E6 in PF2

 I plan to use a bunch of jargon in this post.  I may or may not explain it.

So to explain at least what the title means:  PF2 is, of course, Pathfinder Second Edition.  E6 refers to "Epic 6", a variant invented for D&D 3.5.  The exact details of how it's implemented in 3.5 aren't needed here, but the important parts to know if you haven't heard of it before is that in E6, once you hit level 6, you stop gaining levels.  Instead, every (some number) of XP, you gain a new feat.

Since everything is a feat in PF2, it seemed to me like it might be interesting to implement this idea in PF2 and see what happened.  These mechanics have not been tested and are probably not totally complete, but people were talking about PF2 on twitter and it made me think I should post these up.

First, the core:

E6 Core Rules

-Once you reach level 6, you cannot gain any further levels.
-Every (X) XP after reaching level 6, you can choose a new feat.  You can select any feat that you qualify for, in any category.  The exact number of XP should be chosen by the GM/group to calibrate the rate of advancement to the desired rate.  Some good defaults are 1,000 XP and 500 XP (so you gain feats at the same rate you would gain levels or double the rate, respectively).
-You cannot gain Master or higher proficiency in any game element that your class does not already have Master proficiency at level 6.  You cannot increase your proficiency rank for your Class DC or weapon specialization effect beyond what it was at level 6, except with the Expert Attacker feat (described later).  You cannot gain a class feature higher than level 6 that increases your proficiency rank with any game element.
-You cannot unlock, learn, or cast a spell of 4th rank or higher.  Any cantrips or focus spells you know never scale beyond 3rd rank.  You cannot increase your level for Spellcasting past 6.

That's the core set of rules; with just those, you can get the basic E6 experience.  But to make it better, we have more to add.  You might notice that the core rules reference that you can't gain a class feature higher than level 6 that improves your proficiency; if you stick with the core rules, there's no point in mentioning that, you can't gain new class features.  But if we add new feats, that suddenly becomes an option.  And we have some new feats to add that will help let characters choose to increase some tricks or balance out a few things that might come out odd at 6th level.  It's also worth thinking about whether you want to use the optional rule for Proficiency Without Level (Archives of Nethys can auto-calculate this for monsters so you don't have to subtract it out yourself).  If you do use PWL, you unlock a significant amount of content for yourself, making a lot more monsters usable.  If you don't, it contributes to a dangerous world where there exist monsters that the PCs simply cannot fight in a meaningful way; a level 6 PC against a level 15 dragon, no matter how many extra feats the level 6 character has, is not going to go well.  The sheer weight of numbers that the dragon has will crush the PC.  (Even with PWL, the dragon's higher proficiency will give it an advantage that might be insurmountable, but it'll be less impossible.)  It's up to you which of these options you prefer.  And now, feats to use with E6.

New feat - Expert Attacker

Prerequisite:  You are level 6, have at least 5 additional feats, and have no way to apply Expert or higher proficiency to an attack roll.

Choose a weapon, spell type, or other method of attack that you are Trained in.  You become Expert in that attack form.  If you choose a spell type, you also become Expert for save DCs.

New feat - Hardiness

You gain hit points as if you had gained a level in your class.

Special:  You can take this feat multiple times.  Its effects stack  Each time you take this feat, increase its rarity by one step (Common, Uncommon, Rare, etc).

New feat - Increased Spell Capacity

You can cast one additional 1st-rank spell per day.

Special:  You can take this feat multiple times.  Each time you take it, increase the rank of the extra spell you can cast until you gain a spell of your highest available rank, then reset it to 1st level next time.  For example, a 6th level sorcerer can cast spells of 1st, 2nd, or 3rd rank.  The first time they take this feat, they can cast an additional 1st-rank spell.  The second time, 2nd-rank; the third time, 3rd-rank.  The fourth time they take this feat, they get another 1st-rank spell.

New feat - Increased Focus Capacity

Prerequisite:  Increased Spell Capacity x3

You gain an additional focus point.  Your maximum number of focus points increases to four.

Special:  You can take this feat multiple times.  Your maximum number of focus points stays four, but if you took the feat and still didn't have four points, you can keep taking it until you get up to four.  (You would probably rather learn more focus spells but it's an option if you want to.)

New feat - Advanced Learning

Increase your effective class level by two for the purpose of qualifying for class feats, class features, and ancestry feats.  You do not actually gain any features or feats when you take this feat.  You cannot gain a banned feature or ability through this.  You do not gain hit points, spell slots, increase your proficiency bonus, or any other effect of gaining levels.   If you take a class feat or class feature higher than 6th level, it may need to be scaled down as described below.

Special:  You can take this feat multiple times.  If your effective class level is 10 or higher, this feat is Uncommon.  If your effective class level is 14 or higher, this feat is Rare.  If your effective class level is 18, this feat is Unique.

New feat - Advanced Class Feature

You gain one class feature above 6th level that your effective level qualifies for through the Advanced Class Learning feat.  This feature cannot contain any banned elements and may need to be scaled down as described below.

Special:  You can take this feat multiple times.

Scaling Down Items and Feats

Many items have cool powers but are too high level; they can easily be scaled down.  If the item deals damage, its damage should be reduced to around 6d6 damage (21 average) (and lower if it has utility effects, large AoE, etc).  If it gives an item bonus to something, the bonus should be no more than +1.  If it has a save DC, the DC should be around 20.  If it has an attack bonus, the bonus should be around +10.  If the item is scaled down to a level below 6, these numbers of course should be even lower.  Use the tables for creating items with magic item bonuses, DC by level, etc, in GM Core.  (Remember an attack bonus is just a DC - 10).  If an element would give resistance, it should be no more than 5 if permanent, 10 if temporary, and halved if it is a particularly broad resistance.

The same is true of higher-level features or feats that may be unlocked.  In many cases the GM's discretion will be required to decide what the exact effect available from a higher level feat or feature will be, or if it's available at a choice at all if it cannot be scaled down to an appropriate level of power while retaining the essential effect.  Some feats or class features may simply deal flat damage and can be scaled down, while others provide significant non-numerical power such as teleportation or flight that are inappropriate and can't be easily scaled down.

D&D-Like Ability Scores

 It turns out I can just post whatever I want here and no one can stop me.  

Ability scores are boring in modern versions of D&D.  If your ability scores are randomly generated, then they make sense as-is; but if you can arrange them or use point buy or standard array, there is a trivially correct choice for almost all characters, and that's boring.  Here are a few different options for how to make them interesting again.  These are written from a 5E perspective because I'm just recycling notes out of my ideas file and not really writing them from scratch, you can probably convert them to any D&D-like game without too much effort.

Option 1: Redefine Ability Scores

Redefine the six stats as follows.

Strength - Your ability to project force into the world, whether physical or magical.  Whenever you would add an ability score to a damage roll, add your Strength modifier instead.

Dexterity - Your precision and aim.  Whenever you would add an ability score to an attack roll, add your Dexterity modifier instead.

Constitution - Your health and survivability.  Continues to add to hit points.  No changes except based on what the other stat changes do.

Intelligence - Your ability to do skillful things.  Whenever you would add an ability score to a skill check, add your Intelligence modifier instead.

Wisdom - Your strength of will and ability to enforce your will on the world.  Whenever you would add an ability score to a save DC or use as part of another class feature, add your Wisdom modifier instead.

Charisma - Your sense of self and ability to keep yourself together.  Whenever you would add an ability score to a saving throw, add your Charisma modifier instead.

(Wis and Cha could easily be swapped, idk, I wrote this in two minutes)

With this method, suddenly it's no longer just "Wizards want Int, fighters want Str".  Every stat becomes useful for every character.  (PS - I stole this idea from Pillars of Eternity.)

Option 2: Remove Ability Scores (Flexible Version)

Remove the 1-20 ability scores.  Keep the ability score names.

All characters are proficient in two ability scores, and deficient in one, of their choice.

If you are proficient with an ability score; when you make an ability check, attack roll, or saving throw based off that ability score, add your proficiency bonus.  (This is separate from adding your proficiency bonus if you're also proficient in that roll, so you'd add it twice if you are proficient in both Strength and Athletics and you made a Strength (Athletics) check.)

If you are deficient with an ability score; when you make an ability check, attack roll, or saving throw based off that ability score, subtract one.

If your class references an ability score for a reason (such as save DC or uses per day), use your proficiency bonus.  Optionally, you could require that the character actually be proficient in the referenced ability score in order to use your prof bonus.

You can use an ASI to become proficient in an additional ability score or select a feat.

Option 3: Condense Ability Scores

Remove the six stats.  Replace with Fortitude, Reflex, and Will.  If the game refers to Str or Con, use Fortitude; if it refers to Int or Dex, use Reflex; and if it refers to Wis or Cha, use Will.  Exception:  Hit points gained by level are a flat value for all PCs chosen by the GM based on campaign style (+0 for life is cheap, +2 default, +4 for superheroes) and are not gained via Fortitude.

You have proficiency in one of Fortitude, Reflex, or Will based on your class.  You are deficient on another one, and neutral on the third, also based on class.  Some classes might let you pick between options.  You can go from deficient to neutral with one ASI; neutral to proficient takes two ASIs.

When making any attack or replacing any stat in a class feature, use the new stat that you're proficient in.  (For example, fighters are proficient in Fortitude; a fighter making an attack would use Fortitude for the attack.  Clerics are proficient in Will; a cleric's attack would be based on their Will.  This is true whether the cleric is attacking with a spell or a staff.  Clerics would also use their Will in place of their Wisdom modifier if a class feature says, for example, you can use this Wisdom modifier times per day.)

Diegetic Character Advancement

This post is an overview of different types and methods of diegetic character advancement.  First, a definition.

Diegetic Character Advancement: Advancement based on specific in-world actions, items, or events, rather than abstracted or based on the story as a whole.  Generic experience points would be an example of abstraction, while milestone leveling would be an example of story-based progress.  These options listed here are neither of those.

Why would you want this?  Diegetic character advancement as a concept has two major benefits to it.

1 – It rewards players for taking the actions that you want the game to be about.  If you make a game that’s about shoveling snow, and you reward players for shoveling snow, then players are going to want to shovel snow in game.  They’re not going to want to use a snowblower to get rid of the snow more efficiently or a flamethrower to get rid of the snow in a more ridiculous way, they’re going to want to shovel it.  (Sorry it was snowing as I wrote this.)

2 – It is extremely intuitive and understandable.  By giving players direct goals that interact with the world, players know what they need to do to advance, and they know how to do it.  It being so understandable reinforces player attachment to the fictional world and helps them treat it as a more real place that matters, instead of as a series of setpieces that they travel through without attachment.

As an overview, these types will be listed and given a brief description, but not fully detailed with everything about them, because I don’t have that much time.  In theory future posts might cover one or more types in more detail, but for now we’re just going over the basic things about them, pretty much just what they are and a simple list of pros and cons.

Types of Diegetic Character Advancement

Note:  There is no requirement that a game, or a campaign, use exactly one of these types of advancement.  It is fully possible to mix them, and this is particularly true if using specific advancement; certain game elements might use one type of advancement (say, skill feats being unlocked only through job paths), while other game elements use a different type of advancement (spells being unlocked only through discovery).  You can do pretty much whatever you want with any of these advancement types, it just adds more work the more of them you use.

For each type listed here an example game or source that uses them is listed.  These are not exhaustive at all and are just off the top of my head when putting this together.  Some of them are RPGs, some are video games, and at least one of them is something I wrote that hasn’t been really publicly released.

Gold for XP (Used In:  Older versions of D&D, OSR games)

Or, more generally, advancement through collection of in-game currency or other items.  Whether it’s actually 1 XP for 1 GP isn’t important so much as it being advancement through collection of valuables in-game.  This is distinct from some other categories because the currency or item collected can be used for in-game things as well as for gaining XP; after you get your XP from the gold collected, you can then spend the gold, you don’t need to consume it or any such thing.

Pros: Compatible with existing systems, doesn’t require tracking anything that you weren’t already tracking, intuitive, matches existing incentives.
Cons: “Collecting massive piles of gold” as a primary motivation doesn’t suit every character or campaign style.  Still uses XP as a generic advancement currency.  May require numeric conversion to provide a satisfying advancement rate, depending on game system being used with, if the game was designed with a different advancement mechanic in mind.

Discoveries (Used In: Skies of Arcadia, random reddit post I’m referencing without a link)

One such example of this would be that reddit post where a GM had scattered obelisks around their world, and when the PCs found one, they gained a level.  Another example of discoveries is in Skies of Arcadia, where locating specific landmarks in the world can be reported for a reward.  Discoveries here refers to specific terrain features or environmental effects, and locating or exploring them provides advancement.  (Whether it’s a whole level, a specific feature gained, or a generic partial progress are all viable variants of this method.)  An argument could be made that Discoveries is a subset of Achievements, below, (achievement to reach a specific place), but I separate the two generally in that Discoveries are always a specific location in the game world, while Achievements can be anything but usually require repetitive action.

Pros: Joy of discovery and wonder at finding new parts of the world, incentivizes exploration, provides opportunity for GM to focus on making their world cool and let the players experience it.
Cons: Requires careful GM placement to feel good about how often they’re found, without feeling like they’re guaranteed to appear in the players’ path or otherwise unearned.

Job Paths (Used In: Aces and Eights)

Job paths are when you select a path of advancement, and it provides you with the conditions to advance.  For example, a Fighter path might say “Practice your sword”, while a Dentist path might say “Treat a patient’s sore tooth”.  The biggest distinction between job paths and other types listed here is that job paths allow not only different characters to have different goals required to advance, but allow the same character to have different goals at different times, if the game lets you change between job paths.  (In Aces and Eights, you can change your job path each session, but you only progress towards the goals of your current job path; if the Dentist job path says to treat 50 patients, any patients treated while you’re a Fighter doesn’t count.)  It is also worth noting that these job paths are not classes, and do not affect your actual abilities, just the conditions under which you gain advancement.

Pros: Allows players to define their own advancement in ways that align with their goals, requires little to no GM prep after defining the paths themselves.
Cons: Bookkeeping is required to track goals, higher player investment required for them to select the correct path for them.

Achievements (Used In:  Every Video Game)

For example, kill 100 jerks, collect 14 eggs, slay a dragon, do this list of things.  An achievement can be a single task, a collection of tasks, or a counting repetition forming a collective task.  Achievements can have additional narrative weight added to them by including flavor to what is functionally the same requirement; there’s not a lot of difference in terms of challenge between ‘slay a dragon’ and ‘drink the blood of a freshly slain dragon’, but significant narrative weight to the latter.  (At least, not the way most people would run it.  If the requirement is that you personally deliver the killing blow to the dragon, that actually is different.)

Pros: Familiar to most modern players, strong gratification when completed, provides strong control over advancement to players.
Cons: Requires significant setup in creating achievements for advancement options, repetition can feel like a grind that’s difficult to complete, requires carefully watching events to see if they match any of the achievements (since you can unlock any achievement at any time, not just a subset).

Consumable Items (Used In: Godeaters of the Astral Sea, Sorcerers of Ur-Turuk)

In Sorcerers of Ur-Turuk (which is sort of a d6 system Ars Magica in the desert), there are a variety of magical skills.  The titular Sorcerers go on quests to retrieve items that can be consumed to provide progression towards those skills, with each item being worth progression on a specific skill (to use a D&D example, you might find one item that boosts Evocation or another that boosts Conjuration).  Godeaters of the Astral Sea does it a little more generically, with crystalline godsblood being consumed to raise what is effectively your level cap.  In either case, these are examples of a consumable item; characters must physically locate and consume an item in the game-world to progress.  This is distinct from Gold for XP because the item is not generic, but specific and rare; Gold for XP refers to any kind of currency or fungible item, while Consumable Items are non-fungible.  (They’re never tokens, though, these things actually have value.)

Pros: Provides another kind of exciting loot to find, lets players portion out progression between party in-character
Cons: May require loot distribution between players, potential for in-game shenanigans around progression

Equipment-Based Advancement (Used In: Terraria, other survival games)

With this type of advancement, you don’t necessarily have character abilities.  Your abilities are your equipment (or, at least the type of abilities that are controlled by this advancement type are like this).  If you change your equipment, you change your abilities.  If your character dies, that’s fine, they were just a carrier for the equipment; as long as the equipment survived and was retrieved, your new character can put it back on and have the exact same abilities immediately.

Pros: Highly customizable character options, very understandable progression, exciting loot options
Cons: Limited character investment, possible loot competition between players

Faction-Based Advancement (Used In:  Roguelites, MMOs)

Faction-based advancement is similar to equipment-based advancement in that it has skills and abilities provided by a source other than your character.  As you progress with a faction or factions (whatever ‘progress’ means in your context, but usually it’s filling up a bar somehow), you gain new abilities.  These abilities are maintained between your characters if your character should die or retire, because they’re granted by the faction. 

Pros: Ties characters to in-game groups of NPCs, investment is maintained between characters, provides social choices
Cons: Limited character investment, doesn’t always offer opportunity for players to set their own goals, may cause some players to feel pressured to join a group they don’t totally agree with for mechanical rewards

Expenditure (Used In: Godbound, some OSR games)

With Expenditure, advancement occurs through expending a currency.  This currency may be something specific for advancement, or it might be useful for something else.  For example, in Godbound, characters earn Dominion points, which they can spend to alter the world.  Godbound characters can’t gain levels until they have spent sufficient Dominion.  Some OSR games use a carousing rule, where instead of earning XP for finding gold, you earn XP for spending it.

Pros: Ties characters to find a reason in-world to spend, rewards players who use their resources
Cons: Not all players want to spend in these ways, still requires collecting the currency.

Practice-Based Advancement (Used In: Cyberpunk 2020)

With practice-based advancement, you improve at a skill or feat by using it.  If you want a new ability, you try to use it untrained until you practice it enough to unlock it.  For example, in Cyberpunk 2020, at the end of sessions each of your skills gains IP based on how you used it that session.  You can spend the IP to improve that skill – and only that skill, you can’t spend it on anything else.  This can be done more generally if desired, for example by splitting into categories like Combat, Social, and Magic, and rewarding progression that can be used anywhere in the category when a part of the category is used.

Pros: Possibly the most intuitive system there is, rewards players for doing the things that they did
Cons: Large amount of bookkeeping, advancement rates can be challenging to get right

 

A lot of these options (or perhaps diegetic character advancement in general) are best used in games that have an assumed baseline of character power after which additional power is good to have but not necessary for basic combat effectiveness, such as Godbound and E6.  The reason for this is because when advancement is tied to specific in-game events, the exact amount of time for those events can be estimated, but not guaranteed.  Abstract or story-based progress can be guaranteed to occur at specific times or specific rates; diegetic advancement being based on the actual events in-world can only be guaranteed to the same degree that the events of the game can be guaranteed.  If you are running a game that includes player agency and the events of the game in-world are not guaranteed, then this also introduces uncertainty into your progression.  Uncertain progression that results in some characters having strongly different levels of combat effectiveness from others can be frustrating for groups (though it is also specifically desired for other groups and games; this is a question about asynchronous advancement and the options and effects related to it, which is out of scope for this particular rambling.)  Even with fully synchronous advancement, uncertainty about progression can make gameplay more difficult or less interesting if there are troughs.  It can also make it more interesting.  Having the level of baseline effectiveness guaranteed, and advancement providing things that are useful but not needed, smooths this out and prevents it from causing issues.  You don’t need to have that, but it’s nice.

Specific Rewards vs Abstracted Rewards

Any of the listed types of diegetic character advancement can be split into two general categories; specific rewards or abstracted rewards.  Specific rewards is what most people will think of when they hear or talk about diegetic character advancement, and in my opinion, is the real way that we get a strong benefit out of it.  But some of the same benefits can still be gained from generic rewards so I’m mentioning it as an option.  In a sense, modern D&D-type experience points can be considered a type of diegetic character advancement, because they are rewarded purely from combat* and serve primarily to make you better at combat by gaining level; that would be an abstracted reward.  They aren’t really diegetic the way they’re usually used, because most campaigns do not do the worldbuilding support that’s needed to make experience points actually fit into the world; but if you did, then they could be a weak form.

*(Yes, I know a lot of people reward XP for noncombat tasks, and some DMGs depending on edition mention this as an option.  We all know it’s basically just for combat though and any noncombat rewards are an afterthought.)

The distinction between abstracted rewards and specific rewards is what you put down on your character sheet after doing the thing.  After you complete whatever in-world task it is that allows you to advance, what form does the advancement take?  If the advancement is a specific thing, then we’re talking about specific advancement; if the advancement is something generic that you then spend or otherwise convert into a specific thing, then it’s generic.  For example, if discovering the Lost Temple of Ra gets you an advancement roll, that’s generic advancement; if it unlocks the feat Sun-Kissed for you, then it’s specific advancement.  Generic advancement still offers benefits over non-diegetic character advancement with tying characters to the world, but specific advancement is wrought chains of adamant turning a character sheet into a chronicle of heroism.  The drawback, of course, is that it’s way more work to build a world that has these rewards directly built into it, and it also requires a system that can support it; if your system doesn’t have enough unique game elements for it to make advancement feel different and rewarding, then there’s not a lot of point in using specific advancement, you’ll run out of things to reward.  This kind of specific advancement is analogous to the experience of magic items in old-school D&D, particularly the case in a play culture that prioritized modules.  If you had the Sword of Rainbow Lashes, it meant that you got it by going through the Pit of Despair, because that’s where that sword was.  Modern MMORPGs also brought this style forward, with drop tables in dungeons.  If you want that item, you go to that dungeon, it’s not going to be anywhere else.  As mentioned above I feel that the real power of diegetic advancement comes in when you’re doing specific advancement, because it ties your progression directly to your actions.  Every element on your character sheet becomes a specific thing that happened; it’s not just “I gained a level and selected this feat”, it’s “I stabbed a dragon with a magic sword, was splashed by its blood, and my world was changed thereby”.  A spell you learned isn’t something you selected on level up; you got that spell from the spellbook of the evil wizard you hired a thief to steal, and you avoid casting it with witnesses around so they don’t notice.  Every element on your character sheet becomes not just something you picked while playing a game, but a mark of valor, and carries its own story with it.

I have no ending to this planned and am done with words.  Ok bye.

This was originally posted elsewhere but is being reposted here to pretend that content exists.  You're welcome.

This is an essay about ten rules of RPG design.  They're not the only ten rules, and like any rules they can be broken as needed, but they're a good foundation to start with.

The essay is included in full in this description.  If you prefer it as a PDF, you can download it here as well.

Ten Rules of RPG Design

The Preface

First, this is about roleplaying game design, in which the players assume the role of a fictional character to guide them through an imagined world.  It is notably not about similar types of games, such as narrative engines (sometimes called “storygames”), because frankly those aren’t my thing and I don’t know how to design them.  So we are discussing only roleplaying games here. Although some of the rules here may be useful for other types of games as well, they’re mentioned in the context of roleplaying games.

I believe that these are fundamental parts of designing a roleplaying game.  These are not about my style of design or about how to design a specific game, but should apply for anyone who is trying to design any kind of roleplaying game.  As a result they are very general and if you’re looking for help designing your existing game, they may not be super helpful.  On the other hand, they are things that you should be thinking of before you begin your design.  They won’t tell you how to build a living room, but you can’t have the living room without the foundation.

There are other rules not listed here.  I’m not trying to list every possible thing that you could want to know when designing a game.  But I’m trying to list a good set of fundamentals; if you start with these rules in mind, you might need to learn, discover, or invent other rules, but you’ll at least have a good starting point on how to think about your game.

Every rule listed here has exceptions.  Even if I present something as a hard and fast absolute rule, I’m speaking in generalities here. 

The Rules

Every rule requires a reason.  When you create a rule, you need to think about what it does and why it does that.  Intentionality is the single most important part of any rule; why does it exist?  What effect will this rule have on a player who encounters it?  What effect will this rule have on the imaginary world?  What effect will it have on the table experience?  If these effects aren’t what you want, then maybe the rule isn’t what you want.

Every reason requires a goal.  What is your game about?  What experience should the players have?  When it’s over, what stories should they tell?  Before setting down a single word, you need to start by knowing that your end goal of creation is.  It doesn’t need to be in perfect detail, and you might change your mind at some point, but you need to have a goal that everything is working towards if you want to have a fully coherent and well-designed game.  Your goal should reflect the intended experience of playing your game, including both what kind of stories can be told with it and what genre those stories will fit best in.  For example, a fantasy adventure game, a cyberpunk heist game, or a sci-fi exploration game are viable (if bare-bones) goals.  You’ll need to define what your goal means to yourself, but as long as you have a clear image in mind, it doesn’t matter at this stage if you can explain it to other people (unless those other people are also designing with you, in which case, you should probably explain it to them). Don’t moderate your game design to try to appeal to people who dislike your goal.  Everything you design should be in service of your goal.  If people don’t like your goal, then your game won’t be for them, and that’s ok.

The player’s incentive should match the character’s incentive.  If you have a rule that rewards a player for doing something, it should be a thing that the character also wants to do.  Conversely, if there is a thing that you believe characters should want to do, then you should provide an incentive for players to do that as well. Not everything requires an incentive, and incentives don’t need to be used to try to ‘fix’ players or table issues.  It is not the responsibility of a game designer to deal with bad behavior or other issues.  But if you do create an incentive, then you want to make sure that it makes sense for both the player and the character.  If the incentives don’t match, then you get incoherent results.

System is setting.  The rules of the game dictate the stories that can be told with that game.  Yes, this is fundamental.  This goes back to every reason requires a goal; before you create rule 1, you should know what your game is for and what you want to do with it.  If your system states that all characters heal to full after every fight, then it’s impossible to tell a story about scrappy underdogs fighting the odds and being worn down but never giving up.  This is also true about genre, but is true about the implied settings as well.  Any setting that the system is used in will have certain things about it be necessarily true, because they’re defined in the system and its impact on the imaginary world.   Some systems are more flexible than others, and any system will be able to support multiple settings, as long as the settings all understand that they possess the intrinsic elements of the system.  But in the end, it must be known that the things you define in your system will be true in the setting as well, and that should be taken into account when defining things.

All rules and other game elements should be integrated into a seamless whole.  This doesn’t mean that you can’t have subsystems, but the subsystems should pass back a result that integrates into the core systems of the rules.  Everything should feel like it fits together and makes sense when interacted with.

Balance is a tool, not a goal.  Its purpose is to create interesting choices.  It does not actually matter if every choice is equally effective at all times, and in fact this is actively bad (that’s homogeneity, not balance).  As long as the choices that are provided are interesting, then the balance is sufficient.

The goal of progression is not the same for every system.  Campaign length is one of the things you should consider when designing mechanics.  Progression matters, and the way that your game changes as progression occurs matters as well.  Do you want characters to journey from zero to hero?  Do you want them to start at a particular power level, and just have adventures around there?  Do you want them to start by saving the village, and move up all the way through saving the multiverse?  The progression system is part of what determines the bounds of these stories.  Additionally, the progression system affects the meaningful length of campaigns and should be considered for its impact on abilities.  Many games are considered to “fall apart” after a certain amount of progression because the cumulative impact of progression on character abilities results in an unintended experience.

Design elements can carry both Breadth and Depth.  Roughly speaking, Breadth is how many different choices you can make, and Depth is how much each individual choice affects your gameplay experience.  A game that is Broad can be easily replayed in a lot of different campaigns, while a game that is Deep supports a long-term campaign with a single character.  When designing your game, you should think about this, and what kind of play structure you want your game to have.  Is it a beer and pretzels game that people will play one-shots of repeatedly?  Make sure it’s Broad.  Are you designing to support 20-year campaigns?  Make sure it’s Deep.  (It is possible to be both, if you add all the rules to support both.)

Let players engage with the things that they want to engage with.  If a player repeatedly selects game elements that improve their abilities at a specific thing, that means that the player wants to engage with that thing.  Many games will have these mechanics remove the need to engage with that thing; this is a mistake.  If a player wants to focus on something, and the mechanics are there to allow them to focus on it, then the mechanics should not prevent them from engaging with it.

Games that contain a small but nonzero percentage of ridiculous bullshit are more fun.  A game that’s purely predictable feels sanitized.  Don’t try so hard to make the game balanced or predictable or fair that you squeeze all the life out of it.  Let ridiculous things happen.  Let the players be frustrated occasionally and let things that aren’t fun in a vacuum happen.  Without valleys, the peaks aren’t meaningful.  Of course, a balance needs to occur, and presumably your goal is for the game to be fun overall; but if you try to make it so that the game is never even a little bit un-fun, you’ll end up making it so it’s never exciting and fun, either.