Design vs Balance, and their effects on gameplay.

Following the controversy of the 5.4 changes, I feel a general discussion about balance and design is in order. My hope is to broaden perspectives about champion design for the many players who are now over-reacting over the patch 5.4 changes.

Now, let's preface the discussion with a few assumptions:

1. Riot's balance team is competent in what they do. The designers who work on the game are paid professionals. They have game design experience either from working outside the game, or from working on multiple patches of the game. Riot's designers have all prerequisites for doing a good job with balance.

2. The design and balance teams face demands from above that restrict what changes they can and cannot make. Even if a designer is in charge of how a champion is implemented into the game, they might not be fully in control of at what point the changes they want to make go live.

3. The most popular champions should have the most discretion when it comes to implementing changes. Champions are the main selling point of the game. Radical unwarranted changes to popular champions have the drawback of potentially driving players away from the game. Unless those champions ruin the game's experience for the other players, changes should be implemented only with caution.

4. The design of champions supersedes their immediate balance. Changes are made primarily to fit the idea of champions. From a player perspective, it's more important for a champion to feel right than to be absolutely powerful. The design of champions is what sells them, not their absolute power-level at a specific point of time.

5. Absolute balance between all champions is not desirable from a player or spectator point of view. Having too many viable picks is detrimental to both the players and the viewers. If everything is viable, then a team doesn't have to concede any power picks to gain the champions they want the most. The aspect of making the best with limited resources doesn't exist, and the value of having a strategy is diminished.

6. Certain champions are purposefully stronger than others as a means to encourage the community to do work for the design and balance team. This is most apparent with new champion releases. Most new champions are ahead of the curve in terms of power. The novelty of being new attracts many players to buy and try the new champions, and their relative power is what keeps players playing them. The design and balance teams gather data at a much faster rate when a champion is played more. For junglers, this gives the team data about the new jungle as well. While establishing jungle diversity has been a failure so far, it's probable that the design team has enough tools at their disposal to institute higher diversity now thanks to how restrictive the jungle has been.

7. Win rates aren't necessarily a good indicator of champion strength. From a statistical point of view, this seems like a fallacy. However, synergy with other champions, individual skill, and even which side of the map the champion is played on are major factors to take into consideration. Win rates are one tool among many to evaluate how strong a champion is. They aren't the definitive one.

Now for what I actually wanted to talk about:

• What kind of gameplay is desirable for League, and what kind of gameplay is undesirable?
• Which champions should undergo design changes, and which should just undergo numerical changes?
• In what manner should those changes be implemented?

What kind of gameplay is desirable for League, and what kind of gameplay is undesirable?

This is a very controversial question, and one I'd like to hear a multitude of answers for. SephAgro makes some points regarding the changes to Veigar in 5.4 in his thread here:

The point I'd like to raise is this one:

Veigar is absolutely gutted. Specifically, his flashy-play shut down ability; Event Horizon.

I feel this gives insight into the idea that has been driving Riot's balance team for quite some time now. As much of a meme as it has become, the fantasy of what a champion should feel like at the later parts of a game is a driving force behind how champions are balanced. League currently has a multitude of "flashy play-making champions". If those champions lose that ability, they lose much of their identities. Abilities that threaten that potential are undesirable in the eyes of the design team because of that fact. A good example of this is how the support role has changed since around season 2 of the League. In season 2, supports generally didn't get enough gold to purchase more than one big item to play with. Currently, the support role has gone through a drastic overhaul which mostly allows support players to still feel powerful in the later stages of a game. It's an example of a fantasy of what a champion is being fulfilled. Taking that away would impact overall enjoyment of the game for most support players. Thus, we establish what I think Riot's design team thinks is a mark of bad design: An ability is bad if it's more about restricting others than it is about fulfilling the fantasy of its owner. That's an incredibly broad statement. It would mean nearly all CC in League is bad design when taken at face value, because CC inherently restricts characters from moving, using abilities or doing damage with attacks (or even all three). Yet for many champions, being able to use CC fits their identities. CC fits into the fantasy of what these champions are, and have been included in the kits specifically for that reason. The idea is that these champions would pay a tax regarding their other attributes when compared with similar champions who do not have this CC. So what we can establish is this: If CC fits the perceived identity of the champion, it's justifiable to include it in the kit. If a champion has CC as a tool to protect themselves while they use their other abilities to fulfil their fantasy, then perhaps that CC should be replaced by a different ability instead. Let's use Veigar as an abstract example because of the thread I'm referencing. If Veigar cannot have reliable CC because it doesn't fit his fantasy, then Event Horizon should be redesigned into something else entirely. Veigar is a burst caster who lacks mobility, but compensates for it by having incredible single target damage. Perhaps instead of an ability to restrict his opponents, he should have an ability to defend himself, to ensure he can get off at least one spell rotation? If that defensive ability scales with something other than AP, veigar can make the decision to either be a burst caster with some tools to counter assassins, or a more sustained damage mage with survivability. Properly implemented, it could add variety to a character designed for mostly a single role.

Gameplay that reinforces the core idea of what a champion is is desirable. Gameplay that is added because of abstract reasons to make a champion feel balanced is undesirable. The design team is responsible for deciding what abilities make for gameplay that fits the idea of the champion. The balance team has to work in that frame with the numbers of the abilities until they can reach a point where the champion feels balanced. The best efficiency is achieved when teams are kept as small as possible, and have as much overlap as possible. However, this becomes a problem if the designers aren't suited for balancing due to low personal game knowledge or understanding of strategy and game mechanics. While persons in these teams don't need to be highly ranked players, they need to understand the game on a very deep level.

That exposition allows me to answer my second question: Which champions should undergo design changes, and which should just undergo numerical changes?

If an imbalanced champion cannot be balanced over an extended period of time, then the champion suffers from inherent design imbalance and should be flagged for redesign. So, what determines if a champion is imbalanced or badly designed? We have a couple of metrics we could use for this:

1. Pick and ban rates in competitive games, with context of if the champion is a valid counter-pick or a power-pick.
2. Win rates in competitive games (challenger league, LCS and regional equivalents etc), when context is added to determine how much the champion contributed to the win, and how much of that could be allocated to player skill as opposed to champion strength.
3. Win rates in high level online games and in-house leagues.
4. Riot's in-house testing and the PBE.
5. Player feedback across certain or all skill levels.
6. Purchase rates for the champion, with context of other factors that might contribute to purchases (like a recently released anticipated skin).

The best way to go about this is by using all these sources of information. If a champion is constantly performing better than others in all of these metrics despite balance changes, then the problem should be obvious. When the champion is over-performing in some of the metrics but not all, then that decision becomes murky.

I don't have a list of champions that are unhealthy and require changes, because I can't show that the champions I feel are too powerful fit the metrics I provided. Furthermore, there are plenty of other usable metrics besides those I pointed out that could be used for determining champion balance.

So to recap, there are multiple reasons why it takes a while for champions to reach a balance:

• The process itself takes a long time by virtue of favoring design over immediate balance.
• Unless a champion is completely imbalanced, the metrics that determine if a champion is too strong are murky, and different metrics can give opposing information.
• Champion popularity is a key selling point of the game, and the decision to make changes may not be something the design or balance teams can make.

With the difficulty associated in balancing champions, I come to my third question.

In what manner should changes be implemented?

I'm not privy to complete information about how the balance team works, but I can offer my opinion based on my observations.

I feel that the best way to implement changes is by making as many changes as is feasible at once, with longer times between balance patches. This gives us a couple advantages.

1. The PBE becomes a better tool for determining balance because changes can be tested for a longer period of time. When the PBE patch is gradually updated with changes, the designers and players can get a better feel for individual changes, and their influence on the other planned changes. Changes shouldn't be introduced all at once on the PBE, but they should on live.

2. If a change is too strong, then at least counter-play to that new change exists if the patch seemed sound on the PBE. There should be very few situations where it's impossible to win against a certain comp or champion when the patch hits live.

3. The players on Live have more new things to learn or internalize, giving an advantage to those who can adapt to these changes quicker when they go live. Over time, this advantage will wane, allowing some mechanically weaker but strategically more sound teams to cinch a win they otherwise would not get.

This would increase the value that Riot receives from the PBE. The downside is that since the realm should only be populated by a select few, it places further restrictions on those who should be allowed access. Players should be compelled to test the new changes in every game in an organized manner. Mirror-matches should be discouraged for the most part, and focus should be in synergies between newly changed champions and old champions in both casual and competitive contexts.

I do not feel the PBE currently fills the role it's supposed to. Patches roll over to live too quickly from the PBE, and there are few chances to test multiple iterations of changes there.