Guardrails: avoiding unexpected breakage vs preventing exploits

[PhoenixPhyre]

Bouncing off a quote in a different thread (that I didn't want to sidetrack) on MM3:

But speaking more broadly, M&M has plenty of exploits like that. It's inevitable in a game with such deep customization; the fact that it's a superhero game with deep customization makes it all worse. Including a wide variety of enemies and attack types helps, but ultimately it's up to the players not to be d*cks. Which is...kind of true for all ttRPGs. Heck, all social activities.

The bolded section meshes really well with a realization I came to a while ago--good systems have guardrails and try to avoid broken stuff...but how they do so matters. It's better (IMO) to write in guardrails that reduce the chances of accidental breakage over trying to prevent motivated players from exploiting things. I mean, it's probably useful to think about how something can be abused intentionally and rewording the obvious ones, but that hits diminishing returns.

TTRPGs require a fundamental level of trust between the players (including a GM, if there is one). Those meta-agreements, the resolution to "not be d*cks" (whatever that means at that particular table), that's the main constraint on exploits. No amount of text on the page can materially affect someone determined to exploit the fine cracks; no perfect abstraction exists, and motivated reasoning means you can find loopholes wherever you want to. All that really matters is the willingness of other players to go along with your shenanigans. And shenanigans, when agreed to by the table, don't really cause too many problems.

On the other hand, I find it very important for systems to sand down some of the rough edges and work to make the experience clean for new players and groups. That doesn't mean simple systems only--complexity is ok (like all things, in moderation). But you shouldn't have to reach through shattered windows with broken glass still in the frame to manipulate the game state or build characters. It should be quite difficult (nothing can be impossible) to accidentally create a broken character in either sense of the word. Either one that falls well short of system expectations OR one that vastly exceeds the system's comfortable upper limits. Both of those might be possible, but should require active intent and system mastery.

Ideally, the system's "happy path" (most well-supported character types and power levels) should be the default, easiest path. Taking the "obvious, thematically evocative" options at each step for something the system supports should produce a competent (by system standards) character. Maybe not the most optimized (the floor is not the ceiling), but capable of meeting the system's minimum standards comfortably. This goes extra for DM-side stuff--if the DM has to make sure to juggle and design loot drops so that half of the party can keep up (the others not needing such things) or if the DM needs to carefully design challenges and tell new players about very non-obvious choices[1], that's a problem. Because those things won't happen with a new GM.

One way to do this is to be open and explicit about the "happy path". And then bake the necessary pure numbers straight into the build system. Whether that's classes or not is less relevant (although classes do make this kind of design easier since they channel the choices down predictable lines). It's the difference between base 4e D&D requiring a bunch of magic items (appropriate armor and appropriate magic weapons/implements) and at least one feat (Weapon or Implement Focus) just to meet baseline due to monster values scaling faster than base PC values and 4e or PF with Automatic Bonus Progression in place. One is failable, the other isn't.

TL;DR I prefer when systems focus their guard-rails on avoiding accidental breakage/cleaning up the sharp edges over trying to prevent exploits by locking everything down or focusing on "hard numeric balance".

[1] Is the 3e D&D Tough feat good for people who want to make tanky martial characters? Nope. Not at all. For that purpose, it's decidedly sub-optimal. But it's actually useful for wizards...despite nothing about it making that clear.

[Mechalich]

The thing about preventing accidental breakage is that is requires the developers of the system to both possess extremely high system mastery and to playtest the system extensively to understand where the mines are buried that can cause builds to explode through the intended system parameters. The more complex the system, the more difficult this becomes, since the math-hammering portion of the testing becomes more complicated and the number of elements necessary to juggle when working through the options increases exponentially due to the myriad interactions of different sub-systems. Somewhat ironically, a system's success works against it in this regard, since a more successful system mandates more content, which increases more options, which produces greater complexity, which produces new and unanticipated points of accidental failure.

Traditionally, this has been difficult to achieve. TTRPG developers are generally not mathematically-minded persons and have been extremely hesitant to conduct the kind of statistical probability analysis necessary to properly test the mathematical models underpinning various RPG systems. Playtesting is worse, since it's extremely expensive to conduct and even when done may not reveal problems if the developer team plays the game differently than the majority of the player-base ultimately does (this was, traditionally, a huge problem with basically all White-Wolf publications).

Taking the "obvious, thematically evocative" options at each step for something the system supports should produce a competent (by system standards) character.

One difficulty is that the obvious and thematically evocative options in specific source material may include non-competent characters. With regard to superhero genre, this is a major problem, since that genre hides the fundamentally squishy nature of certain builds behind a series of dodges that while acceptable in single-author fiction don't work in a collaborative tabletop game. A proper superhero game probably needs to forbid Batman, since without his plot armor he gets shot and killed sometime before the third mission is finished, but it's really hard to enforce a 'no Batman' design code in a superhero scenario.

[PhoenixPhyre]

It's actually easier than that if you don't try to prevent accidents entirely, just make them harder and make the system more resilient to their presence. This generally means having a broader, flatter power curve for characters, as well as reducing the supported range of character traits.

A tightly wound, complex system is more fragile and rigid than one with seams and expansion joints. One that tries to codify everything and is designed to be played "RAW first" is more fragile than one that teaches players and DMs where the happy path lies and what the system expects, while giving DMs lots of leeway to adapt.

Class based systems can be (but aren't necessary) easier to guard in this way than classless ones, simply because there are fewer interacting parts and more is decided in advance. And they can declare exactly what is supported and what is not. You can rule out the accidentally incompetent characters to a substantial degree if every class is competent at a baseline level by default. That has other effects, so it's not a panacea. And many designers fail to do make use of this. Some (ie 3e D&D) glory in creating trap options.

Generally, if you try to make a game generic (within a genre), you have to rely more on human judgement because you can't rule out broken combinations globally without unacceptably reducing the range of characters. But then again, I prefer games that don't try to be build a bear, everything to everyone types. Choose your supported character types and support them well. And be explicit about what is supported and not supported. Anyone playing outside that takes that risk on their own, but does so knowingly (and thus not accidentally).

[Reversefigure4]

I'm all for side-bars that openly lay out 'this ability is potentially abusable, and here's how you'd do it'. It flags it openly to the GM so they see it coming, it flags it to the players so they don't feel the need to prove their cleverness by building it.It makes it nice and clear when someone is violating the 'don't be a ****' rules, particularly if it flags that Ability A is fine, and Ability B is fine, but they shouldn't be combined.

Equally, I'm all for example characters or sidebars that say "a competent character should be able to do X". No system in the world will hold up against a player trying to deliberately make an incompetent character (one level in each class usually produces something functionally useless), but if we can see what a character should look like and have good guidance on building it, we should be able to tell why that doesn't work.

[Satinavian]

Balancing is a good thing. Some abilities are just too good and warp the game around them whenever they are in play. Others are so weak that no one ever takes them, making them a waste of space (or worse, someone actually takes them because they are not good at rules).

But i think a system is fine, if the parts get balanced, trying to fight synergies and redundancies is a never ending struggle and ends in convoluted mess with so many restrictions that you hardly can build a character you want. And it takes so much work better applied elsewhere. I think, synergies should only be considered for balancing for the thematically obvious combinations at best because those tend to pop up when players follow an inspiration, not when thy try to get power through rule mastery.

I am generally also very much in favor of diminishing return and/or caps. Both tend to limit the effectiveness of min-maxing naturally without going through all the combinations. It doesn't matter how you get a stat so high, the payoff is just not that big anyway.

I dislike classes. Sure, they are easier to balance, but only because they restrict viable characters to a very limited set of archetypes. That is to much of a price to pay imho, especially when it is still far from easy to balance class systems.

I'm all for side-bars that openly lay out 'this ability is potentially abusable, and here's how you'd do it'. It flags it openly to the GM so they see it coming, it flags it to the players so they don't feel the need to prove their cleverness by building it.It makes it nice and clear when someone is violating the 'don't be a ****' rules, particularly if it flags that Ability A is fine, and Ability B is fine, but they shouldn't be combined.

If the designers actually noticed a combination that is too bad, they could as well just ban it and they usually do.

I think those sidebars are more useful for abusable subsystems where the problematic combinations can be reached in various ways. Like "Our animal companion build rules allow a huge variety of results. However if you were to stack only templates of similar kind, your companion would exceed values of a normal character of your experience in a narrow niche. Please don't do that."

[KorvinStarmast]

This generally means having a broader, flatter power curve for characters, as well as reducing the supported range of character traits.

BitD does this well enough. They also did a good job on the strong archetypes / class thing.

But then again, I prefer games that don't try to be build a bear, everything to everyone types. Choose your supported character types and support them well. And be explicit about what is supported and not supported. Anyone playing outside that takes that risk on their own, but does so knowingly (and thus not accidentally).

Your last bit I fully concur on.

[PhoenixPhyre]

I dislike classes. Sure, they are easier to balance, but only because they restrict viable characters to a very limited set of archetypes. That is to much of a price to pay imho, especially when it is still far from easy to balance class systems.

I agree with the rest of the post (in broad strokes), but do want to say that for me, personally (and this is entirely opinion), I prefer when games limit themselves to a narrow set of archetypes, so having classes is a strong plus for me. But then I don't play in genres where that doesn't make much sense (like superheroes or a lot of modern stuff). I'm pretty firmly in the "fantasy adventuring" camp as far as genre preferences, and there having strong archetypes makes everything work much more smoothly and allows stories to focus on the adventuring part. And interactions with the world have nice, well-defined interfaces.

Basically, "mechanical character customization" isn't the highest good for me. Not at all. In fact, it's somewhere down middling or lower on the importance scale. TBQH, I've never really noticed that it makes much of a difference in the characters that result, which seem to be much more differentiated by different people playing them than anything mechanical. I've had a number of D&D fighters whose mechanical differences were small who all had very different impacts on the setting. I've had characters whose only unifying factor was "casts spells" (very different spell selection and other mechanics) who had basically no impact on the setting or play--they were just there. So as far as differentiating factors go, player >>>> characterization >>>>>>>>>>> mechanics IMO. As a result, I'm happy when games limit the archetypes sharply. And then individual campaigns limit them even further. I'd be happy to play in a "you must all have at least one level in fighter" game. Or a "no full casters" game. Etc.

[Satinavian]

Well, it is preference.

I don't like classes in fantasy games either. And if i think back to the last dozen or so characters i played in fantasy games, pretty much none of them would fit well with the classes most fantasy systems offer (In the core books that is. If you have 300+ different classes over dozens of supplements that is another story. Still not to my liking as it would be probably easier to just go classless)

As for limited campaigns, i prefer to just go with a theme instead of any mechanical class restrictions. Sure, you can go "One level of fighter for everyone". But you could also do "you are a military unit in conflict X" or "you are a group of young mage aristocrats of empire Y and their close associetes trying to impress their elders" and use classless systems.

[KorvinStarmast]

I don't like classes in fantasy games either.

You are an experienced player.
For the new player ~ I have done quite a bit of DMing for teens and pre teens ~ a class unburdens the player a great deal in terms of removing a barrier to entry. A lot of work is pre done for them.

As the player gains more experience in the TTRPG form of play, I can see where some players find classes too limiting. Very much a matter of taste.

As but one recent example:
We are playing a space/horror game called Mothership that has pretty well built classes / archetypes.
It was a hell of a lot easier for me, a new player to this game, to have a few choices (less than 12) of "kit" to use to try the game out and play it than it would have been for me to create, from scratch, the Teamster I am currently playing. I don't have to know the game inside and out at the Chargen stage. That's beneficial.

[Atranen]

Well, it is preference.

I don't like classes in fantasy games either. And if i think back to the last dozen or so characters i played in fantasy games, pretty much none of them would fit well with the classes most fantasy systems offer (In the core books that is. If you have 300+ different classes over dozens of supplements that is another story. Still not to my liking as it would be probably easier to just go classless)

As for limited campaigns, i prefer to just go with a theme instead of any mechanical class restrictions. Sure, you can go "One level of fighter for everyone". But you could also do "you are a military unit in conflict X" or "you are a group of young mage aristocrats of empire Y and their close associetes trying to impress their elders" and use classless systems.

Which classless systems do you like the most for fantasy?

[Satinavian]

Which classless systems do you like the most for fantasy?

That is easy : Splittermond

Unfortunately not available in English.

I also played a lot of TDE in the past but depending on the edition it is not completely classless, more like a hybrid. Also it is not nearly as good. I also remember fondly some campaigns in SIFRP but that is otherwise restricted in scope.

[Atranen]

That is easy : Splittermond

Unfortunately not available in English.

I also played a lot of TDE in the past but depending on the edition it is not completely classless, more like a hybrid. Also it is not nearly as good. I also remember fondly some campaigns in SIFRP but that is otherwise restricted in scope.

Ah too bad! I would love to try it out.

[gatorized]

The idea that it must be impossible to write a system without including blatant game-breaking exploits is a result of only having experience with systems made by incompetent people, mainly D&D and related D20 systems.

You are an experienced player.
For the new player ~ I have done quite a bit of DMing for teens and pre teens ~ a class unburdens the player a great deal in terms of removing a barrier to entry. A lot of work is pre done for them.

As the player gains more experience in the TTRPG form of play, I can see where some players find classes too limiting. Very much a matter of taste.

As but one recent example:
We are playing a space/horror game called Mothership that has pretty well built classes / archetypes.
It was a hell of a lot easier for me, a new player to this game, to have a few choices (less than 12) of "kit" to use to try the game out and play it than it would have been for me to create, from scratch, the Teamster I am currently playing. I don't have to know the game inside and out at the Chargen stage. That's beneficial.

What classes actually do is give players a bunch of abilities they're not aware of / don't understand / have no interest in using / don't fit their character concept. When you pick each and every power your character has - and you don't need fifty powers to be effective - you're far more able to make meaningful decisions about what your character will do.

[BRC]

What classes actually do is give players a bunch of abilities they're not aware of / don't understand / have no interest in using / don't fit their character concept. When you pick each and every power your character has - and you don't need fifty powers to be effective - you're far more able to make meaningful decisions about what your character will do.

okay, this is system dependent, but

What classes do for new players is give them a solid starting point for building their character.


If a new player comes to a class based system and says "I want to build a Sword Guy!", you hand them the Swordguy class, and it's got 4-10 abilities around Swords.

Your new player now must merely learn 4-10 abilities in order to engage with the game, and they have their sword guy.

This is inherently limiting, but also more accessible. The GM only need to understand the system well enough to identify the Swordguy class and explain the abilities. The player only needs to learn the handful of abilities they actually have, and if the game is decently designed, picking a class guarantees a certain level of capability and cohesion with your powerset. The player shouldn't have to worry that they've "Failed" character creation in a class-based system.


By comparison, a build-a-bear system might have 50 abilities for Sword users. Even if the player only gets to pick six of them, their first exposure to the game is "Comprehend these 50 abilities and pick six of them." That's a lot of decisions the player may not have proper context for. It's more versatile sure, but also less accessible, an inherent part of character creation becomes the "Game" of Character Optimization, simply to end up with a reasonable character.

And even build-a-bear systems are often less versatile than they advertise themselves. "I want to play a magic archer!" "Sure, this is a non-class based system, build whatever you want!" only for the Magic abilities and the Archery abilities to require totally different stat setups, and have the system assuming everybody made a focused, synergistic build. Your "magic archer" ends up a mediocre archer with mediocre magic.

A Class-based system is inheriently limiting, but it contains the promise (True or false) that there is a list of things you can pick from that will result in a decent character.

A game with true power customization, like HERO, can be either better or worse depending on your GM. Rather than picking from a list, the player is given the nebulous concept of "Whatever" and has to work through a complex system of power customization. If you've got a GM who really knows the system, they can work with the player to build an appropriate power list. If your GM is also new to the system, then building something as simple as "Good at Swords" can be a frustrating headache, and everything above about pitfalls is even more prevalent.

[PhoenixPhyre]

okay, this is system dependent, but

What classes do for new players is give them a solid starting point for building their character.


If a new player comes to a class based system and says "I want to build a Sword Guy!", you hand them the Swordguy class, and it's got 4-10 abilities around Swords.

Your new player now must merely learn 4-10 abilities in order to engage with the game, and they have their sword guy.

This is inherently limiting, but also more accessible. The GM only need to understand the system well enough to identify the Swordguy class and explain the abilities. The player only needs to learn the handful of abilities they actually have, and if the game is decently designed, picking a class guarantees a certain level of capability and cohesion with your powerset. The player shouldn't have to worry that they've "Failed" character creation in a class-based system.


By comparison, a build-a-bear system might have 50 abilities for Sword users. Even if the player only gets to pick six of them, their first exposure to the game is "Comprehend these 50 abilities and pick six of them."

A game with true power customization, like HERO, can be either better or worse depending on your GM. Rather than picking from a list, the player is given the nebulous concept of "Whatever" and has to work through a complex system of power customization. If you've got a GM who really knows the system, they can work with the player to build an appropriate power list. If your GM is also new to the system, then building something as simple as "Good at Swords" can be a frustrating headache.

And even more so, build-a-bear requires you to come into character creation with a fully-formed idea of what your character WILL BE. Especially if it (as most do) has different costs at character creation compared to advancement. Then you need to basically have your entire power set set in stone, because buying anything new and relevant later becomes prohibitively difficult. The level of commitment you need up front is huge, and you have to do all the work yourself. And if you don't have a strong character concept, you'll either flounder or you'll end up making a fairly useless character. And either way, you're stuck.

Basically, a build-a-bear system starts with "I want a character with X power set" and then translates that into a mechanical construct. Benefit--you can play exactly what you want (most of the time). Downside--you have to know exactly what you want ahead of time. And characters don't end up nearly as organic.

Whereas with a class system, you can start out at the broadest archetypes and grow into the character. I've had lots of players who picked their class entirely based on a cool mini. "This one's got a big axe, gimme that one." Short discussion of fighter vs barbarian and they went barbarian. System mastery required? basically 0. And things work just fine. No need to scour through bunches of books and do all the math--it's defined for you. And you're guaranteed to get something that fits the adventures and gameplay and setting aesthetic if the DM has done even a half-way job of curating a much smaller list of options, compared to filtering through hundreds, if not thousands of granular options, most of whom are vague or ill-defined thematically.

--------

But that's all secondary IMO. The key point of the OP is that bold sentence. No one should be able to "fail" at character creation, at least not accidentally. You can choose to create an incompetent character--no system can stop you. But that should require active working against the system--the system should lead you to something acceptable (from the system's perspective). That's possible in both a class-based and classless system. The methods are different, and class-based can be easier to safeguard. But it's still something designers have to explicitly and intentionally do--3e D&D, for instance, utterly 100% failed at even trying to do this. You can lose D&D at character creation without knowing you've done so for several levels. And the game in fact glories in presenting outright traps as viable options. That is bad design IMO.

[BRC]

In defense of Build-a-Bear systems, sometimes pitfalls can be avoided with simple guidelines.


For example, in Deadlands classic ( A game I adore for a variety of reasons), you have a build-a-bear character system befitting the overdressed 90s RPG that it is. One part of that is that the "Character Points" you use to buy skills during character creation are distinct from the "Bounty Points" you use to Improve a character. Notably, you can buy any skill up to 5 by spending an equal number of character points, but using bounty means improving a skill one step at a time.

So buying "Shootin' 5" at character creation is just "Spend 5 character points", but going from 0 to 5 shootin' afterwards means spending a total of 15 bounty points over at least 5 sessions.


This means that if you want your character to be really good at shootin' and riding horses, but only okay at, say, playing Harmonica, the thing to do is spend your character points as efficiently as possible, buying up your core skills with those, and then use bounty points later on to fill in the things you only want one or two ranks in.


So a simple guide like "Avoid spending character points on skills with less than 3 ranks, also make sure you have a good Guts skill" goes a long way towards avoiding those pitfalls.

[PhoenixPhyre]

In defense of Build-a-Bear systems, sometimes pitfalls can be avoided with simple guidelines.

And to be very clear, I'm totally fine with that kind of explicit guideline, put somewhere obvious. Hidden away in a sidebar...not so useful. In fact, that's the kind of guidance I want in most places. Tell me what the system expects, and if there's something non-obvious as a result of how the math ends up working that could result in accidentally-broken characters, flag it with obvious markers.

When I say "guardrails", I don't generally mean mechanically-self-enforcing hard barriers. Because those generally result in brittle systems that can't be modded effectively, which I don't like. And still fail to resist motivated individuals. Because nothing can keep people from breaking a system if they try hard enough. Like the common locks on most front doors, they're designed to keep accidents from happening and keep honest people honest. Not provide a hard guarantee of safety.

[Telok]

I've always found example characters and quick-build templates useful in classed, build-a-bear, and hybrid systems.

Often class systems assume that because they're a class system you can't screw up the build, but then they fail to make their actual inspirations workable and don't bother telling you this. Wanna play a D'Artagnan like character? Surprise! You need to use the thief class instead of the warrior class, learn lock picking, and have to stab people in the back to be relevant in combat.

On the other hand a couple base templates and a "pick 2 from each lists A, B, & C" quick build has made creating effective characters in Champions take under 10 minutes.

Useful examples and clear decent guidelines about how the game is intended to be run are really helpful. Stuff like "you can run the game rolling for everything, nearly nothing, or anywhere in between" with no further discussion of what those look like is a useless waste of space.

[Easy e]

If looks like much of this discussion is about class and Level vs Build-a-bear so of course I am going to go in a wildly different direction......

There is no system in the world a motivated person can not break..... someone, somewhere can and will find a way to break it.

Therefore, I don't bother to stop those folks. You can not stop them.

Here is what you may want to do instead:

1. Write to lean into your design goals and explicitly state your goals to lead to #2

2. Write it for your target audience who align with the design goals in the first place or they would get a different game

3. I lean heavily into the idea that Role-Playing Games are collaborative exercises engaged with for entertainment in which the players and GM are complicit in each others fun.

Therefore, breaking things doesn't matter as much because people have less reason to break anything. It would go against the explicit design goals, the target audience, and the core assumptions of collaboration in the games.

[PhoenixPhyre]

If looks like much of this discussion is about class and Level vs Build-a-bear so of course I am going to go in a wildly different direction......

There is no system in the world a motivated person can not break..... someone, somewhere can and will find a way to break it.

Therefore, I don't bother to stop those folks. You can not stop them.

Here is what you may want to do instead:

1. Write to lean into your design goals and explicitly state your goals to lead to #2

2. Write it for your target audience who align with the design goals in the first place or they would get a different game

3. I lean heavily into the idea that Role-Playing Games are collaborative exercises engaged with for entertainment in which the players and GM are complicit in each others fun.

Therefore, breaking things doesn't matter as much because people have less reason to break anything. It would go against the explicit design goals, the target audience, and the core assumptions of collaboration in the games.

I agree and I disagree. I'm not worried about motivated people breaking stuff. As you say, that's inevitable.

I am worried about, and want to make more difficult[1], accidental breakage. In either direction. Someone who wants to break something should have to do so knowingly and intentionally by cobbling together options with malice aforethought, not just by picking the straightforward path.

As you say, having clear design goals and explicitly and clearly stating them makes a big difference. As does flagging some of the (inevitable) pitfalls along the way. But so does making sure that
1) archetypes (where they exist) are explicit
2) any sample characters/template bundles (in a classless system) produce valid (by system standpoints) characters who have clear upgrade/progression paths
3) classes (in a class-based system) similarly produce valid characters. Where you have a closed list of options within a class (such as a 5e Battlemaster's Maneuvers), you can actually make sure that none of them, in isolation or in combination, produces an out-of-bounds character. Open-list, shared elements (like feats or spells in D&D) should either have guidance (e.g. pick at least one spell that does damage, don't heavily specialize in only one element/type unless you have a feature that lets it apply more broadly, etc) or should have "obvious" (or worst case suggested) choices. And those should work.
4) none of the game elements, applied in its straightforward use, should be out of bounds. No one-button nukes. No useless spells or feats in isolation. Everything, used in isolation, should be an acceptable outcome. Yes, even if spammed. Because resource constraints aren't really binding constraints.

Note that #4 doesn't restrict combinations of game elements much at all. It's about stopping the obvious ones. If you have any ability that, standing alone, produces unacceptable results, change it so it doesn't or remove it.

[1] nothing is impossible--it's important to never underestimate the boundless power of people to screw things up accidentally. But you can make it less probable.

[Chronos]

Quoth Mechalich:

One difficulty is that the obvious and thematically evocative options in specific source material may include non-competent characters. With regard to superhero genre, this is a major problem, since that genre hides the fundamentally squishy nature of certain builds behind a series of dodges that while acceptable in single-author fiction don't work in a collaborative tabletop game. A proper superhero game probably needs to forbid Batman, since without his plot armor he gets shot and killed sometime before the third mission is finished, but it's really hard to enforce a 'no Batman' design code in a superhero scenario.

Not necessarily. One superhero game I play (a computer game, but the principle should still hold) has separate build space for origins and powers. Everyone, of any origin, selects their powers in the same way, and the same powers have the same mechanical effects for everyone, regardless of origin. And one of the origins is Natural, which is described as you not actually having superpowers, just being very highly trained and skilled. Some powers are an easier thematic pick for some origins, and you can end up with silly combinations like someone who constantly spews radioactive fire from their body just by practicing really hard at it, but mechanically, it's no problem, and you can cut down the silliness (or encourage it, if that's your thing) just by talking with your players and not being jerks.

[KorvinStarmast]

What classes actually do is give players a bunch of abilities they're not aware of / don't understand / have no interest in using / don't fit their character concept. When you pick each and every power your character has - and you don't need fifty powers to be effective - you're far more able to make meaningful decisions about what your character will do.

Your bias is noted.
As a generalization, the bolded bit is utterly false.
In some cases, yes, there's some "what is this doing here?" in class-based games.
In the recent example I offered, Mothership, that is not the case and I also had four skills to select from (in terms of boosting a capability like first aid or repair) that were beyond the class template.
The class template saved me a lot of work/prep in terms of having to know the whole game in order to create a character.


In Blades in the Dark, there are a limited number of upgrades available based on class, but you pick one. Then, through play, as you earn enough flags/points to choose another, the ones you choose are still defined by your class/archetype. (I have a Leech and a Hound in play at the moment, depending on the night and what the Crew needs).

[Pex]

Probably a side concern but I still think relevant is the knowing the difference between what is breaking the game in the overpowering perspective and what is a character being powerful as intended. At the extreme end it's easy to determine what is breaking the game in the overpowering perspective where the Thing that is breaking is making the game unplayable. However, different people have different tolerance levels of PC power so one person's game breaking bug is another person's powerful feature. The guidance given to the DM should be clear the game expects this particular power level, the players are to benefit and enjoy it, and here is advice on how to use it. If the game can handle it there's no harm in offering suggestions and advice for those people who don't want such high power and the consequences there of in how the game will function. However, the game should own the power it gives, not apologize for it, and those players who refuse to accept it are advised this game is not for them.

[PhoenixPhyre]

Probably a side concern but I still think relevant is the knowing the difference between what is breaking the game in the overpowering perspective and what is a character being powerful as intended. At the extreme end it's easy to determine what is breaking the game in the overpowering perspective where the Thing that is breaking is making the game unplayable. However, different people have different tolerance levels of PC power so one person's game breaking bug is another person's powerful feature. The guidance given to the DM should be clear the game expects this particular power level, the players are to benefit and enjoy it, and here is advice on how to use it. If the game can handle it there's no harm in offering suggestions and advice for those people who don't want such high power and the consequences there of in how the game will function. However, the game should own the power it gives, not apologize for it, and those players who refuse to accept it are advised this game is not for them.

Yeah. I'm all for clearly marked boundaries on what the system supports. Going outside that should be knowing and intentional, and buyer beware. It's not the system's fault if you use it outside the clearly marked safe zone and it shatters. I do feel more comfortable blaming system designers who don't mark that area clearly or at all or when it shatters when you combine obviously synergistic elements and stay within the marked zone

[Pauly]

One of the best guardrails a system can have is to deliver on what it promises.
For example D&D is often sold as having highly tactical combat, which in reality is complete bollocks. Character creation has far more to do with the ability to win fights in D&D which is why you have so many pixels slaughtered on threads about how to build [X] class and so few on what are the best actual tactics in a given encounter.

Games like old school Traveller with randomized chargen get people more engaged with who their character is as a person instead of treating their character as a shopping list of bonuses.

[Zombimode]

OCharacter creation has far more to do with the ability to win fights in D&D which is why you have so many pixels slaughtered on threads about how to build [X] class and so few on what are the best actual tactics in a given encounter.

That a fallacious argument. The reason why you don't see online discussions about tactics for D&D is that there is no common ground. Every encounter for every party is it own unique tactical scenario. And those are all private.

I don't know about your games, but in my games the players do analyze and discuss their tactics, especially after great victories or great failures. This suggest that tactics not only exists in D&D but also matter quite strongly.

[Telok]

That a fallacious argument. The reason why you don't see online discussions about tactics for D&D is that there is no common ground. Every encounter for every party is it own unique tactical scenario. And those are all private.

Hogwash, people play the modules with the same encounters all the time, with parties that commonly map to things like "2 warriors + mage + priest + 5th wheel". There's no real D&D tactics discussion because nearly all martial builds only really get one effective option and everyone else has the top 3 spells for their class & level and they're used the same way nearly every time.

There just isn't the depth of options to support tactics in current D&D once you're past which spells to cast in what order. Melee walks to the closest enemy to smash, archers shoot whatever is dangerous, casters drop the same patterns of "conc buff/cc + nuke/cc & repeat" basically every single encounter.

Game I'm running there's real trade-offs between light armor and sealed space suit armor for everyone. Suppressing fire works as an alternative to just shooting for hit point depletion. The shooty guys are also very good brawlers with excelent perception and first aid skills, plus they're the pilots & drivers. The sword-mage is also the party hacker in a Shadowrun-like environment. They can all effectively stealth in one way or another. They have options for flying, burrowing, and submersible vehicles, armed and civilian (which matters in many many places). And then there's always explosives. They've got choices on how to approach their problems. If they were a D&D 5e group the approach would always be the same "walk up for SMASH while the cleric casts bless and the sorcerer casts haste" like the previous year of D&D we played.

[Pex]

One of the best guardrails a system can have is to deliver on what it promises.
For example D&D is often sold as having highly tactical combat, which in reality is complete bollocks. Character creation has far more to do with the ability to win fights in D&D which is why you have so many pixels slaughtered on threads about how to build [X] class and so few on what are the best actual tactics in a given encounter.

Games like old school Traveller with randomized chargen get people more engaged with who their character is as a person instead of treating their character as a shopping list of bonuses.

I'd rather play what I want to play than what the dice tell me I must play. That and not die during character creation.

[Pauly]

That a fallacious argument. The reason why you don't see online discussions about tactics for D&D is that there is no common ground. Every encounter for every party is it own unique tactical scenario. And those are all private.

I don't know about your games, but in my games the players do analyze and discuss their tactics, especially after great victories or great failures. This suggest that tactics not only exists in D&D but also matter quite strongly.

Basically what Telok said. I play a lot of wargames and have achieved a few 3rd place finishes in nationals in different systems. I’ve also flamed out in more than a few tournaments too. Which is to say I’m not consistent enough to be a top tier wargamer but I know my way around a TTMWG. D&D has close to zero tactical depth or breadth. A good example of a RPG which has decent tactical combat is Cyberpunk.

I'd rather play what I want to play than what the dice tell me I must play. That and not die during character creation.

Playing what the dice gives you is a different experience to choosing what you want. My experience has been that the randomized characters work well enough as long as the players are open to non optimized characters or preset ideas if what they want to play. I personally get more invested in my rolled up characters than my optimized chracaters.

[Pex]

Playing what the dice gives you is a different experience to choosing what you want. My experience has been that the randomized characters work well enough as long as the players are open to non optimized characters or preset ideas if what they want to play. I personally get more invested in my rolled up characters than my optimized chracaters.

That can be your preference in taste, but it doesn't dictate players choosing the character they want to play exploits a game.