Status report 2

Meanings and expectations

The word "Balance", like all words, is something people use to communicate with one another, and as such means what people mean it to mean. This article is going to be me giving my thoughts about how the concept of balance (at least based on my guess of what people generally mean) ties into players' enjoyment of the game.

Balance in Duelyst is the idea that certain choices a player can make in their efforts to win matches are equally effective at winning matches ("viable"). For example, the notion that the six factions are balanced is that, when starting to build a new deck, which faction you select won't impact your chance of winning your next match (on average at least: people still say factions are balanced even if some matchups are lopsided, as long as they average out).

To help explain this perspective let me contrast with another common meaning of balance: avoiding intrinsically advantaging one player over another. Competitive Duelyst is a symmetric game from the perspective that ladder and tournament matches don't inherently differentiate between different player accounts. After deck selection, the game is asymmetric, but at the very outset, the situation is symmetric, and asymmetries arise as a result of player choice. This paragraph specifically is presenting the perspective that one can say that "every symmetric game is balanced" in this sense, as a tool for highlighting how the meaning of balance presented in the previous paragraph differs from this.

Usually players report that they would indeed like for the six factions to be balanced in the sense of equally viable. However, there are many situations players encounter in their journey to the end of a match where there is no desire for "balance" in the sense of options being equally viable (meaning effective for winning). For example, players are fine with in-match decision making being imbalanced, i.e. that some plays are better than others.

Moreover, there is generally not an expectation that every deck, as in every single possible combination of 39 cards, be balanced. People not only accept but actively enjoy that deck building is a skill, and for it to be it must be possible for one to build both good decks and bad decks. I don't think there's any desire for e.g. the deck one obtains by clicking neutral cards in order starting from the end to be as powerful as the decks people are actually playing.

There are some common situations where it's not entirely clear what is meant by "balance". When people ask for archetypes to be balanced I think they mean that the best possible decks from each of the archetypes be balanced. Pretty often it's unclear where an archetype ends and the next begins, and also in those cases the spirit of the request for balance among archetypes is often undermined.

When people ask for individual cards to be balanced, I think that's sometimes about how often a card appears in optimal decks, but actually I think more often it's about how in-match gameplay feels, and is a pretty separate usage meaning of the word "balance" than what I've been talking about so far. The in-match feeling that often causes players to call an individual card imbalanced comes from when that card excessively dictates play in the match: if matches are feeling like they're all about one card, that's when usually people call a card imbalanced. To highlight the difference between imbalance because of matches revolving around one card vs. imbalance because of prevalence or lack of equivalence in choices, consider for example that Healing Mystic is more common than many cards that get called imbalanced more often than Mystic.

It's important to realize that the notion of balance meaning equal viability across some set of options necessitates a choice of said set. Often the choice of set is left implicit, and communication is harder when it's unclear if discrepancies stem from differences in set choice ("intent") or evaluations of viability ("execution"). Similarly, the notion of balance pertaining to in-match impact has both an intent component and an execution component.

Roles of balance

One must keep in mind that the ultimate goal of Duelyst is to be "fun". This is a subjective metric based on human emotion, and I think "fun" here is by definition the feeling that Duelyst is striving to evoke in players. In other words, if you keep asking "why?" about Duelyst design decisions, "It's fun" is the answer you end up at, at least before having to move outside of the realm of game design.

Recognizing that balance is not an end goal and does not have intrinsic value, we're lead to consider the question of when balance is fun and when it'd be preferable to compromise or eschew it. This then leads to another component of what I've been calling "intent". Here is an example:

Earlier I said that players often want a pretty wide range of archetypes to be balanced and optimal. However, an important competitive element of CCGs is picking a strong deck, and accompanying it the skill of discerning a deck's power level at the level of archetypes. There is often an expectation that this important competitive element will be relevant for determining the winner of a match, and that players' masteries of the associated skills will be tested. If, by design, archetypes are equally powerful, then this deck selection component is lost. For this reason, many players, the players who think it's fun to care about deck selection, do not want balance among archetypes. Instead, these players prefer when what is optimal is opaque. Not only does that mean they can puzzle over what they find fun to puzzle over for a long time, but, essentially by the definition of opacity, it allows for bigger gaps in power levels, leading to proficiency in deck selection skills being better rewarded.

I've found that players often view balance as an objective thing, probably because the execution component of it is in principle objective. However, I think viewing balance as an objective metric overlooks the necessary and subjective component of choosing an intent. Moreover, my experience is that the latter typically has a much larger impact on the end result. Overall, my opinion is that balance is largely subjective.

Novelty and richness

A core principle for me in game design of the sort relevant here is that there is an implicit "contract" between players and designers: players should try to win matches, and designers will ensure that that experience is fun. Tools for creating fun are important for designers, so that they can hold up their end of the contract.

Looking back at the examples discussed thus far it jumps out to me that one major source of fun for Duelyst players is learning, viewing many aspects of Duelyst as a puzzle and taking pleasure in discovering elements of that puzzle and its solutions of sorts. This concept was touched on briefly in the previous Status Report.

For the process of discovery and learning in Duelyst to be engaging and fun, it can't be obvious what the most effective way to win matches is. Two tools that help avoid this are what I'll call "uncertainty from novelty" and "uncertainty from richness".

Uncertainty from novelty is the idea that it's difficult to evaluate situations that are substantially different from all past experiences. When radical changes are made to the game, e.g. through the introduction of new cards, uncertainty from novelty is being leveraged to create opportunities for learning and fun.

Uncertainty from richness describes situations where a great deal of experience and expertise is needed to reach clear answers. Chess has a tremendous amount of this for instance.

In principle, uncertainty from novelty and uncertainty from richness operate independently from each other: having more or less of one doesn't necessarily constrain the other in any way. In practice, however, there is often tension between them, and putting an emphasis on one frequently comes at the expense of the other. One direction is that novelty is wont to undermine the relevance of deep, fine knowledge of minutia. Ergo conversely making use of uncertainty in richness naturally incentivizes stability in design.

Delineations associated with these two sources of uncertainty are fundamentally subjective, and I think of them as soft and fuzzy ways to describe narrow aspects of player experience. In particular, they hinge critically on the evaluation skills of players. When players are able to feel confident in their assessment of how best to win matches, the lustre generated from the uncertainty fades.

From this perspective, balance, in the sense of apparent equality in viability, is relevant. This too manifests a tension: design emphasizing novelty often yields compromises in the execution of making power levels be as intended, which is a key component in making compelling design that emphasizes richness. One way to frame this is through consideration of the notion of a "threshold for mystery" beyond which players don't feel that they've yet discerned what is optimal for winning matches. Uncertainty from novelty makes this threshold of mystery large, but relies on design which usually makes differences in viability large. Conversely, uncertainty from richness corresponds to small thresholds for mystery, and thus necessitates a certain type of balance.

To give some examples of how the balance framework that's been presented can help reason, here are ways in which one may frame a couple familiar views that highlight their wisdom.
"Balance doesn't matter" - One of the major strengths of uncertainty from novelty is that it functions even when the difficult execution of intended power level is lacking.
"If you can tell that something is balanced then it isn't interesting" - After players feel like they've crossed the threshold for mystery, they no longer have the opportunity to have fun through discovery and learning, and designers might now be breaching their side of the implicit contract.
"My opponent's deck is bullshit" - The details uncertainty from richness is asking players to ponder aren't something players are interested in thinking about; there is an issue with intent, in contrast with execution.
"The meta is stale" - Players yearn for the discovery that stems from uncertainty from novelty as it evokes a unique kind of fun; now would be an opportune time to design with an emphasis on novelty, even at the expense of other sources of fun.

Design directions

The concepts related to balance that I've presented above help inform the design of Duelyst: Origins. I believe different situations, different states of the game, call for different approaches. Each of the various metas, configurations of card designs, and community sentiments will naturally cause certain approaches to fun to be more or less effective, and it's essential to be flexible, seizing opportunities and playing to strengths.

With cognizance of balance's lack of intrinsic value and instead having relevance insofar as pertaining to fun, I anticipate that Duelyst: Origins will frequently find itself in situations where the greatest benefit comes from choices which eschew emphasis on equality in the viability of strategic options. There are many ways for things to not all be equal, and matters of intent are of central importance.

As mentioned in the previous Status Report, the processes of discovering and learning are major sources of fun for many players. Uncertainty from novelty and uncertainty from richness correspond to tools for creating this fun, which vary in effectiveness depending on the situation. I anticipate a natural ebb and flow in our use of these tools, as further novelty in the immediate wake of a major shakeup is is naturally harder to make compelling than usual, but is unusually compelling after longer periods of stability. Moreover, sometimes the game may find itself, for example, in states where the process of discovery after major changes is particularly deep and constantly evolving, or states where the fine considerations in expert play are particularly gripping, and I believe we are best served by staying attuned to such things and adapting our choice of tools appropriately.