Design Brainstorming: Psychopomps Are Missing, Pokémon, and Types

I've been replaying Pokémon Omega Ruby, and it's been getting me thinking about the element system in my own game, Psychopomps Are Missing. I have 9 types in Psychopomps: Slashing, Striking, Piercing, Fire, Ice, Lightning, Arcane, Divine, and Abyssal. I like these types, as I feel like they give me good coverage for the full range of damage types I want in the game. The way I'm using them is very analogous to Pokémon: weaknesses and resistances to types use multiples of two. However, unlike Pokémon, what weaknesses and resistances exist is not immediately obvious.

I do try to be consistent among enemy types. For example, the bandits in an early area are generally weak to Arcane damage (there are caster versions that are weak to Piercing instead). However, what types enemies will be weak to is not obvious, nor is there some clear mnemonic one can use to remember. Again, I try to make it feasible to group enemies into categories to remember that way, but this only goes so far and only lets you intuit so much.

By contrast, Pokémon types are distinct and clear about their relationships between each other. While the 18 types and their relationships are a lot to memorize, they remain consistent. Most Pokémon make their type at least somewhat obvious, between general design, inspiration, and color palette. For example, if you see a Pokémon that looks like an insect, good chances are it's Bug type. If it has wings, it's likely also Flying type.

This ability to intuit a Pokémon's type from its appearance, and thus what moves will likely be good and bad against it, is missing from my game. This is something that I find dissatisfactory, given that types are incredibly important to effectiveness. It turns out, dealing two—and especially four—times as much damage due to good type selection has a massive impact on combat. Pokémon's system makes this work well, but I don't feel like my implementation does for a lot of reasons that stem from a very different design.

At this point, I'm thinking I either need to come up with a system that standardizes type weaknesses and resistances, or I need to change to a different system that lowers the emphasis on type. Both are valid approaches, but I'm not sure which is better. Indeed, I'm writing this very article as a way to force myself to properly process this issue!

Both systems are trading pro/con pairs. For example, the Pokémon system makes types matter in a very significant way, but that's both a pro and a con simultaneously, depending upon how much you want types to matter and how that affects the design of your game. As I think about it, part of what makes the Pokémon system work is that there are a lot of types. This makes for more possible combinations.

There's more than that, though: my game has four active party members at once, while Pokémon has only one. The large number of active party members means that ganging up on targets is very strong, which further amplifies the power of strong type effects.

As I think about it, the main advantage I particularly like with sticking to a rigid structure of power-of-twos like Pokémon uses is that it is simple. It makes it easy on me as a designer (this thing is weak to this, that means it takes double or quadruple damage) and it's easier for the player, as they don't have to worry about whether or not a move that's got more raw power is only slightly resisted (e.g. losing maybe 10% damage) or heavily resisted enough to use something with less raw power that isn't so impacted by typing. But without the clear rules of Pokémon, I'm not sure that ease actually exists for players—there is no type chart to look up, no way to verify what types are good against what. You don't have to compromise as much in Psychopomps, either. Characters can use more than four moves, player characters are not locked to specific type weaknesses and resistances based on something (changing gear will adjust these), and so on.

As I'm thinking about it, perhaps a solution to this could be a mix of hybrid stuff. One big difference that I overlooked between Pokémon's elemental system and my type system is that I currently only have damage types. If I introduced defensive types, I could use them as a way to introduce consistencies into the system. And now I'm thinking about types in a more complex way. I also like the idea of adding a display for these types to enemies after the player has defeated at least one.

I'm now thinking I'll have enemy categories (for example, insects or plants) and equipment-based categories (for example, chainmail or robes) that have set values. However, I might make these a bit lower than Pokémon's powers of two. I like this idea a lot!

Time to go make a spreadsheet.

You can support Sientir in his creative endeavors by subscribing to his Patreon.