban gets sniped by the enemy team feels bad.

We know changes like this have far-reaching implications so we’re proceeding cautiously – we’ll have more to share in the coming weeks.