The Stuff of Legend: Negative Design

Design and development for games can be a lot of fun. We get to throw new ideas against the wall and see how they stick, then mix them in with what’s already there and see what comes of it. Nobody at Rule of Cool Games does this more than me; my unofficial job is to barge into meetings with half-baked concepts and throw them at people until someone takes up the challenge. Positive design (“You can do this now!”) is the impetus that gets a lot of us going on a new project. After all, more often than not, delineating what you can do and opening the door for people to explore through means constructing the appeal of the game.

For the first Monday of August, I’ve decided to turn the column toward a trickier subject: negative design. The Legend system in particular dwells on negative design to maintain a state of balance. Fury and Precision don’t get along. Nonlethal conditions that might completely sideline a character are hard to inflict and are never permanent. Internal development limits exist for how and when things scale, and design is restricted in certain areas based on what’s already in place. Negative design is the work of saying “no,” and while it doesn’t always take such a solid and intractable form, it’s hard to fight the sense that negative design tells you to take that bright, colorful new toy and don’t play with it in the house.

Today, therefore, I’m going to address what many have commented on as being unusual exceptions to Legend’s largely free-for-all multiclassing model, detailing the negative design decisions that have gone into this state of exception and whether such decisions will be made in the future. I’d like to use the following to provide answers to the question: When should negative design be employed?

For emphasis: Legend has a small number of tracks that are locked to the class that features them. This lock both prevents multiclassing into them and multiclassing out of them. You either have the track and the class it comes with, or you are of a different class and cannot get the track. As negative design goes, this is a hard limit: there is no way around it within the rules, and as such there is no way these tracks can coincide on a player’s sheet. Esoterica Radica, Judgment and Shaman’s Path all stand apart among the 50+ tracks that comprise the core of Legend for being expressly incompatible, not by dimidiation or resource conflicts, but by requiring you to spend a non-track resource (your class) to get them.

The reason for this exclusivity is twofold, and begins far in Legend’s past. The first classes to transform into track structures were the Paladin and the Rogue, and at the time the design goal called for each of those classes to have an essential skeleton of “paladin-ness” or “rogue-ness” and choices to play off of that skeleton (note that these classes have the highest number of selectable tracks). That these tracks were basic foundations for a Paladin or Rogue meant that the core features of someone playing either class could be moved there to remain independent of your other track choices. You could construct your particular flavor of Rogue or Paladin without worrying about loss of vital functionality and balance, and the skeletons could reference involvement in certain abilities (Bastion and Once More!). Had Legend design not shifted toward vertical multiclassing and away from encouraging single-class play, the other classes would most likely each have a similar track today; however, as priorities changed to promote more flexibility, other classes were set up with less directly interrelated abilities.

When the new development team came in to work on 1.0, the question of these exclusive tracks was raised. Ultimately, given their history, design, balance and appearance in play, two overriding factors pushed the developers to retain the exclusivity. Firstly, each of these tracks represents, in some way, power a bit over the curve. That is not to say that they are imbalanced; rather, they meet the power level expected in the game coming down the stairs, so to speak. The value of establishing a precedent for the upper curve paired with the difficulty and general unpopularity of rebalancing something that was not in and of itself problematic, leading us to conclude that the game was better served by keeping “great-but-not-too-great” discretely available. Secondly, each represents a great flavor structure – not an explicit fantasy flavor that we expect people to conform to, but the broader strokes of the tropes that Judgment and Esoterica Radica each represent. Shaman’s Path gets this a bit less so; it’s intended to represent a Shaman’s chosen devotion, whether to the ideals of a deity (compare D&D’s cleric domains) or perhaps a manifestation of spiritual power in the physical realm (go on, use it for Vigilante. Drive a car spawned from the star-dreams of Xibalba).

By making these sorts of choices and offering mechanical reinforcement, we can create alternate, more experimental kinds of chassis on which you can build your character. Will hard restrictions be a common negative design choice in the future? Absolutely not, but establishing a precedent justifies future experimentation. Emphasizing certain characteristics of a chassis, a track or a class creates a new way of thinking about character-building. We won’t do it often, but when we do, it’s because a conclusion has been reached that it’s worth doing.

To partition, streamline and guide: Some things just go very well together. People tell me peanut butter and chocolate make for a lovely unison, though for the life of me I’ve never understood why. Some things just don’t. Peanut butter and chocolate oh fine. Tuna fish. Peanut butter and tuna fish. And licorice. I’m pretty sure that would be awful.

In a game like Legend, where character creation consists of a vast amount of choice, it’s important for us to work against trap choices to prevent new players from feeling like choice is just an illusion. The skill gate in Legend exists where people step away from playing a class straight with a full buy-in or with the free multiclass and move toward planning out complementary actions, checking math on comparative offense vs. defense requirements for a given level and sketching every character as though a “class” was some boring thing you had to attend earlier in the day before getting in to play some Legend.

What is the skill gate? It’s the partition between play and play, between casual and expert, between pick-up and pro. Simply put, we look at the skill gate as the margin between sitting down at the table to join your friends for their game and sitting down at the table ready to make the game yours. It’s so named because the transition requires picking up skills in character-building, math and creativity to work on your own ideas without a guide or someone else handing you a build. It’s an important concept to recognize in a game where mastery is possible and beginner’s luck isn’t a practical factor. Working on only one side of the skill gate results in either a higher upfront time investment required to participate or zero reward for getting deeper into the game and its rules. We try to cater to both groups: the new players who want to learn and the experienced players who want to rewrite expectations of how the game plays.

For the first group, the people learning the system and those who just want to pick up and go, peanut butter and chocolate is where it’s at. Things that inherently complement each other right out of the gate let you feel powerful and relevant, providing the opportunity to learn the ropes of playing the game without demanding mastery of character-building beforehand. For the second, you’ve got your Heston Blumenthals who will take peanut butter, tuna fish and licorice, laugh at the notion that they can’t possibly go together, and whip up something preposterously creative that shatters your notions of what tuna could be. The skill gate exists, and is the reason behind tracks that call on a specific ability score.

For most tracks in Legend, we’re quite pleased to key abilities off of KOM or KDM (and occasionally other special ability modifiers). For several tracks (particularly Barbarian and Paladin tracks), however, a specific ability score is called out: Constitution for Rage, for example, or Charisma for Smiting. These represent a soft limitation; while you can still build a character using both Celestial and Smiting, you have to understand that they don’t lean on the same ability score. You have to plan out a way to ensure that both Wisdom and Charisma are relevant to your character. You get the opportunity to be Heston Blumenthal, putting together odd combinations to wondrous effect (for instance, Shaman’s Path to take Smiting and toggle it to your Wisdom modifier). The skill gate will always exist; as long as it’s going to be a necessary weasel, it may as well reward you for being invested rather than penalize you for impatience and unfamiliarity.

Tracks that directly reference ability scores are a great tool to emphasize a number of things in the game; the inherent compatibility of a track and your class or racial chassis, for instance, or the utility of using two tracks in tandem. Some unusual choices are there to demonstrate that the roles an ability score might play in a different game (perhaps one with prisons and pyrolisks) do not limit its use to your character in Legend. Others are selected to contribute to their desired playstyle in ways both direct (CON will give you DR to help you out in melee, Ragers and Dervishes!) and indirect (the more terrifying your Terrifying Presence is, the harder you are to personally intimidate and the better your Will save is likely to be). These soft limitations can be danced around with Shaman’s Path, Multiclass Flexibility and other clever planning, but serve an important function as guideposts to carry newer players through the skill gate.

To mitigate trap choices: While it’s awesome to be an Army of One, it can be debilitating to have to split resources between two different major competencies. Plenty of players of some other game (likely involving warrens and wyverns) have discovered to their chagrin that uniting the awesome powers of the divine with epic sorcerous power leaves them little better than the poor second cousin of either cleric or wizard. The warrior tired of being kited at a distance may find himself incapable of effective combat either in melee or at range as he swings with a bow and shoots with a sword.

Different playstyles that fit well under the same general banner may work terribly with one another. Apples and caramel? Good. Chocolate and caramel? GOOOOOOD. Apples and chocolate? It’s not awful, but when’s the last time you had anything featuring both together? When we look at the ranger, with Professional Soldier and Battle’s Tempering, we have two skillsets that are effective and useful both for a melee combatant and a ranged expert. However, with Reign of Arrows and Iron Magi coupled into one track, there exists a hard limitation preventing them from coexisting. Dimidiated tracks like this are uncommon, but serve to divide out two options that are at odds with each other, typically because of a serious standard action conflict. Can you still find ways to screw yourself with conflicting standard actions? Absolutely, but we’re not going to pack them together and say “here, this is smart, you should try it out.” It’s a fundamental lesson of action conflicts being taught via forced choice.

Never: When all that is said and done, our priority with future content is going to almost always be maximum flexibility with maximum compatibility. Multiclassing is fun. Builds are fun. If we didn’t believe that, if we didn’t have daily confirmation of that from fans, we wouldn’t be pushing tracks. You haven’t seen a new class out of us yet (repeat: yet) because the basic unit of character-building is simply so popular. Hard and soft limitations born of negative design will appear on occasion, but we don’t stand by legacy come rain or come shine.

Case in point: Guild Initiation. For those of you with us prior to 1.0, you may remember that at one time, Guild Initiation was a prerequisite for taking either Mechanist Savant or Knight as a track. That’s right – tracks with external prerequisites. You’ll note as well that this put them in conflict, as they competed for a resource they could not share. At the time this design was still current, the thinking was that certain tracks requiring GI as a prerequisite would allow them to be more powerful as a tradeoff, since players couldn’t have more than one and there would be an associated cost offsetting the extra strength.

Unlike the situation with ER and Judgment, which had distinct advantages to development, we came to see the concept of tracks powerful enough to require not just exclusivity but a feat tax on top as a poor choice for both balance and play purposes. It wasn’t fun to sacrifice a feat slot simply to play with an interesting track, and it set a precedent that we would either have to continue using (making it less and less fun each time) or abandon (leaving a bizarre and unpopular legacy tumor in the core rules.)

This is where development stepped back and looked at negative design values: what was it emphasizing, was it serving the interests of new players, and did it fence off trap choices? We determined that it emphasized imbalance and exclusion; that it told new players “cool stuff doesn’t come without a cost;” and that the two tracks didn’t feel like they were setting up a trap at any level that any other two given tracks weren’t. From a positive design standpoint, we wanted to remove the prerequisite. From a negative design standpoint, there was nothing the prerequisite was doing that hadn’t already been accomplished by making the tracks optional instead of class-based. So we threw it out.

Negative design has a role to play and gives character to choices and options, but it’s always going to be a secondary tool in the Legend developer’s kit. The fact that exceptions are questioned is a good thing to us, as it means they’re exceptional enough that our positive design vastly outweighs the negative elements that help give the game its underpinning. I look forward to seeing your thoughts on the forums.

Mr. A


Leave a Reply

Your email address will not be published. Required fields are marked *



HTML tags are not allowed.